Silas poses the following question in the comments (link)
Let’s say a few scientists have announced that they’ve discovered an effectively unlimited/costless energy source (like the cold fusion situation). A few of their friends have gotten it to “work” too, in that it appears they’re really on to something — perhaps energy problems are about to be solved forever.
On the other hand, most claims like these turn out to be fraudulent or mistaken. And even if they’re doing everything properly, it could be that they’re not tabulating the energy flows correctly (i.e. not realizing that something critical is consumed in the process, meaning it requires more resources and energy input to keep it going indefinitely, which they hadn’t accounted for on deeming it costless). But still, there is much evidence for and against, and the truth or falsity of the claims really matters.
In this situation, it would be very wise not to commit resources to narrow uses, as the eventual unveiling of the truth about the energy claim will obviate many kinds of economic activities. If it turns out to be true, huge sectors of the economy become worthless: the drilling equipment for oil, energy prospecting equipment, coal mines, natural gas pipelines, equipment for maintaining these pipelines, transportation networks that exist because all of this, and, of course, all of the specialized workers and knowledge sets so related. Investing *more* in these will mean you’re “caught with your pants down” when the claims are verified to be true.
On the other hand, investment in the *opposite* direction will be painfully obviated if the claims turn out to be false. So all investments and most purchases carry an additional risk-cost, and so holding off on doing so is a rational response to the economic realities involving high uncertainty about the usefulness of particular goods. In this situation, if you got people to spend and invest for the sake of propping up spending and investing, you’re causing a huge waste of resources: since people don’t yet know which course of action is right, many more goods than otherwise are going to be committed to wasteful production structures.
First let me say that I haven’t come up with any satisfying answers to Silas’ question. I don’t know how to think about it properly, so I will be more humble.
There are a few reasons why the idea that a general reduction in spending due to increased money demand is a desirable signal:
- A strong signal to avoid investment in affected sectors is already present, people who invest before the uncertainty is resolved will make worse investment decisions than people who don’t.
- There are lots of reasons why money demand can increase and many of them do not have to do much with uncertainty like this. For example, some new financial technology might require lots of transactions and people need money in order to conduct the transactions.
- A general spending reduction sends a very blunt signal. A shortage of money is likely to cause people to reduce their spending on many different margins, most of which will not be relevant to the sector. For example in the example you gave, the uncertainty is around one sector (energy) which may have little effect on some other sectors, so it likely makes sense to shift resources that cannot be transferred into the future into those sectors (say entertainment and housing)
- The signal is fleeting, as prices adjust, the signal will go away even though the need for transfering resources into the future may continue. In a world with no sticky prices this signal would not exist; nominal spending would fall, but real spending would be stable.
I think one of the key issues is the time frame over which this plays out. I am not sure what kind of time frame Silas had in mind: ~2 months or ~2 years.
Over a time period of ~2 months, I am especially unsure how to think about the issue, but keep in mind that most proposals for nominal income targets focus on income expectations ~2 years out, and would only try to stabilize short term expectations to the extent that they affected longer term expectations (and only by trying to stabilize longer term expectations). I am unsure about the desirability of stabilizing income over short time scales, but in any case, I don’t think it is possible.
Over a time period of ~2 years, I think I understand the issue a little bit better. Over the course of 2 years the appropriate response for the economy is to shift production away from investment in affected sectors and toward resources that can be stored until the uncertainty is resolved investment in unaffected sectors, consumption and leisure. It is important to note that there are many kinds of resources that you cannot transfer into the future easily, for example lots of kinds of labor. The proper response in these cases, is generally to do something else with them rather than do nothing with them. Prices are probably mostly flexible over this time period, so allowing nominal income to rise or fall will not affect resource allocation a lot, but will affect the dynamics of the transition.
Good Morning, Economics 
161 comments
Comments feed for this article
August 23, 2010 at 11:34 am
Silas Barta
Thanks for your reply, John. Here is the part I most disagree with:
Over the course of 2 years the appropriate response for the economy is to shift production away from investment in affected sectors and toward resources that can be stored until the uncertainty is resolved investment in unaffected sectors, consumption and leisure. It is important to note that there are many kinds of resources that you cannot transfer into the future easily, for example lots of kinds of labor. The proper response in these cases, is generally to do something else with them rather than do nothing with them.
Labor is more complex than just an issue of whether to work now or later (in which case it’s indeed wasteful to simply stop working — when you value the money more than the lost leisure, of course). Rather, when the economy invests in a certain kind of production, more retooling is going on that just workers shifting jobs. Rather, things will be built that requires certain other things in place to work best, which require other things, and so on. An entire “ecosystem” develops around new projects.
This is the real (employment-wise) problem with government stimulus projects (not that you’re advocating this, but to show the reasoning): if the economy re-tools around there being a new e.g. road system being built, then when the project ends, all of the resources (capital as well as labor) have to re-tool right back out of it. What’s more, it has to bear the risk cost I talked about before, since there’s the big question mark about how long this project will be propped up.
The same point applies to monetary stimulus, even over the 2-year horizon. So you get people to spend for the sake of spending, against their own judgment. So purchasing patterns are distored. So the economy retools around an unnatural set of preferences that will fade away as soon as you stop rewarding people for spending for the sake of spending. And the resources (including labor skills) committed to that pattern of behavior isn’t costlessly recouped, but has to go through yet another painful adjustment.
My alternative is: don’t introduce these transient episodes. Employment is maintained by sustainable patterns of specialization and trade (to borrow Arnold Kling’s phrase). Let people adapt to the new economic realities, and the economy will re-tool around the new wants (including the new lower productivity of some existing modes). Jerk around the sources of demand, encourage people to do things they didn’t see a benefit to before, and it’s that much harder to recognize the appropriate specializations for everyone.
Economists have gotten too locked into the mindset that “hey, if people are spending, that’s good”. But that’s just the result of the historical correlation of spending and productivity; it doesn’t mean that if you force up spending, things will be good.
August 23, 2010 at 2:38 pm
jsalvati
It’s not totally implausible to me that there are situations when an increase/decrease in the demand for money communicates a true signal, but I think it’s definitely not the case that all such increased/decreased demands for money correspond to true signals.
Think about the following situation: someone develops a new technology that significantly lowers people’s demand for holding money (maybe some kind of electronic payment technology). Without a balancing decline of the supply of money, the price level must rise. People will try to get rid of their excess money balances by spending more (both real and nominal), and this will eventually cause vendors to raise prices and people will reduce their real spending back down to normal levels. I would characterize this as inefficient; people’s incentives are distorted by having too much money floating around. How would you characterize this?
August 23, 2010 at 3:01 pm
Silas Barta
It seems like such an event would cause a one-time emptying of individual money stocks, rather than a permanent run-up in prices. A temporary increase in prices is exactly how you keep people from overloading merchants with one-time purchases.
(Perhaps I’ve misunderstood how this new technology works?)
I also don’t understand the skepticism to the idea that changes in levels of nominal spending can be meaningful. Trading goods for money is a conversion to the most liquid resource. In doing so, you take a cost to get that liquidity. If you could find a steady barter-partner, you could save a lot of money and increase efficiency (this is what friendships and marriages are). If the economy as a whole experiences a huge drop in spending because of recognition of these “no-intermediate liquidity” exchange opportunities, that is a gain in efficiency, and does not ened to be corrected by giving huge injections of money to favored parties.
August 24, 2010 at 10:16 am
jsalvati
I should have specified that the technology allows people to need less money on an ongoing basis. It is a permanent reduction in the demand for money.
As for my skepticism, truthful signals arise in special circumstances, like when the decision maker bears the effects of their decisions. We have a good understanding of how fluctuating demands for money have effects beyond the person making the decision. Thus I should be surprised if it turns out that the signals caused by fluctuating demand for money are something totally desirable.
August 24, 2010 at 12:40 pm
Silas Barta
I’m still not clear on the technology you’re describing. If you mean some kind of “universal constructor” device that allows you to manufacture your necessities without having to buy them like you usually do, the market will react (in the absence of compensatory monetary policy) by bidding down the prices of the things people would normally have to buy (and for which producers are now less efficient), and bidding up the prices of all other goods (which are now relatively more scarce). Note that demand can shift to goods that didn’t previously exist.
This is exactly what should happen; what’s the problem?
August 24, 2010 at 2:35 pm
jsalvati
No, that’s not what I mean. I mean a technology that means that people will have to hold less money. For example, when credit cards were first invented or some sort of technology that means people don’t need to carry around cash for the sorts of things they use cash for now a days. Or perhaps some sort of financial innovation that means banks need to hold fewer reserves.
August 25, 2010 at 7:52 am
Silas Barta
In the absence of knowing why people are holding less money, it’s hard to say what will be affected. It’s this blackbox model of money (not to mention the people using it) that I find problematic in the first place.
But per your latest comment, I refer back to my first answer — the only effect is a one-time flushing out of the current level of paper cash holdings.
August 25, 2010 at 7:58 am
jsalvati
Then there is some serious misunderstanding between us. Have you read the Selgin paper (http://www.cato.org/pubs/journal/cj10n1/cj10n1-14.pdf ) (20 pages)? I don’t think he discusses changes in money demand, but I think the framework might be useful for understanding my understanding.
August 25, 2010 at 8:02 am
jsalvati
Also, when you say “one time flushing out” do you mean their cash holdings are permanently lower (I agree)? The next question is, when they get rid of that money, how do they do so? I do not think it is possible for people to get rid of their money without eventually increasing prices (burning it seems to be very unpopular 😀 ).
August 25, 2010 at 10:11 am
Silas Barta
Yes, I agreed with that: because people are doing a one-time draw-down, prices have a one-time spike (perhaps distributed over time *because* the prices discourage immediate spending).
So what’s the problem again?
August 25, 2010 at 10:45 am
jsalvati
Putting aside the question of whether it’s a good or bad thing, do you agree that this flushing out of money holdings will eventually lead to an increase in all prices (given that productivity has not increased to completely balance this)?
The mechanism being, that people want to hold less money than they currently hold, so they spend more in order to get rid of their excess cash (or in a totally different kind of economy they might try to sell less). They might buy consumer goods/services or they might buy assets or some combination. This makes it attractive for suppliers to raise prices in the medium term.
August 25, 2010 at 12:22 pm
Silas Barta
Yes, this will lead to a temporary increase. Like I said, this solve the coordination problem by preventing everyone from blowing the money all at once.
Again, what’s the problem?
August 25, 2010 at 12:37 pm
jsalvati
Temporary increase in sales? Or temporary increase in prices?
I assert it will cause a temporary increase in sales and a permanent increase in prices.
August 25, 2010 at 2:23 pm
Silas Barta
What’s happening is equivalent to everyone consuming a few more resources than they otherwise would. In the sense that resources have contracted, the aggregate long-term supply curve has contracted, so somewhere, prices have permanently gone up, even if it’s divided evenly over the infinite future — and in that case, does it really count as a permanent price increase?
What if everyone takes the money from this draw-down to buy an extra taco today? Taco prices today spike. Does that also correspond to permanent increase in prices? Maybe, maybe not. Same thing for the scenario you described.
But I don’t see the point of this hairsplitting. So people bid up prices one time because of a new technology. The Fed should print money and make underpriced loans to big players to combat this because … ?
August 25, 2010 at 2:59 pm
jsalvati
I think we are talking past each other again. My point is not that because we have consumed more resources now, there will be fewer resources in the future, so things have gotten more expensive, though I don’t have any problem with that point. My point is that if people need to hold less money than they used to, the price level will rise. You can think of this as a change in velocity (though I normally don’t talk about velocity), people hold money for a shorter time on average before spending it. Since the need for holding money has gone down, money is worth less in terms of goods, therefore the price of goods in terms of money will go up.
Although generally macro assumes prices are relatively rigid in the short term (if everyone buys a taco it the price of tacos will stay mostly the same), it does not matter for this point.
Also, in this case, I would advocate the exact opposite of the Fed printing money, the Fed should take money *out* of the economy in this case ( which leads to a loss for the fed/government) so that people don’t spend more (though maybe this will lead to slightly increased consumption if it is a productivity increase).
August 26, 2010 at 6:39 am
Silas Barta
people hold money for a shorter time on average before spending it.
Well, then that’s where I disagree. If people simply hold a lower “stock” of money, the pass-through rate is unaffected; the only changes in flows are in the one-time “flushing out” from the higher stock holding to the lower stock holding.
And how does the Fed take money out of the economy. I know the standard answer is that it sells its bonds, but people still spend as if they had that money in reserve, and the bond has to mature, right?
August 26, 2010 at 7:31 am
jsalvati
I should note that what I am claiming here is not at all controversial; basically just the Quantity Theory of Money (in the long run). In any case, you can think of it like this, all money must be held by someone, if someone buys something, they got rid of money, but someone else got more. People can individually hold less money, but they cannot do so in aggregate without a change in the money supply. What actually happens when the demand for money falls is that rising prices causes money demand to increase (money demand is an increasing function of prices) until people want to the same original quantity of money.
I used to be confused by how the Fed takes money into/out of the economy as well. The answer is that, excess (or negative excess) money the fed has is given to (taken from) the government. If the Fed wants to take money out of the economy, it sells a bond and destroys the proceeds, the bond is paid back with tax money (which will come from elsewhere in the economy). The result is that the money supply changes immediately (though I don’t think this is especially important).
August 26, 2010 at 9:04 am
Silas Barta
Okay, then let’s say there is a permanent increase in the price level. What catastrophe results from not somehow reversing it through central monetary policy (of doing ____) ?
August 26, 2010 at 9:14 am
jsalvati
The basic problem is twofold, first people must undertake the costly process of finding the new equilibrium prices and adjusting to those prices. Second, people inefficiently consume more now (instead of later or consuming leisure). You can think of the second effect as coming from a burst of inflation, people are caused to spend money now because inflation will erode their excess purchasing power.
August 26, 2010 at 10:08 am
Silas Barta
And how does the policy you advocate (of ____) accelerate this process or somehow make it less painful? How do you know those intermediate “bad” prices serve no function?
August 26, 2010 at 12:57 pm
jsalvati
The policy I advocate (of adjusting the quantity of money to changes in the demand for money) reduces the need for a change in the price level, so people do not have to find new equilibrium prices. This also means people will not inefficiently consume more/less now because the amount of money people hold will be the amount they desire to hold.
We know that the price level is not inherently meaningful, but you could imagine that changes in the price level is meaningful. However, we already know why this change is occurring, and the explanation does not suggest that the change is important (to me anyway). It *could* be important, but you’d have to come up with an explanation within this framework.
August 26, 2010 at 1:25 pm
Silas Barta
“Adjusting the quantity of money…” is not a policy; it’s the goal of a policy. The policy would be to purchase bonds, etc. But already that assumes away the problem: that this monetary policy somehow introduces fewer distortions that the market’s normal reaction to the introduction of some new game-changing technology.
Yes, exogenous shocks affect how much money people will want to hold. But the new equilibrium won’t simply be “all prices scaled by some factor”; rather, there will be different relative prices, too. While some aggregate measure of prices might not be meaningful, the relative levels change when the over level changes.
You call prices that change in response to shocks “inefficient” and say that they can somehow be improved by a monetary policy that quickly bumps money’s value to some new equilbrium to save on some frictional relocations. But the monetary policy *itself* introduces dislocations as well, and you would have to somehow not that these aren’t worse than the existing adjustment costs.
Also, I don’t think you’ve answered the question of how you know the costs resulting from different money holding patterns *aren’t* sending meaninful, important signals. For example, if the economy suddenly lost its coordination, shouldn’t liquidity be at a premium and therefore dollar-costs throughout the economy be lower? Isn’t this exactly how economic actors know to do those things that restore coordination, as regards any exogenous shock?
August 26, 2010 at 8:26 pm
jsalvati
I want to note that you seem to have accepted the majority of my premise, that macroeconomics economics is not empty and that it suggests that it might be a good idea to adjust the money supply to changes in the demand for money.
Anyway:
1) I will readily concede that measuring inflation is difficult. If your monetary policy focuses on inflation, this certainly makes determining appropriate policy more difficult.
2) Yes, in order for it to make sense to adjust the quantity of money for changes in the demand for money the costs have to be lower than the benefits. I have already explained how adjusting price level is an expensive process. It might be the case that adjusting the money supply is an even more expensive process, but I don’t think I have ever heard anyone argue this before, and doesn’t seem to fit the facts (conducting open market operations doesn’t seem to be a very expensive process, nor do financial market transactions generally have high social transaction costs).
3) I don’t have a way to prove definitively that those costs do not have some alternative desirable function. However, we would none the less be surprised if those costs turned out have a desirable function. Why? Because we know what causes those costs, and it is not actual resource costs; in a barter economy or an economy with very flexible prices, you would not have analogous costs. It would be a fairly big coincidence (not to say they don’t happen) if those costs served some other purpose, independent from their cause.
As for your example, I can’t say I understand what you’re saying precisely enough to address it.
August 27, 2010 at 8:24 am
Silas Barta
I want to note that you seem to have accepted the majority of my premise, that macroeconomics economics is not empty and that it suggests that it might be a good idea to adjust the money supply to changes in the demand for money.
Not without a few additional jumps you need to spell out. Could you point to what I’ve agreed to and how it implies that it might be a good idea to adjust the money supply?
I have already explained how adjusting price level is an expensive process.
No, you’ve explained how it imposes costs on people. Whether those are “expensive” depends on whether you believe those costs reflect real scarcity. If you believe, as I do, that the premium on liquid assets reflects such a scarcity, than this is only an “expensive process” in the sense that it’s an “expensive process” to buy wheat after a disease has killed off the crop.
It might be the case that adjusting the money supply is an even more expensive process, but I don’t think I have ever heard anyone argue this before, and doesn’t seem to fit the facts (conducting open market operations doesn’t seem to be a very expensive process, nor do financial market transactions generally have high social transaction costs).
Those aren’t the (only) relevant costs. The costs I’m concerned about are the distortions introduced by selling goods into the economy. It’s not like the Fed shifts the value of everyone else’s money; like “normal” adjustment, the changes start somewhere and propagate through the economy. Any discoordination and misallocation resulting from these operations counts as part of its costs. If “no one’s argued this”, well, all the worse for their models.
And economists are starting to agree that previous open-market actions have had such huge costs, which is what they’re saying when they e.g. blame the housing bubble on Greenspan.
3) I don’t have a way to prove definitively that those costs do not have some alternative desirable function. However, we would none the less be surprised if those costs turned out have a desirable function. Why? Because we know what causes those costs, and it is not actual resource costs; in a barter economy or an economy with very flexible prices, you would not have analogous costs.
Yes, that’s exactly my point! In a barter economy with flexible prices (so flexible they overcome the coincidence of wants problem in most cases), people pay no liquidity premium. (I was actually planning a post about this.) If everyone instantly knew which trades to make to go to the Pareto optimum, there would be no need for money. The use of money reflects the inability to solve this difficult computation problem. So any time someone converts their goods to money before buying something, they’re incurring a cost to get that liquid good.
How *much* of a cost? That depends on the economy. In a well-coordinated one, the premium will be lower, while one that just experienced a bust, and so is discoordinated, will have a higher premium. This is the cost that people are responding to when they hoard dollars.
It’s also why the reasoning you and Sumner advocated leads to the absurd conclusion that we should punish or ban barter (like prohibiting people from cooking for themselves or getting goods via non-monetary exchanges), even if such economists quickly realize this absurdity and stop short of advocating it without knowing why. This barter is an efficient response to “hey, it’s harder to coordinate buyers and sellers, so if you see such an opportunity, use it”. Trying to suppress this signal leads to misallocation for the same reason trying to suppress any market signal does so.
August 27, 2010 at 10:24 am
jsalvati
Perhaps, I misinterpreted you. You started focusing on topics we had not been focusing on, so I interpreted that as abandoning that line of reasoning; my mistake. I certainly don’t doubt you if you say you disagree with me!
When I said ‘expensive’ I meant precisely “it imposes costs on people”.
1) This post is addresses most of your point: http://monetaryfreedom-billwoolsey.blogspot.com/2010/05/malinvestment-and-monetary-equilibrium.html . The basic idea is that if markets are efficient, when people hold more money, and the the CB issues more money to offset their holding more money by buying some assets then the effect is the same as if the original people who chose to hold more money had chosen to save by buying assets. This sends a signal to the market that more resources are available for use. The alternative way this signal can reach the market is through the slow and costly process of price adjustment.
2) It has always been my impression that “too lose money -> bubbles/problems that only show up significantly later” theories (as opposed to “too lose money -> inefficient spending/overworking etc.”) were necessarily ‘behavioral’ type theories, of which I am naturally skeptical. Thus I have not taken a close look at such theories. If you have links/references to an argument you think is particularly convincing, I’ll take a look.
3) What you say sounds like it might be and interesting point, but I think the word “coordination” is very much underspecified in this context.
4) Now, I readily concede that I am not sure how NGDP targeting interacts with shifts between market and non-market activities. It may be that it interacts well or it may be that it interacts poorly; I haven’t spent a lot of time thinking about this topic, so I just don’t know. However, it is far from obvious that monetary-equilibrium type thinking implies one would want to ban barter. A straightforward application would imply barter during a recession is good since more barter means lower demand for money (don’t need money for the relevant transactions), and lower money demand is helpful if you are on the “not enough money” side of monetary equilibrium. Remember to think in terms of monetary equilibrium.
August 27, 2010 at 11:10 am
Silas Barta
Perhaps, I misinterpreted you. You started focusing on topics we had not been focusing on, so I interpreted that as abandoning that line of reasoning; my mistake.
It’s not this assumption I object to, but your claim that not-endorsing a line of reasoning implies acceptance of something else. Please be specific.
1) If that’s the point, then I don’t need to read further — it’s non-responsive. You need to compare to the case where people hold cash and the Fed *doesn’t* offset the signals that this sends. (And that involves more than just assuming that non-moving money = the apocalypse.)
2) I’m not sure what the distinction is between inefficient spending and a bubble is — in either case, you have to unwind stupid investments that were only spurred on because of monetary policy, and this means unemployment and wasted resources later on.
3) I think I gave a sufficient operational definition of coordination: the ease with which supply and demand can find each other, which numerically shows up in the premium placed on cash. Discoordinative events include anything that renders existing production structures (factories, supplier relationships) incapable of satisfying (something contributory toward) actual consumer demand.
4) Well, I suggest you remedy this gap in your understanding, because if you only think in terms of money exchanges, and view that as the only relevant part of the economy, you’re setting yourself for a Goodhart trap, a Lost Purposes trap, an understanding below Level 2. As I keep emphasizing, you need to be able to justify policies to help the “economy” in a way that’s grounded all the way down at what the layman considers a “good economy”.
And my reference to banning barter was a reference to this part of our last exchange. Yes, economists are smart enough to see why it would be stupid. But they can’t articulate *why* it would be stupid, despite “getting that money flowing”; like you, they can only say that there are “better ways” to get the money flowing. But any model that tells you even this much … well, it needs to be fixed.
And I don’t see why I should think in terms of monetary equilibrium until you show me the full Level 2 connections for why I should care about it. I also don’t see why you think your favored economists would agree with you about barter being good during a recession. Imagine that all this time, the drop in spending had been due to people bartering for their goods. This would show up in the stats as lower consumer spending and (most likely) lower NGDP.
Are you telling me Scott Sumner would say, “Oh, well, this lower NGDP and consumer spending — well, I guess that’s okay because people are getting what they want, which is really the important thing, and monetary policy is just another means to that same end”? Or would he robotically say, “Low NGDP. Bad. Must make people spend.”?
Barter and money are not independent topics, and you can’t understand one without understanding the other. If you don’t understand barter, you can’t explain why someone sells their goods for money instead of other goods, and so you can’t account for the premium on money, and you end up thinking people are holding on to their dollars because of mental issues that we have to fix by flooding dollars into the economy.
August 27, 2010 at 12:58 pm
jsalvati
1) Perhaps I am misunderstanding you, but in case not let me try to be clearer. In the case where the money supply is not adjusted, in order for the relative prices of assets and the prices of the resources that go in to making those assets (which are mostly sticky) to reflect the true cost of making those assets, you must wait for the prices to come to their new equilibrium level. You will incur both the cost of of adjusting prices and the cost of having non-equilibrium prices. I thought that you had accepted this point, but perhaps I misunderstood you before or misunderstand you now.
2) I think this is a fair point. Your original point seems to be the topic of 3) so I am going to merge those threads.
3) I think I understand your idea a bit better now, and I find it interesting. However, I want you to notice a couple of weird things about it, and clarify some aspects.
3.1) The size of the required price adjustment isn’t proportional to the size of the discoordinating event; many kinds of technology that would reduce the demand for money would not necessarily affect the kinds of business activities that are profitable in real terms very much even if they affected the demand for money a lot (for example some kind of untraceable card that replaced the use of cash would decrease the demand for money by quite a lot, but would probably not discoordinate the economy comparably).
3.2) The costs of price adjustment are something a person inflicts on someone else, when a person changes their demand for money, they impose the cost of disequilibrium on the rest of the economy, not merely on themselves. It is unusual for externalities to work out so they counterbalance some other externality or otherwise give proper incentives.
You have also not specified a mechanism though which the cost of liquidity is determined by the level of discoordination in the economy. You have also not specified how the cost of liquidity is the same as the cost of adjusting prices. Can you clarify?
4) Think about it this way, barter is good during monetary disequilibrium because it relieves some of the demand for money, but barter is surely not the most efficient for of transaction. Offsetting an increased demand for money by adding money to the economy will relieve that demand for money without incurring the inefficiency of barter. You’re asserting a false dichotomy; barter can be better than the alternative of no trade, but be worse than the alternative of allowing people to spend more by increasing the supply of money (“barter good, low NGDP bad”). It’s right to say that my previous response was wrong or not well thought out.
It’s not that I don’t understand barter in general. It’s that I m not sure I understand how an NGDP-like rule will react to a massive (but not during a recession) increase or decrease in barter (such as from some kind of new home production technology or some such). The case of barter in response to monetary disequilibrium I understand.
August 30, 2010 at 7:48 am
Silas Barta
Please pardon the delay in response.
At this point, I want to try to identify where our disagreements have consistently been. I see two related areas:
A) Regarding frictional adjustments toward a monetary equilibrium (the issue in (1) above): You seem to be saying that “We have to reach a new monetary equilibrium anyway, so if monetary policy can do it cheaper than the regular adjustments the market would make, this makes everyone better off.” I disagree that the new equilibrium will be a simple shift in the price level, since relative price signals show up that convey important information. Monetary policy necessarily makes these signals less informative, and therefore prolongs readjustment, which would require that resources fall into a sustainable pattern of satisfying genuine consumer demand.
Therefore, I regard as unhelpful to think in terms of how monetary policy can avoid the pesky readjustment costs for the same reason I think it’s unhelpful to ask how policy in general can avoid the pesky higher (relative) prices associated with a failed crop — they must show up *somehow*; the only question is how well you can hide it.
B) Regarding barter as a reaction to (phenomena associated with) recession and NGDP contraction (the issue in 4): My position is that mainstream monetary economists (MMEs) have overcompartmentalized their understanding of an economy to the point that they ignore the reason for money in the first place.
This is fine when the relative merit of using money stays high (i.e. liquidity premium stays low), but “Goodharts” when it doesn’t, causing economists to make poor inferences — such as believing that it would alleviate the recession to ban “cooking for yourself”, even if they’re smart enough not to advocate it. Unfortunately, they can only justify rejecting this policy because they can find more efficient ones, *not* because they see the fundamental reason why this is destructive.
(Like in all areas, poor understanding that leads to such a failure mode *really* bothers me.)
Furthermore, I don’t see how barter somehow substitutes for or alleviates demand for money. In the absence of countercyclic monetary policy, and during a recession, people will want to hold more money and buy more goods through barter. This is because:
i) Money is more valuable since, unlike most resources at this point, it can definitely be converted into something satisfying consumer demand;
ii) People will want to swap out more of their purchases made with money, with some kind of barter;
iii) Liquid, near-cash goods (like savings account dollars) aren’t as attractive, because that money is seeking out uses that the economy can’t find for it right now, given the greater discoordination.
So it doesn’t follow that MME-favored policies will appropriately favor barter transactions, and so I dispute that you’ve appropriately characterized the role of barter in restoring equilibrium.
Please let me know if you consider this a fair characterization of our fundamental disagreements, and if so, how you reply.
August 30, 2010 at 8:50 am
jsalvati
Well, I think we are finally making good progress.
A) I think this is a good characterization, well put.
A lot of what you’ve been talking about in your last post really has to do with changes in productivity. Changes in the level of coordination can be usefully viewed as productivity changes. I urge you to read Selgin’s paper (if you have not already) which I linked to before as it is precisely about how the money supply should respond to productivity changes. I suspect you will like the answer, which is basically ‘by changing very little’. You could alternatively read Selgin’s 80 page book Less Than Zero (mises.org/books/less_than_zero_selgin.pdf), which I am currently reading, and covers the same material, probably more in depth.
I order to respond more fully, I think I have to understand why you think that a change in the demand for money will lead to relative price changes. Can you explain that more fully? I can understand if you think that changes in the demand for money will often be accompanied by changes in productivity, which will lead to relative price changes, but by themselves changes in the demand for money will not lead to permanent relative price changes.
B)
I am a little confused how you could disagree that barter is a substitute for money. Maybe there’s something I’m missing, but didn’t you just get done telling me how the reason that money transactions exist is because they substitute for barter more efficiently? You state this directly in B.ii : “People will want to swap out (substitute) more of their purchases made with money, with some kind of barter.” It is totally consistent for people to want to hold more money and do more barter during recessions and for barter to be a substitute for money. If someone’s demand for money went up for other reasons (say B.i or B.iii), then in order to economize on money they may substitute some barter for money transactions. If you have fewer money transactions, you will need less money than you otherwise would.
August 30, 2010 at 9:16 am
jsalvati
This nick rowe post is also relevant to the barter discussion: http://worthwhile.typepad.com/worthwhile_canadian_initi/2010/01/is-barter-countercyclical.html . I should have linked to it earlier.
August 30, 2010 at 10:49 am
Silas Barta
A) Before reading that, I want to first make sure it’s responsive to our disagreement.
I agree with the Nick Rowe post that you linked — that says it very well, except that I would object to his claim that it’s inconsistent with the recalculation story: money is more valuable because the economy is “recalculating”. (You recognize a particular barter as a good idea *because* of the incentives produced by a recalculating economy.) Since the values of resources are less certain, the relative value of those closer to satisfying a more stable consumer demand will go up. Money is the most valuable in this respect.
I think I have to understand why you think that a change in the demand for money will lead to relative price changes. Can you explain that more fully?
A change in the demand for money is necessarily tied to change in relative prices, since goods differ in the ease with which they can be traded for money. And this isn’t arbitrary (and thus something you could sweep away with an injection of money); rather, it reflects a real scarcity: the scarcity of “identification of production methods that satisfy consumer demand”.
B) I think this is an issue of equivocation between meanings of the term. Barter does substitute for money in the sense that someone will want to substitute a barter trade for a purchase with money. However, the kind of substitution I thought you were referring to, because its relevance to the cash hoarding you found harmful, was the kind where bartering disincentivizes holding money.
This kind of substitution, I claim, does not happen. From the fact that you discovered an opportunity to barter for (what would otherwise be) $100 worth of services, it does not follow that you should reduce your money holdings by $100. That’s like renting a hotel room that, it turns out, comes with free snacks, and then waiving the snacks because “I brought my own”. No, the $100 (or your own snacks) is still valuable, even though you discovered this windfall.
August 30, 2010 at 1:03 pm
jsalvati
A) I am afraid I am still not getting it. Your explanation seems like it might imply temporary changes in relative prices, but it doesn’t seem like it implies permanent changes in relative prices. Perhaps it would help if you explicitly addressed the example gave : “example some kind of untraceable card that replaced the use of cash would decrease the demand for money by quite a lot, but would probably not discoordinate the economy comparably”.
B) The no substitution claim I think is at odds with the description of the cause of barter you agreed with earlier. Doing more of your transactions though barter means having it is less important for your life than it otherwise would be, so on the margin you would give up less to get it, thus you have a lower demand for money.
The only way you could have no substitution effect is there was it was as or more efficient to do the transactions through barter than through money. These kinds of situations do crop up, but you’d have to have a reason for why they are more abundant during recessions.
Back to the original issue: while it could be that increased barter does not decrease the demand for money, it definitely does not *increase* the demand for money, so monetary equilibrium thinking does not naturally lead to the conclusion that it would be good to ban barter.
I want to point out that both Recalculation and monetary disequilibrium can be true at the same time, though I think monetary disequilibrium folks generally tend to talk about changes in productivity rather than Recalculation. I think Sumner and Rowe have both stated explicitly that negative changes in productivity/Recalculation is something that actually happens and probably has contributed to the current recession, they just don’t think that’s the primary source of current problems. I think Rowe’s point is that the cause of the increase in barter is monetary disequilibrium, which may be caused by a productivity shock, but that if the monetary disequilibrium were relieved you would not get the increase in barter. I think you acknowledge the first part at least, “money is more valuable because the economy is “recalculating””.
August 30, 2010 at 1:24 pm
jsalvati
Wait, I have a clearer example I think: Lets say we live in a world where everyone takes the bus to work. In this world, ever since buses were invented they have only taken cash, so in order to the the bus to work people had to hold a significant amount of cash for busfare. One day someone invents a method for using credit cards on buses. This reduces the demand to hold money substantially because busfare can be handled through short term credit. This means that the equilibrium price level is now higher than it used to be, and without an adjustment in the money supply, prices will have to rise. What is your theory for why this adjustment is a desirable process?
August 31, 2010 at 12:27 pm
Silas Barta
A) I admit, I see I may be in error here. But what if I argued that: This new credit card capability decreases the liquidity premium on money, thus decreasing money’s value relative to everything else? This would mean that productivity has gone up, but not expressed in dollar terms.
(Still, I think the setup of such problems assumes an impossibility, but I can’t quite articulate that yet.)
B) I still don’t see what’s unclear — if money has become more valuable, you don’t decide to ignore that additional value because you can get “enough” some other way. Why would you eat the hotel’s snacks in the example I gave (or save them for later), instead of just rejecting them? Well, that’s the same reason people don’t throw away value by holding fewer dollars than they can simply because they found another way to meet their needs.
Or look at it as being similar to Gresham’s Law: if people normally trade gold coins, and then it’s mandated that they trade fiat coins of base metal at the same value as gold coins, then obviously, they will use fiat coins for purchases wherever they can, and hold on to their gold coins (which trade above face value). But they don’t somehow reduce their holdings of e.g. 3 gold coins just because they got a windfall of 3 new coins.
Or to put it another way, what I claim is going on is that a recalculation (i.e. recognition that existing production methods don’t actually satisfy consumer demand) cause *both* a gain in relative value of dollars, and a gain in the return to barter.
I think Rowe’s point is that the cause of the increase in barter is monetary disequilibrium, which may be caused by a productivity shock, but that if the monetary disequilibrium were relieved you would not get the increase in barter.
And my point is that the “reason people value cash so much more” (which you say is monetary disequilibrium) cannot be eliminated by injections of new money, since they necessarily distort the relative price signals and disrupt the adjustment to a new equilibrium. The productivity shock can only be fixed by 1) identifying which non-money goods are most certain to satisfy consumer demand, and then 2) identifying which goods n-levels removed will contribute to those goods.
The higher relative prices of closer-to-consumption goods reflects a real economic phenomenon, and they are necessary for entrepreneurs to make the right decisions that re-establish a new equilibrium.
August 31, 2010 at 1:57 pm
jsalvati
A) I can understand that sentiment well, but I don’t share it. I am not sure how to prove that one it is correct or incorrect. Can I ask what what shifted your perspective?
If you see that this situation might be different, shall we proceed on the assumption that you are arguing that there are different kinds of changes in the demand for money?
I will naturally agree that it causes the equilibrium value of money to drop. I don’t think there’s anything that implies that the liquidity premium will drop, you will still have to make the same concessions for liquidity in percentage terms.
B) You have to think about *how* people increase their money holdings and *why* they hold it. They get it by either buying less or selling more. Shifting the quantity you hold is not free, you will have sacrifice consumption or make asset price concessions or pay interest if others are also looking to get more money. If you have a smaller need for money (for example if you managed to satisfy some of your needs though barter) you will be inclined to undergo less of the costly process of increasing your money holdings.
C) I am curious why you say closer to consumption goods have a high relative price. Can you explain that? Are you talking about the low price of financial assets (and other flexible price goods) or do you mean something else?
Keep in mind that I don’t advocate adjusting the money supply to adjust for the supply shock (selgin’s paper goes into why you wouldn’t want to do that, and might be compatible with your thinking); I want to adjust the money supply for the change in the demand for money. If there’s a productivity change (happens all the time), but no change in the demand for money the money supply should stay the same.
September 1, 2010 at 7:57 am
jsalvati
I have another example, maybe it will help clarify what you think is off about my examples.
Consider a world with two countries, A and B, who use different currencies. The two countries are close to each other, and have been trading for many years (so they are in equilibrium with each other and have exploited specialization and economies of scale). One year as a engagement gift to the princess of A, the King of B decides that his country will switch to using the currency of A. In order to remain in monetary equilibrium, the currency authority of A will have to print more money in order to accommodate the increased demand for A money. If they do not print more money, then prices will have to adjust downward.
September 1, 2010 at 8:46 am
Silas Barta
A/C) I’m going to have to call a time-out to read the selgin paper, since you’ve convinced me it makes an important distinction about when the money supply should increase, which means that, at least, it doesn’t start from a fundamentally flawed premise.
As for the “closer to consumption” point in C), my point was that in a recalculation/discoordination scenario, money is the most valuable (can be directly converted into something satisfying real consumer demand), widely used goods are next (corn, oil, ball bearings), inputs to make these are next, etc. This is exactly the relative price structure necessary to re-coordinate the economy, and loans to preferred players distort these signals..
I will give you my responses to the other points now though:
B) Okay, I can somewhat agree with that — as people satisfy their wants through barter, then, even despite the higher (implicit) value of dollars, they will “sell off” their dollars, reducing demand for money. My point is only that this is not a one-to-one substitution. If you’ve found that you can get $100 worth of goods through barter, that might cause you to trade away $50, not the full $100 you’ve saved. And this below-1-to-1 substitution rate represents the remaining level of discoordination, and will increase to 1-to-1 with time.
Re: 2-country example: I agree that country A would need to replace country B’s money units at some pre-defined rate to prevent avoidable inefficiencies. But this is critically different from monetary policy advocated by MMEs, which involves making loans to specific, concentrated players in the economy, distorting prices as they do so, and trying to get people to make loans for the sake of making loans.
September 1, 2010 at 9:03 am
jsalvati
OK, it seems like we are inching towards mutual understanding.
B) I agree, except about the part about discoordination, as you haven’t convinced me of this link yet.
Re: That’s a fair response; I can see viewing those as different.
I see you talk about loans a lot. I think a lot of people advocating increased credit are confused; you’ll hear a lot of monetary equilibrium types would refer to people “confusing money and credit”. Is this something that bothers you about loans in particular or do you have similar problems with having the Fed make asset purchases?
September 2, 2010 at 12:58 pm
Silas Barta
Good thoughts. I’m still on time-out to read Selgin, so I’ll start answering when I’m done with that.
Convincing me of the relevance of a MME paper *is* progress, so this is a good sign.
September 9, 2010 at 8:10 am
Silas Barta
Okay, I’ve read most of the Selgin paper, and it didn’t do much to change my mind, but it did help clarify where I think we diverge, which is that: I dispute that recessions are a case of a monetary disequilibrium that is not also a real disequilibrium.
That is, the relative price changes that happen due to a recession are not merely a friction that is pure waste; they indicate real changes in relative scarcities. (Selgin’s paper made a big point out of how the changes propagate slowly and uninformatively, but I think this is compared to an inappropriate baseline.)
I view a recession as a large-scale loss in knowledge of which activities (and uses of resources) really satisfy consumer demand, and that this is a kind of loss of productivity, but not equivalent to all losses in productivity. (For example, if tools degraded faster than expected, this is a loss in productivity, but would not lead people to infer that they need to fundamentally change patterns of economic behavior.)
In a recession, money increases in value relative to all other goods, and this is what needs to happen, and it should not be fought, for the same reason that you should not try to “fight” an increase in grain prices due to a failed crop — any method will just make the shock worse.
Money’s gain in relative value is just a special case of the more general increase in demand for more certain, “sure-fire” ways to satisfy consumer demand. Because money is the surest, it clearly must increase in relative value. But similarly, goods that are more general-use or liquid (oil, grain, multi-purpose equipment) also increase in relative value until people can find new sustainable production methods (what you might call a new monetary equilibrium).
Because a recession is a “shortage of knowledge” of what satisfies consumer demand, people spend effort to find (double) coincidences of wants that they would not otherwise bother to look for, which is why you see more barter, and the signal that tells them to look for barters is the higher relative value of money.
Monetary policy does not help this process. When the central bank introduces money through purchases, it necessarily introduces the money in specific places, where it distorts the relative price signals that are necessary to identify what the new uses of resources should be. Similarly, stimulus projects are by their nature temporary and so cannot point resources in the direction of sustainable, consumer-satisfying production structures.
As evidence for my position, I remind you that viewing the monetary disequilibrium as fundamental (which MMEs like Sumner do) leads one to advocate clearly absurd policies. Can we agree that, if your theory tells you that it would be helpful to:
a) Make ultra-underpriced loans to clumsy banks, and
b) Prohibit non-monetary-exchange-based ways of satisfying wants (e.g. cooking for yourself rather than paying someone else)[1]
then it’s made a big mistake somewhere? This point ties in with my broader theme that MMEs’ understanding has become far too decoupled from the fundamentals, from recognizing what is *really* meant by a “good economy”. Because they don’t recognize that a “good economy”, in the only sense we care about, satisfies individual wants, they only judge policy effectiveness by whether it meets certain metrics, even when those metrics no longer reliably measure what we care about.
Please let me know if you find this responsive, and if there’s anything I haven’t addressed.
[1] Yes, I know that they don’t advocate this; I just said they think it would be helpful, which is still a problem for the theory — it should say this is harmful, not just “a subpar way of solving the problem”.
September 9, 2010 at 9:29 am
jsalvati
I think this is a bit clearer, but basically the same thing I think you’ve been arguing, which is good.
A) Do you deny that monetary disequilibrium is an actually existing disequilibrium? My understanding of your position was that it was a desirable disequilibrium, not that it didn’t exist. Keep in mind that when the demand for money increases without a corresponding increase in the supply of money, the *equilibrium* value and price of money increases, but the *current* price of money does not (when prices are not completely flexible anyway) even if the value of money to people has increased.
B) I should say that I don’t find it totally impossible that money and information about “doscoordination” are tied together in the way you have described, but you haven’t really given any reason for why and how this would be so. I think you find the idea intuitively attractive, and I find it somewhat attractive as well, but unless you give a theoretical reason for why they are tied together as you postulate, I doubt I will change my mind on this.
C) I am not sure why you say ‘tools degrading faster’ is qualitatively different from other kinds of productivity shocks. If tools are found to degrade faster then expected, people will likely need to adjust for this; perhaps some activities will become unprofitable, or perhaps people must shift resources into tool production.
D) I will now make a point that is related to your point about “money increasing in value”, but that we have not discussed before. Money is the only good in the economy that is not traded in it’s own market. There is no single market price to adjust when the price of money is out of equilibrium, the prices of all the other moneygood_i markets must adjust.
E) No, we cannot agree:
a) I might agree that this might be helpful, but I think it’s irrelevant. There are crappy ways to solve any problem. For example, in the case of pollution externalizes, there are plenty of crappy ways to ameliorate the problem (say CAFE standards), but their existence doesn’t cast doubt on the theory of externalizes. So the existence of ways of ways of handling a problem that clearly introduce problems of their own does not generally cast doubt on the existence of that problem.
b) I am confused how we still disagree on this. I was under the impression that I had convinced you that even if you didn’t think that barter decreased the demand for money (though we never came to a satisfying explanation for why you rejected this), it certainly did not *increase* it.
F) I haven’t responded to your point about a ‘good economy’ before because I thought it was mostly a rhetorical device. Unless I misunderstand your point, it is easy to deal with. Monetary Equilibriumists simply think that monetary disequilibrium causes people to focus energy and resources on their desire for money (who’s quantity is easy to change) instead of real projects. They think monetary disequilibrium issues distract people from satisfying their wants at little benefit.
more B) I think the clearest way to advance the conversation is for you to explain why you think money and ‘discoordination’ are tied together as you claim they are.
September 9, 2010 at 11:14 am
Silas Barta
Starting with B), since you think it would advance the discussion the best: I thought I already did explain the link between discoordination and money. But to say it a different way: money is valuable because it can *definitely* be traded for something you want. Recessions happen when people realize that the current production structures (i.e. ecosystems built around this or that factory/method being used for this or that intermediate good) aren’t able to satisfy consumer demand. This is what I call discoordination: “Oh no, this factory is worthless.” “Oh no, this skill isn’t actually needed.” “Oh no, we have far too many people doing X.”
The existing patterns of specialization and trade are broken, and need to be changed, but people don’t know what they should shift *to*. So, the closer something is to satisfying that end demand, the more relatively valuable it is during the recession. Money gets the greatest premium, while the now-worthless capital goods get the greatest discount. And so as people begin to re-identify how farther-from-consumption capital goods can be used to satisfy genuine demand, this premium structure diminishes, and with it, the premium on money due to the recession.
(And so in response to D, yes, I understand that money is unique like that, so a change in money’s value is equivalent to a change in all other prices, as denominated in money. And for C, the difference is that with tools-degrading the necessary adjustments are not fundamental, involving fundamental re-organization of economy-wide production methods.)
A) I agree that it’s a disequilibrium, just that the real disequilibrium is more fundamental. So I think it’s unhelpful to look at satisfying the monetary problems apart from the real/coordination problem.
when the demand for money increases without a corresponding increase in the supply of money, the *equilibrium* value and price of money increases, but the *current* price of money does not (when prices are not completely flexible anyway) even if the value of money to people has increased.
Okay, then that’s what I disagree with, though only in the context of when the increased demand for money is caused by a recession. As the economy finds its new methods of satisfying wants, the premium on liquidity goes back down, so when equilibrium is re-established, the equilibrium price of money has not increased.
E) a) That’s not the appropriate comparison — I agree with you that “X is a crappy way to solve problem Y” does not prove that Y is not a problem, and I wasn’t claiming otherwise. What I take issue with, instead, is when you are in the position of saying, “Obviously-stupid X is *better than* doing nothing about Y [though still not as good as X2]”.
The analog in pollution would not be CAFE standards, but rather, pollution subsidies. If economists were in the position of saying, “Hey, I don’t think it’s the *best* policy, but subsidizing pollution is better than nothing!” then I would make the same critique — that whatever theory got you to that point has a big mistake, and you should be re-thinking what you mean by a “good economy”.
And that’s what I think is happening with MMEs — they’re in the position of saying, “Hey, a ban on home-cooking would be better than nothing” — even though it clearly diverges from what we really want when we talk about a “good economy”: efficient satisfaction of wants.
E) b) This is just an issue of sorting out what the equilibrium is relative to, and what is a response to what. Here is how the causality flows:
Discovery of discoordination –> increase in demand for goods that are sure to satisfy wants –> increased demand for money –> increased seach for barter opportunities –> more barter
Then, *as* these barter opportunities are found (and as the economy rediscovers sustainable production methods), people can “loosen their grip” on money, bidding its price back down.
F) Yes, certainly MMEs believe they are promoting a good correlate of what is meant by a “good economy” (here, the desideratum is “low resources spent on money” or something like that). But as we revealed in our earlier discussion about the role of barter, they do not understand the dynamics of *why* that correlation exists (i.e. why good economies will have people spending little effort adjusting money holdings), and therefore when propping up that correlate *won’t* actually cause a “good economy”.
In particular, I made the point about: what if NGDP hugely contracted because a) people found barter-type ways to satisfy most of their wants, and b) saved more of their money? Would Scott Sumner shrug and say, “Okay, NGDP is well below trend, but I’m completely fine with that, because people’s wants are being satisfied, and that’s what really matters, after all”? Or would he say, “Oh crap! Let’s debase the currency until people *have to* spend their money [and thus abandon the current, efficient arrangements they’ve made]!” ?
And by the way — this is no rhetorical device; being able to trace your policy justification all the way down to what a layman means by “good economy” is the difference between a real and a fake understanding.
September 10, 2010 at 8:18 am
jsalvati
I think we are talking past each other again. You are talking about temporary changes in the demand for money, and I am mostly talking about permanent changes in the demand for money (as in the bus example or the reverse). I would expect monetary disequilibrium from temporary changes in the demand for money to be undesireable, but my understanding here is much less clear than about permanent changes.
Permanent changes in the demand for money are both real and nontrivial. For example in today’s climate, both production and inflation expectations fell at the same time, prices were expected to fall (or rise less anyway) at the same time as production was falling and the Fed had added lots and lots of money to the economy. It is difficult to explain this without a permanent increase in the demand to hold money.
I’ll stick up for thinking monetary disequilibrium caused by temporary changes in the demand for money, but I need to organize my thoughts.
E) I think I must be missing something here, as both your answers seem like you are deliberately missing the point. Since I actually think this is unlikely, I will have to come back and try to understand this better.
I said your point seemed rhetorical because it’s a lot like saying “You should include all the relevant features of reality in your model!”: it’s true, but usually totally besides the point since usually your opponent already thinks they have done so; you disagree, but that’s what the argument is about in the first place. I don’t think this point adds anything to the discussion.
September 10, 2010 at 10:44 am
jsalvati
Perhaps a better way to make my last point is the following:
“MMEs’ understanding has become far too decoupled from the fundamentals, from recognizing what is *really* meant by a “good economy”. ” is basically stating that you think MMEs (using your term, I am not sure what the first M is supposed to be) need micro-foundations. Since Monetary Equilibrium is supposed to *be* those micro-foundations, I don’t think this is a constructive point. It would be constructive to say ‘you’ve missed feature X of reality in your microfoundations’ or ‘you have made mistake Y in your theory about what your microfoundations imply about the economy’.
September 10, 2010 at 11:42 am
Silas Barta
MME = mainstream monetary economist = economists thinking like Sumner or Blinder on monetary issues
You are talking about temporary changes in the demand for money, and I am mostly talking about permanent changes in the demand for money (as in the bus example or the reverse).
No: in my last reply, I limited my remarks to recessions, in which I claim that changes in the demand for money are (mostly) temporary. This is because you were talking about recessions too. Separate from that, I addressed the issue of permanent changes in money, like the bus case: I do consider it appropriate for the equilibrium value of money to go down (prices to go up), because the liquidity premium on money has decreased, although productivity would increase when expressed as output per unit of labor.
It is difficult to explain this without a permanent increase in the demand to hold money.
Which is why I didn’t dispute demand to hold money went up; I just claim that this is self-correcting as the economy re-establishes real equilibrium. And if increased demand persists even after this, so what? That, too, reflects genuine economic scarcity and serves a vital role.
I said your point seemed rhetorical because it’s a lot like saying “You should include all the relevant features of reality in your model!”: it’s true, but usually totally besides the point since usually your opponent already thinks they have done so; you disagree, but that’s what the argument is about in the first place. I don’t think this point adds anything to the discussion.
No, that’s not what I’m saying. I’m saying, “Make me care about your predictions for what would happen if we didn’t follow your advice.” Quoting from my blog post that started this:
In other words, why should I care if prices fall? If more people barter instead of contributing to the NGDP god? If inefficient factories have to close? If clumsy banks can’t make haphazard loans with cheap Fed money? Don’t tell me it’s to “help the economy”, because MMEs clearly have a different definition of “good economy” than most people, and I have yet to read an explanation for why I should care about theirs.
As for micro-foundations, I wouldn’t put my objection in those terms. The only “aspect of reality” that’s missing from their models is why I should care. What horrible consequence will befall me if the government doesn’t ban homecooking to prop up NGDP? What horrible consequence will befall me if the Fed doesn’t seize my savings through a round of think inflation? If incompetent banks don’t get free money?
Have I not already made that objection clear?
September 10, 2010 at 12:32 pm
jsalvati
Can you say more precisely what you mean by liquidity premium? I have my own understanding, but I want to see if we are on the same page.
I would still characterize what you say as ‘you need microfoundations’. I have already answered this question before (I think), you care because price adjustments are a costly process and because while prices are not at equilibrium people will have less than optimal consumption. I agree that economists should do a better job of explaining how macro phenomena related to micro phenomena (and I think I have said this before), but that’s not really relevant to the discussion at hand.
September 10, 2010 at 1:39 pm
Silas Barta
Can you say more precisely what you mean by liquidity premium?
The discount someone is willing to take by trading their goods for dollars rather than seeking out a direct double-coincidence of wants.
you care because price adjustments are a costly process and because while prices are not at equilibrium people will have less than optimal consumption
And I’ve answered this too: In short, “buying grain” is also a costly process after a failed crop, but it’s _necessary_ that it be costly, and attempts to make it less costly just shift the cost somewhere else. The premium on money (due to a recession) serves a purpose: it redirects resources from venues that aren’t supporting the satisfaction of consume demand, to those that are. Just like high grain prices after a failed crop.
I agree that economists should do a better job of explaining how macro phenomena related to micro phenomena (and I think I have said this before), but that’s not really relevant to the discussion at hand.
If economists aren’t saying this because they *don’t know how*, and because the usual assumptions for that connection to hold *aren’t true*, and they’re missing this because of a poor understanding on their part, all of which I hold to be happening, then I would consider that to be extremely relevant.
When Scott Sumner is unable to explain why it’s a good thing for the Fed to be making underpriced loans to failing, clumsy banks to get them to make their own questionable loans out of fear of their money losing value — that is, why the obscenely high costs don’t exceed the supposed benefits — then he is guilty of exactly what I am accusing MMEs of.
September 10, 2010 at 2:15 pm
jsalvati
Ok, now we are getting somewhere.
A) Right, so basically you are positing flexible prices, but the whole reason why monetary disequilibrium can have real effects is that not all prices are fully flexible. Many asset prices are near perfectly flexible, but most goods prices are much less flexible. If all agents simply charged a liquidity premium monetary equilibrium would come about very quickly and increases in the demand for money would not be an issue. When some agents do not do this then you get real and inefficient effects while those prices adjust.
B) I didn’t say you had not answered. You are basically saying the MME answer is wrong, which is fine, but then asking them to answer the question is not a very useful thing to do, because they think they’ve already answered. You should really focus your attention on convincing them they’ve made an error.
September 11, 2010 at 8:46 am
Silas Barta
A) Right, so basically you are positing flexible prices, but the whole reason why monetary disequilibrium can have real effects is that not all prices are fully flexible.
Wrong on both counts. My position is that:
i) The “monetary disequilibrium” (which I believe is a wrong way to view the problem) is not independent of the real disequilbrium, so we should *expect* there to be real effects, and these real effects are necessary to re-coordinate the economy into satisfying actual consumer demand through sustainable processes.
ii) My point doesn’t require flexible prices — those are just another cost that have to be worked around in either case; monetary policy can only hide them.
B) If their models assume away or neglect the inefficiencies of clumsy banks getting free money with which to make hastily-considered loans not economically justified (in the absence of this jawboning), then there’s not much I can do except say, “Sorry, you made a big mistake somewhere. You’re unfairly ignoring the inefficiencies from implementing your solution, which have to be compared against the inefficiencies you’re trying to solve.”
Remember, I had pointed out that Sumner’s position would imply that the Fed should also make 0% interest loans to individuals, and he agreed. But if you believe this is a good idea, you’re claiming that the interest rate plays no role, and interest costs can be eliminated without real effects, which would be overturning most of economics as we know it.
September 11, 2010 at 12:00 pm
jsalvati
i) I think you should keep in mind the bus example. This was a case where there might be a significant monetary disequilibrium without a significant real disequilibrium. You said that it seemed like I had said something self contradictory, but you have not yet shown that to be the case.
ii) OK, then there’s something in your argument I am confused about, but I am not sure what it is. Let’s see if we can find out what it is. Earlier you said that an increased demand for money “reflects genuine economic scarcity and serves a vital role”, a) I think you would say the ‘genuine economic scarcity’ is the scarcity of deals that satisfy consumer demand, is that right? b) What is the vital role it serves? I think you would say that it signals to people that the liquidity premium has gone up, is that correct? c) Who does your model imply will have increased money demand when there is a productivity shock? d) My model implies that part of people’s reaction to holding less money than they would like will be to cut back on purchases that were successfully ‘satisfying consumer demand’; do you deny this? If you do, where do see people cutting back?
Help me understand your model better.
B) I think you’re just being argumentative here. I have no problem with the point you’ve made here (other than disagreeing). My point in this thread was not that you were wrong in general, just that the particular point you had repeated (that MMEs needed to trace their theory down to laymen conceptions of what it means to have a “good economy”) was not a useful one. It was at the very best a restatement of the fact that you disagreed with MMEs on microfoundations without specifying how, and seemed like a way of scoring points without advancing the conversation.
September 12, 2010 at 9:27 am
jsalvati
I have realized that your concept of liquidity premium is not quite the same as the finance concept of liquidity. The finance concept of liquidity is fundamentally about the cost of doing trades immediately vs. stretching them out. All trades (immediate and delayed) could be in cash or they could both be barter trades. Perhaps your concept would be better named ‘cash trade premium’ or something? Tell me if I misunderstand something.
I also cannot shake the feeling that frequently when you talk about liquidity premium (cash trade premium) you mean some related but somewhat different idea from your definition and from your usage at other times, but I am not certain what it is.
For example in passages like:
1) “I do consider it appropriate for the equilibrium value of money to go down (prices to go up), because the liquidity premium on money has decreased, although productivity would increase when expressed as output per unit of labor.”
2) “This new credit card capability decreases the liquidity premium on money, thus decreasing money’s value relative to everything else”
A change in the cash trade premium seem like a natural consequence of a change in the equilibrium value of money, but you seem like you’re trying to convey a more substantial point since you’re trying to explain why the change is good. An arbitrary increase in the supply of money would also decrease the cash trade premium, but I don’t think you would regard arbitrary increases in the money supply as appropriate. Am I misunderstanding you? Other thoughts?
September 12, 2010 at 10:32 pm
jsalvati
I should also mention that in finance liquidity is a feature of a particular market, not the economy as a whole. You could have correlated changes in the level of liquidity in many markets, however.
September 13, 2010 at 8:13 am
Silas Barta
Starting with the first post:
i) I thought I made very clear that a disequilibrium due to a recession (which, in turn, is a *kind* of productivity shock) is different from that due to changes in how people pay, which is why I said there should be different consequences, and why there should be different consequences.
ii) To answer your questions:
a) Almost right, but I would say it more like, “scarcity of knowledge of what processes need to happen to satsify consumer demand”.
b) The premium on things “more certain to satisfy consumer demand” reflects the above scarcity, and the premium on cash is a special case of that.
c) Everyone, but less so if they if they are owners of resources more likely satisfy consumer demand.
d) I’m not sure I understand why you’re saying people would want to hold less money; I thought we were talking about the case where their demand went up. In that case, the economy has become less productive, and it makes sense for people to cut back consumption.
B) You already admitted that you don’t know how Sumner’s model handles barter (i.e. how to identify “acceptable” contractions of NGDP that are due to people satisfying their wants through barter and saving more). That is a serious problem for any claim that it’s based on a genuine understanding, as it cannot connect the reason for caring bout NGDP to the reason for money’s existence. This remains a problem even if Sumner thinks he has convincing reasons why his policies pass a cost/benefits analysis.
Next: “Liquidity” might not be the right term to use, but I hope I’ve made clear exactly what good people are getting when they trade their goods for cash rather than directly for other goods: they save on search costs, and increase the probability that they will find a merchant willing to trade that merchant’s goods for what the cash-holder has.
Hope you find that responsive to your latest posts.
September 13, 2010 at 10:31 am
jsalvati
i) fair enough.
c) Can you explain how the productivity shock leads to an increase in the demand for money for everyone? Do you include people who were not producers or consumers of the good that has the productivity shock in ‘everyone’?
I will give you an example of my perspective hopefully that will help you recognize my source of confusion. It would be helpful if you answered in this question by modifying the example, but not necessary if you think it’s not a good idea.
Consider an economy where granola bars are produced from only wheat, wheat producers sell wheat to granola bar producers. Lets say that granola bar production becomes impossible (aliens steal the world’s supply of the granola-bar-reaction catalyst). Prices in the economy are sticky. The producers get and spend less money (lets say their profit was 0). The consumers have extra money since they can no longer spend on granola bars; they can either spend their extra money right away, or they can hold on to it while they decide what they want to spend it on. In the first case their demand for money stays constant unless the new transactions they engage in are more or less money intensive on average than granola bar purchasing (perhaps, their new transactions are cash-only, or the other way around). In the second case, their demand for money increases unequivocally. Wheat producers get less money. They must make the opposite kind of decision that consumers did, they may either cut back on their expenditures (wheat tree growing inputs) right away or they may hold them constant while they decide what to do. If they cut back right away, their demand for holding money goes down. If they wait a while their demand for holding money also goes down. The wheat grower suppliers must make similar decisions. The net change in the demand for money is ambiguous.
I expect that your answer has to do with uncertainty about how consumers will choose to shift consumption.
d) I think you misread me, but you interpreted me correctly anyway. My question was one I had asked before differently: I think that when people cut back to increase their money holdings they will not necessarily only cut back on things that have lost productivity, do you also think this is the case? If not why not? However, I am not sure this question is going to advance the conversation at least until I build a more detailed question, so feel free not to answer.
B) I am not sure how Sumner’s *rule* would handle a non-recession shift to non-traded production; probably not well. Sumner’s particular monetary policy rule would not not handle a big change in the usage of competing private currencies either. But this is different from saying that Sumner’s model (presumably monetary equilibrium) doesn’t know how to think about such a change. I didn’t know earlier because I hadn’t thought about it a lot, but I think I know how to think about it now. A secular increase in barter would probably lower the demand for money and thus call for a decrease in the money supply (if it did not then it would not call for decreasing the money supply). An increase in barter due to people being short on money is symptomatic of an increase in the demand to hold money, and calls for an increase in the money supply. Both these facts follow from the idea that barter is a substitute for money-facilitated transactions.
I agree that Sumner’s rule is probably not be the very best possible monetary policy rule, but I do think it (or some similar rule) is likely the best *known* policy rule.
Liquidity) Yes, I agree with what you have said. I propose we adopt the phrase Cash Trade Premium in order to avoid false associations. I think I understand your argument more in some ways and less in other ways now.
You have stated a few times that you consider it appropriate for the price level to drop if there is an increase in the demand for money because the Cash Trade Premium has risen. This seems almost tautologous to me, and in any case does not communicate your substantial point to me. Why do you think it is good for the Cash Trade Premium to be at that higher level; why should it be nonzero? At first I thought this was related to the reason why you think the demand for money rises when there is a productivity shock, but I don’t think that’s the case, since you stated that it’s also appropriate for the price level to rise in the case of the bus example. My argument has been that there are costs to changing prices, and your argument has been (I think) that there are corresponding benefits; lets drill down into how those benefits come about. I think this question is more central to our disagreement than my question about productivity->increase in demand for money.
September 13, 2010 at 10:50 am
Silas Barta
Before I reply, was there more to your comment that was cut off?
September 13, 2010 at 10:57 am
jsalvati
Oops, I was moving things around and forgot to delete the last passage. I have deleting the offending passage.
September 14, 2010 at 8:09 am
jsalvati
Ok I think I found a way to ask my last question a bit more specifically. I think at least part of your argument has been that a higher Cash Trade Premium increases the rate at which ways of satisfying consumer demand are found. How does this happen? You have mentioned barter several times; why do barter trades find new ways of satisfying consumer demand faster than cash based trades? Why is this necessary in the bus example (I like to focus on bus example because I think that is where our disagreement is most stark)?
September 15, 2010 at 2:03 pm
Silas Barta
Sorry for the delay in responding, there are a lot of issues here to separate; and I was just about to answer when you made comment 57.
I think most of the misunderstanding between us is in what the direction of causality is. For example, about your last comment, I’m not saying that higher CTP increases the rate at which consumer demand-satisfying methods are found; rather, the shifting supply/demand curves do that, like in any market out of equilibrium. I’m claiming that the higher CTP is an *effect* of this more general phenomenon, and looking for barter opportunities in that case is a rational (and efficient) way to respond.
To explain why discoordination leads to this, let me explain the phenomenon in more detail and how it differs from other productivity shocks. Let’s say the market has settled on method M1 of satisfying a very common consumer demand (where M1 could be a specific kind of factory, a business model, etc.). And since that demand is stable, people can reliably go into business doing method M2, which augments and amplifies the effectiveness of M1. And into method M3, which augments M2. And … M15, which augments M14.
But then it turns out that M1 “stops working” — either because the people stop wanting that output, one of its other inputs is no longer obtainable, or there’s too much uncertainty about how it will be regulated. In that case, that whole “ecosystem” built up around M1, which simply isn’t viable anymore — and the people working in M13 might not recognize their dependence on M1. Instead of supporting a non-functioning structure of production, people need to shift to things which *do* satisfy demand, and so it’s necessary that these obsolete modes liquidate, and for the “economic search” to restart from “What is *known* to satisfy consumer demand?” and work outward from there. (And government policies — including the ’08 bank bailout — tend to prop up businesses with questionable relationship to satisfying consumer demand.)
To summarize the difference between this and other kinds of productivity shocks, I’d put it like this: simple productivty shocks (i.e. buses changing methods of payment, tools needing more maintenance, etc.) simply require that people recognize that they’re not longer at a *local* equilibrium, and move to it. In contrast, in a discoordination, people most move out of their “domains of attraction” entirely, and to a more globally optimal point.
I want to offer an example of my own to show the problems of treating discoordination with injections of money: For any problem with monetary disequilibrium you find, I want you to replace “money” with “wheat” and see what happens. For example, if there is a failed crop, then “wheat is tight” and there is a higher premium paid for wheat and wheat-related goods (just like there’s a higher premium paid for money and other liquid goods in monetary disequilibrium) . And in that case, the government can’t promote efficiency by “injecting wheat”. Why? Because it has to buy the wheat from somewhere else using tax revenue, shifting the inefficiency elsewhere. Moreover, the injections cannot last forever, and stunt the growth of more long-term, sustainable solutions for providing wheat. Plus, consumers are deceived about how much more they need to economize on wheat.
So why does it become a good idea when you switch from wheat to money? I know you see the price level as arbitrary, but the same things happen: the injections of money inhibit the formation of sustainble modes of production, as well as the economizing that needs to happen.
And to return to your point B) (the issue which most concerns me and is the source of my frustration): Do you at least agree that it was premature of you to conclude that Scott Sumner’s position is extremely insightful and, and that it gave you a deeper understanding of monetary economics, given that you were not, at the time, able to see the inferential connections between his reasoning and the role of money and barter? Even if Sumner can integrate his view with that phenomenon (and he’s given no reason to suggest he can), that suggests that you did not appropriately account for the “fundamentals” — the things we care about when we mean a “good economy”. So your enthusiasm for Sumner’s analysis is misplaced.
(I keep bringing this back to the point in B because this is fundamentally what bothers me: not the mere fact of counterintuitive economic policies, but the complete lack of an understanding among MMEs that allows them to ground their pursuit of a “good economy” in what we care about, and to know when crucial assumptions stop holding.)
September 15, 2010 at 3:54 pm
jsalvati
OK, I am starting to understand you better, but I don’t think I’m there yet.
Before I respond, can you clarify something? Do we agree that in the bus example a reduction in the money supply is good (and more generally counterbalancing permanent changes in the demand for money)? I thought you did at first, but then you said something that made me think you didn’t agree. Can you clarify? If you do agree here, I think our agreement is much larger than we had been thinking, because NGDP targeting is primarily about counterbalancing permanent changes in the demand for money (Sumner clearly says he wants to target ~2 year NGDP expectations). I think it might be rather difficult to effectively counterbalance temporary changes in the demand for money. If you do disagree then I think many of the other issues we have been discussing are not central to our disagreement (though they may be interesting in their own right).
September 16, 2010 at 8:08 am
Silas Barta
Yes, I agree that a reduction in the money supply is good, because that is what results from people responding to the shifting supply and demand curves. It does not, however, follow that it’s a good idea for the governemnt to “sop up” the extra money, for the same reason it’s not a good idea for the government to increase its wheat purchases when wheat is overproduced.
September 16, 2010 at 8:14 am
jsalvati
I am confused by your statement. How does the money supply change from a shift in supply and demand curves? The demand for money changes, yes, but not the supply.
September 16, 2010 at 9:10 am
Silas Barta
I mean the demand and supply curves change for various goods, not specifically money.
September 16, 2010 at 9:50 am
jsalvati
I interpret you as disagreeing that a decrease in the money supply is good, and that you think it’s better if the demand and supply curves for goods shift out in response to the change in the demand for money.
Here’s the rest of my response:
I assume you agree that a small productivity shock in one sector can cause a massive shift; if producing batteries gets very slightly cheaper, now suddenly it doesn’t make sense to have gasoline cars etc. . If you do, I don’t disagree with your logic here, and it’s and interesting point. However, you have still not a) connected this type of productivity shock to a change in the demand for money or b) explained the benefits that this increased demand for money brings (that would be eliminated by increasing the quantity of money). Lets focus on b since this is where our disagreement is clearest and most important (I am only moderately skeptical of a).
Perhaps the following description will help you understand and address my skepticism. When you describe uncertainty based temporary increases in the demand for money, this sounds like an increased demand for savings (transferring resources into the future). However, when people cut back on their consumption to accumulate more money, they reduce their use of resources without allowing anyone else to access the forgone resources. If someone forgoes buying granola bars to accumulate money, someone else can buy those granola bars, but they must give up buying something else (meaning some other good is now in the situation granola bars were), or lower their money stock. Unless someone reduces their stock of money, some resources will go unused, even for projects to try to move those resources into the future. If on the other hand someone forwent buying granola bars by saving and buying assets (stocks or bonds etc.), they would make those resources available to others, including for projects that shift the resources in to the future. The most natural way to resolve this difficulty is for the money supplier to buy assets in the market, giving the same result as if the money saver had bought assets in the market.
I think your example ignores 2 of the critical features of what makes money different from other goods. 1) money, unlike wheat, money is basically free to produce; if wheat was free to produce, do you think we should not produce more if we have less than we want? 2) you seam to imply that these wheat related markets have flexible prices, but the whole point about how changes in the demand for money without changes in the supply of money are inefficient comes from the fact that many markets do not have flexible prices so that money/wheat will not trade at a premium in those markets. Do you see why I would find this example completely off topic?
B) Perhaps. If by ‘Sumner’s position’ you mean ‘we should target NGDP’, then sure, we can agree targeting NGDP does not ensure we have optimal monetary policy. If you mean Monetary Equilibrium, then no, we do not agree. Monetary (Dis)Equilibrium is a real thing; you agree with that, but you say that there is an additional feature of reality that MMEs have missed so Monetary Equilibrium is not the whole story. You find this feature obvious, but I (and apparently others, since I haven’t seen this theory before) do not. When we agree, if you turn out to be right, we can agree ‘Silas is smarter than John; some things that obvious to Silas are nonintuitive to John’ (the reverse is not true, however, since I didn’t invent Monetary Equilibrium).
September 16, 2010 at 11:15 am
Silas Barta
I interpret you as disagreeing that a decrease in the money supply is good, and that you think it’s better if the demand and supply curves for goods shift out in response to the change in the demand for money.
No, that’s an oversimplified version. Money doing “whatever supply and demand curves suggest they do without artificial manipulation of the restorative processes” is good — if that means money contraction, that’s good *as an implication of this*.
The most natural way to resolve this difficulty is for the money supplier to buy assets in the market, giving the same result as if the money saver had bought assets in the market.
No, because the results are quite different; the supplier distorts the restorative processes that would have corrected the granola disequilibrium and returned the economy to sustainable patterns of specialization and trade.
1) money, unlike wheat, money is basically free to produce; if wheat was free to produce, do you think we should not produce more if we have less than we want?
This is not a relevant similarity. Tokens for *representing* money are free to produce. However, money is really information. The economic information conveyed by money — the relevant quantity here — is *not* free to produce. Printing money and loaning it out is not some free, helpful gift; rather, it degrades the information contained in prices (including and especially the price of money). Because relative prices shift during a recession, in proportion to the distance of the product from satisfying an actual consumer demand, and because money is on the far end of this scale, adding money in arbitrary places distorts the information contained in prices that are necesssary to restore equilibrium and economic efficiency.
2) you seam to imply that these wheat related markets have flexible prices, but the whole point about how changes in the demand for money without changes in the supply of money are inefficient comes from the fact that many markets do not have flexible prices so that money/wheat will not trade at a premium in those markets.
My argument does not rely, in any way whatsoever, on wheat related markets having flexible prices. If wheat markets are very slow to adjust, that *still* doesn’t support the case for government injections of wheat. Do you at least agree with this much?
B) Perhaps. If by ‘Sumner’s position’ you mean ‘we should target NGDP’, then sure, we can agree targeting NGDP does not ensure we have optimal monetary policy.
More precisely, by “Sumner’s position”, I mean Sumner’s (IHMO weak) understanding of economics that causes him to believe that NGDP pumping (specifically, making it grow 5%/year, a number he got from God) is some kind of cure-all, without being able to identify in which standards this metric breaks down. If this is his level of understanding, that casts all of his recommendations into doubt, and you can reliably predict it will lead him to suggest obviously-stupid policies (like making 0% interest loans to households) and to misjudge policies he dosen’t advocate (like banning household production).
This bothers me.
If you mean Monetary Equilibrium, then no, we do not agree. Monetary (Dis)Equilibrium is a real thing; you agree with that, but you say that there is an additional feature of reality that MMEs have missed so Monetary Equilibrium is not the whole story. You find this feature obvious, but I (and apparently others, since I haven’t seen this theory before) do not.
Okay. Then let me put it this way: to change my mind, you must *make me care* about monetary disequilibrium (MonDis). Right now, I don’t care, because I see all of them (in the context of recessions, NOT the bus example) as simply artifacts of real disequilibrium, in which the economy is extremely discoordinated and needs fundamental shifting of production methods.
Here’s what you need to do to make me care about MonDis: explain what’s wrong with an economy (that you regard as being in MonDis) *without* reference to MonDis — in other words, how it deviates from a “good economy” in the sense of “good economy” that normal people care about i.e. getting the maximal (and desired) combination of consumption and leisure that they want. (For example, “low NGDP” is *not* something
September 16, 2010 at 11:20 am
Silas Barta
Sorry, too long I guess. Here’s the rest of my reply:
(For example, “low NGDP” is *not* something normal people care about, as it’s screened off by those other measures.) Then explain how your proposed solutions would fix that problem.
If it becomes increasing clear that you have a coherent explanation and solution for the problem, which would become easier to explain by defining and using the concept of “monetary equilibrium” (or some isomorphic concept), then — and only then — will I agree that I must care about this concept as differentiated from the “real equilibrium” that I’m concerned about.
Let me know if this helps to clarify our differences.
September 16, 2010 at 1:40 pm
jsalvati
First a partial response:
I think perhaps I see the source of our different intuitions. You say: “Tokens for *representing* money are free to produce. However, money is really information. The economic information conveyed by money — the relevant quantity here — is *not* free to produce. “. I think this is a serious mistake, money is not information, money is physical currency or the appropriate bits at a computer at the central bank (we call these separate things the same because they are freely interchangable and useful for the same things). You can claim that money transmits information, maybe even always, but you can’t simply assume that; it’s what our argument is about in the first place. And needless to say, it would be an abuse of language to redefine ‘money’ as ‘X information’. I am not even sure the phrase ‘money is information’ makes a lot of sense
For example, “(Money is) the relative amount that the world (believes that it) owes you” is false. You can be extremely wealthy (say in the form of bonds) and yet hold zero money.
I don’t have a problem with these kinds of statements in principle, but they are not definitions, so you must show that they hold, not simply assume it.
September 16, 2010 at 2:12 pm
Silas Barta
You’re already showing that you hold that position that money is information when you agree that “money is … the appropriate bits at a computer at the central bank”. That’s an example of an institution (like an MMORPG) assimilating the insight that money is information.
Even for physical currency, it is the information *represented* by the currency that matters, not the paper. If someone stole your paper dollars, gave you blank pieces of paper, you would not feel compensated. If you were issued new dollars, and the bills the thief took from you were somehow rendered “unspendable”, you have not lost any money — because the public knowledge of your net debt is untampered.
Your example of the bond-rich individual does not prove what you think it does. He is not owed $X worth of stuff until he trades the bonds (incl. at maturity) for $X.
And irrespective of all of that, it remains true that the *value* represented by the money is *not* costless to create. Printing $1 trillion, for example, does not mean we have goods worth $1 trillion more.
In any case, do you seriously think these distinctions matter for the point you’re making? Even if we disagree on this point, we really shouldn’t dwell on it unless it’s actually relevant to your advocacy of the position that printing up money and loaning it to favored players can be a welfare-improving move over the long term. For example, what *exactly* do you regard as costless about money production, and how is *that* cheapness leveraged by monetary policy to make us all richer.
September 16, 2010 at 8:38 pm
jsalvati
I attacked this position because I have been looking for the source of our differing thought processes on this topic. Your statement “money is information” sounded a lot like an idea that could lead to many of your views that I strongly disagree with via conflating various subtly different kinds of information (I apologize for accusing you of this if this is not the case).
I think you are conflating different kinds of information when you say money is information and use this to argue things. You could say that money represents *A* quantity that the economy owes you, not the full quantity as you seem to have implied. We can increase or decrease *A* quantity that the economy owes you without trying to increase or decrease the total quantity of wealth. Thus statements such as “Printing money and loaning it out is not some free, helpful gift; rather, it degrades the information contained in prices (including and especially the price of money” do not necessarily follow.
The following is an attempt to most directly query the reason for our disagreement:
I want to zoom in on a variant of the bus example because we seem to agree on the facts, but disagree on the desirable outcome. Lets consider the opposite of the bus example: aliens steal all of the rare compound necessary for operating bus card readers, so we must all switch to using a cash based bus system, and people must hold more money for this. The equilibrium value of money/price level is now lower than it would otherwise be. In order to accumulate more cash they reduce their expenditures and sell assets. These actions by an agent reduces the cash balances of other agents and also reduces the first agent’s need for cash. Everyone’s trading is reduced to the point where the total demand for money matches the available supply of money. This is inefficient because trades which still make economic sense (neither preferences nor production capabilities have changed very much) are not undertaken and because people must change all the prices in the economy. Alternatively, the Supplier Of Money (the Fed or otherwise), can decide to conduct an open market operation, trading a newly created quantity of money equal to the change in the demand for money for the same dollar value of assets (bonds stocks etc.). To make the example simple, lets say that all the agents who are trying to increase their money balances recognize the Supplier Of Money’s action as the easiest way to increase their money balances, and get the amount of money they want by selling some quantity of assets they hold to the Supplier Of Money. This means that those trades that were forgone before take place as before the aliens came and prices do not need to be changed, a welfare increase over the alternative.
I take it that you see some costs in the money supply increase case which I do not see which counterbalance the benefits, what are they?
Until we can agree about this situation, I doubt we will be able to agree about more complex situations.
September 20, 2010 at 8:26 am
Silas Barta
Please pardon the delay in response.
I think I know what the problem is with using your bus example as an illustration of frictions due to money shortages and how monetary policy can efficiently fix them.
Basically, in the example, you are positing a lot of “knowns”. That is, for all this extra money that’s needed, people know what it will be spent on, the suppliers of the bus services, know that the demand for the use of the buses will come, and the central bank knows which people temporarily need this cash.
This abstracts away the major purpose of money, which is that you can sell goods without knowing what good you will ultimately exchange it for. The additional money therefore does not help in this situation, in which people know they (indirect) exchange they will make, and could, in principle, adapt without additional money.
(For example, since they know what the money will be spent on, they could just issue bus passes where people make a one-time purchase to cover multiple uses of the bus. Predictably, upon entering the bus, all they need to do is make an informational adjustment — how many bus rides have been debited.)
Because your example does not show money being demanded for its unique purposes as money, I do not think it can be used to derive implications about the effects of monetary policy in a sitution where this increased demand for money is (as I have argued) an artifact of the greater uncertainty (discoordination) in the economy, and in which the higher price of money encourages effort toward re-coordinative actions; and in which the cental bank’s decisions about which people will get money fundamentally changes the resulting (transient) coordination, and therefore how sustainable it will be.
September 20, 2010 at 8:28 am
Silas Barta
Sorry, poor proofreading. The third paragraph should read:
“This abstracts away the major purpose of money, which is that you can sell goods without knowing what good you will ultimately exchange them for. The additional money therefore does not help in this situation, in which people know what (indirect) exchange they will make, and could, in principle, adapt without additional money.”
September 20, 2010 at 11:26 am
jsalvati
I think I am making headway.
I don’t think it’s correct to say that “you can sell goods without knowing what good you will ultimately exchange it for” is the “unique purpose of money”. Certainly it is a benefit of money, but I would say the “unique purpose of money” is that you can buy and sell goods without having to know what *other* people want, which applies to this model.
On whether I posit lots of knowns:
The extra money is not known to be spent on bus transit. It could easily be the case that bus transit is extremely important to people (say everyone takes the bus to work) and they are very reluctant to cut back, so when they try to increase their cash balances they cut back on other things.
If you want to say that I assumed that the CB knows *when* people will need more money, I will agree with you. I have assumed this, and actually figuring it out can be difficult. I don’t deny this. This is what interest rate targeting, NGDP futures markets targeting and other rules are supposed to try to accomplish (no doubt they do so imperfectly).
I didn’t say that the CB knew which people wanted money (I also said permanently, not temporarily), I said that the people who knew wanted money thought the easiest way to get it was via asset sales.
There are many ways for money in the economy to get from people who don’t want it hold it to people who do want to hold it: there’s loans, asset/money exchanges, or simply waiting for people who have hold more money than they want to spend it on consumer goods and the people who hold less than they want to cut back on consumer goods (yes I would say that people changing their consumption patterns due to holding more or less money than they want is bad, but this process does avoid the compounding process where one person’s change in demand affects everyone else’s cash balance in the opposite direction which in turn affects the first agent and so forth). I took the most direct route possible because it simplified the example without changing the point.
A more major point:
My goal here is to convince you that people changing their patterns of production and consumption because they lack (or have too much) money (distinct from wealth in general or real productive capacity. You seem to agree in this very simple example, but disagree in more realistic situations. Do you disagree for more realistic situations where the change in demand is permanent? If so why; what is the difference in fact which makes the difference in policy? Later we can discuss the case where there has been a temporary change in the demand for money. I understand you want to get to the topic of discoordination/recoordination, but I think in we have to build a consensus slowly.
September 20, 2010 at 12:38 pm
Silas Barta
I don’t think it’s correct to say that “you can sell goods without knowing what good you will ultimately exchange it for” is the “unique purpose of money”.
I said major, not unique, and I think “major” is justified in light of the fact that something becomes more money-like as you can more plausibly expect to be able to trade it for something you want later (roughly, becomes more liquid).
Certainly it is a benefit of money, but I would say the “unique purpose of money” is that you can buy and sell goods without having to know what *other* people want, which applies to this model.
I take your comment here to be saying that you regard the main function of money to be that it allows you conduct a massive, multi-party, indirect exchange while only knowing your “piece” of the exchange. That is, if there’s a Pareto achievement possible by party A giving good ‘a’ to party B, who gives good ‘b’ to C … who gives ‘y’ to Z, who gives ‘z’ to A, then money allows something like this to take place without a massive coordination effort on the part of individuals A through Z. Rather, A only needs to know that he’s willing to sell ‘a’ and buy ‘z’.
And this is indeed a perk of money.
However, I think it is a major disagreement between us (and, IMHO, something that misleads your thinking on this issue) that you do not see the other purpose as being significant, while I do. The “other purpose” I’m referring to is that it allows you to satisfy a demand without knowing which of your own demands you want satisfied. In other words, in the above scenario, the existence of money makes it so that A only needs to know that he’s willing to sell ‘a’ — not even that he wants ‘z’!
If people had to meet your criterion (knowing that they want to produce ‘a’ *and* that they want to consume ‘z’), then it still wouldn’t be necessary for there to be money, as they would still have to identify this “trade cycle”, or otherwise coordinate their actions very precisely. Only by accounting for money’s facilitation of “production without foreknowledge of consumption” can you understand why it arose, what produces the premium, and what cost is paid by “injecting” money into the economy.
I think this difference permeates the rest of our disagreement. I see the recession as (something equivalent to) intermediate producers having greater uncertainty about which goods they want. This uncertainty is (efficiently, correctly) reflected in the higher premium on cash. But if you don’t see money’s role as signifcant in allowing you to conduct exchanges despite not yet having decided what you want to *receive* in exchange, then you will (incorrectly, in my view) see it as a “shortage of money” that can be corrected with more money.
But if the “shortage of money” reflects a fundamental scarcity (as I claim it does) rather than an arbitrary accounting metric, then “injecting money” makes as much sense as the government “injecting wheat” during a wheat supply contraction (and subsequent price jump): it has to get the wheat from somewhere, simply introducing an inefficiency that counterbalances the “savings” from the lower price of wheat.
My goal here is to convince you that people changing their patterns of production and consumption because they lack (or have too much) money (distinct from wealth in general or real productive capacity). You seem to agree in this very simple example, but disagree in more realistic situations.
I don’t believe I did agree. Rather, I pointed out how, under the restrictions in the example you gave, people *don’t* actually need a higher money supply, as the situation as been limited to one where the critical function of money is not necessary — which is why I showed how they can alleviate the supposed problems of too little money without a change in the money supply (a method that is borne out in practice: when people know that a large series of transactions are going to be made, then can and do reduce the number of instances when cash has to be carried around as physical tokens).
September 21, 2010 at 10:13 am
jsalvati
Yes, you read me correctly.
I think it’s important to point out that in both barter and in monetary systems, many kinds of non-money assets can serve as stores of value (this is the essence if what you’re talking about). In a monetary system, if I am trying to save without knowing what I will want to purchase in the future, I can buy some financial asset which I can sell at a later date (so my average demand for money is low). In a barter system, I can trade for a large quantity of grain (or gold or to a lesser extent housing etc.) and then trade for the things I want at a later date when I know what I want. Acquiring any store of value will allow you to postpone your consumption decision into the future.
It’s also important to note that this is basically irrelevant to the point of the example; I can rework the example to be about increased savings in the form of cash instead of an increased demand for money for transactions. After I wrote this example I realized that this second example is essentially the same as the first, so it’s probably not any more convincing to you than the other example. It should suffice point out that the increase in the demand for money could easily come from a perception that money will give them more flexibility in future purchases (of unknown type) instead of a need for money for bus transactions, without changing the example very much. However the example may be worth keeping since it says things somewhat differently:
The increased demand for money could just as easily come from an increased desire to save (either long or short term). Lets say people have a change in preferences and want to save a little bit more terms of money and a little less in terms of buying financial assets (perhaps someone requires them to have cash collateral or perhaps they have just grown more frightened of some event or something), to make things simple lets assume they are *new* financial assets (they could put their cash in a bank who lends it out, but for the sake of simplicity lets say they don’t; little in this example changes if they do). Also for simplicity of language lets say it is a one time change, they hoard a specific quantity of money instead of buying an equivalent amount of financial assets (not a stream of money financial assets) and then continue buying financial assets. Without these purchases, the producers of financial assets (companies which do do real investment projects) have less cash to undertake their real investment projects. In order to maintain a cash balance, they must reduce their investment activities (a one time reduction) or attempt to get cash from elsewhere. This has the same effects as I described in the bus example and is in itself inefficient because time preference has not changed. Since the equilibrium price level. On the other hand, if the Supplier Of Money prints money and purchases financial assets in order to offset the change in the amount of cash saved (and assuming financial markets are efficient), the producers of financial assets will not need to reduce their investment activities and the rest of the economy will not also suffer the effects of a reduction in spending or the need to adjust the price level, a clear welfare increase over the alternative.
If you do still disagree, then I think you have just not responded to my example. What are the costs that I don’t see of this increase in the money supply? You could alternatively explain why increases in the demand for money like the ones I gave in the examples will never happen, not merely that there is an incentive to reduce your demand for money (I agree), but that by some law of economics, they just cannot happen.
To respond to your specific example, lets say that preventing counter fitting requires a lot of specialized capital equipment, so basically it only makes sense to do with money.
I feel like I still do not know what the cause is behind your disagreement, what is your theoretical framework for thinking about this issue? I continue to suspect that you are using some idea along the lines of “money is information” as an intuition pump, but I don’t know for sure.
September 21, 2010 at 11:01 am
Silas Barta
Okay, I think now that you’ve adjusted the example, I can better explain why I disagree with you, and what costs I think you’re missing.
In this scenario, you’re basically saying (and correct me if I’m wrong), “When people increase their cash hoarding, while the money supply is not increased, this decreases the economy’s ability to undertake real investment. These missed real investment opportunities represent an inefficiency because a simple injection of money would have permitted them, without costing anything.”
But now look at what you’ve posited about the hoarders: they prefer to put (more of) their savings in a store of value, rather than as a financial asset. This sends a fundamentally different signal, and reflects a fundamentally different scarcity, than if they preferred to put (more of) their savings in financial assets. Specifically, the very things you were positing as the reason for why they prefer to hoard (fright about events, financial intermediaries being more skeptical of opportunities and so requesting more solid collateral) represent a scarcity, the very one I’ve been emphasizing hin this discussion: scarcity of knowledge of which processes will really satisfy consumer demand, which I call discoordination.
(Keep in mind, for a bank to be willing to make a loan, it must believe that the borrower will be able to produce something of value to sell over the term of the loan, which depends as much on the borrower’s competence as demand for what the borrower intends to sell.)
Curtailing real investment is exactly what should happen: because of the greater uncertainty about which modes of production will still be viable several years from now, fewer projects can be economically justified until entrepreneurs find
The injection of money from the central bank, then, is not costless: it props up projects that are unsustainable in light of the shifting modes of production, and will have to be (expensively) liquidated and repurposed at a much earlier time than if long-term projects had held off until entrepreneurs could identify viable ones. In essence, it “tricks” doomed businesses into thinking they can stay alive because they got “short term working capital” during a crunch. (And this is why I say that MMEs go astray when they neglect to think about whether the economy is actually satisfying consumer demands, and instead count aggregate flows.)
So, in short, that preference for a store of value rather than a less liquid, potentially-interest-yielding instrument makes a big difference. And MMEs shouldn’t keep thinking in terms of “how to keep businesses going”, but whether those businesses should be there in the first place. (I’ve gotten increasingly annoyed at the implicit assumption that everything must return to its 2005 levels.)
September 21, 2010 at 11:03 am
Silas Barta
Yikes, fifth paragraph was cut off. It should say,
“Curtailing real investment is exactly what should happen: because of the greater uncertainty about which modes of production will still be viable several years from now, fewer projects can be economically justified until entrepreneurs find out which production methods more stably work with the new preferences.”
September 22, 2010 at 8:08 am
jsalvati
I would make one major correction. The capacity of the economy to undertake all sorts of activities is reduced, not just investment (though that’s certainly one important activity). It could even be the case that people are reluctant to cut back on their investment spending for institutional reasons (say they save through pension funds), so investment gets cut back the least (though in practice I very much doubt this is the case).
The costs you describe are already reflected in my example. When people know that certain investment activities are a bad idea, the market *already* has a great way of discouraging those kinds of investments, via the fact that investment projects related to those fields will be less profitable. If there is more uncertainty about which projects will be profitable in the future, people will pay less for financial assets linked to those projects (which could be all projects). Investors already bear all the costs of poorly planned investment projects. Why is an additional incentive to halt activity necessary; what is the externality, and how did the solution get in to money?
You ignore continue to ignore all the really weird things about your position:
1) Lets say money demand goes up because Scrooge McDuck has taken up money swimming. Or say someone decides to save in the form of money because they are financially illiterate and don’t know how to save in the form of stocks or bonds. Or say there’s a wave of immigrants who come from a barter system they are ready to both consume and produce, the economy will simply have more transactions and therefore people will try to hold more money to conduct them (this is true even if everyone knows exactly what they want and how to produce it). These all have the same results, but I don’t think you can claim that they have the same cause of “uncertainty”. None of the arguments you have made even attempt to make sense of this.
2) The actual activities which get cut back don’t necessarily correspond to the activities which are related to any increased uncertainty. Someone could hoard money and cut back on oatmeal because they are worried about the future in general. Oatmeal production is affected regardless of whether the uncertainty has anything to do with oatmeal, it’s just the easiest margin for them to cut back on.
3) Activities which cannot be transferred into the future are also cut back on. In other words consumption today is sacrificed for no gain in the future.
4) With permanent changes in the demand for money the overall price level must change, this cries out for explanation.
I have mentioned these before, and you have yet to respond to them.
September 22, 2010 at 10:43 am
Silas Barta
I have mentioned these before, and you have yet to respond to them.
Do I sense a bit of hostility there? Please have patience: there is a huge inferential and intuitional gap between us, and I believe I could point out numerous instances where you haven’t answered problems that I have posed (though perhaps this is to your credit, since you are at least willing to admit when you don’t have an answer to something).
First of all, regarding your 1-4), I believe I have already agreed that there are circumstances where increasing the money supply is an acceptable, efficient move. For example, the previously discussed case where country A converts to using country B’s currency by having A’s notes replaced with B’s at a pre-defined rate (though as you’ll remember, at the time I questioned the relationship between this kind of conversion and that conducted by standard monetary policy, which puts the money first in the hands of preferred players).
For some other examples: In a gold-based monetary system, obviously mining new gold increases the money supply, and I have no probelm with that (and before you ask, it’s because more gold means more real resources for the economy). Or, for any economy, if a good (other than the national currency and its derivatives) becomes so liquid that people take it as payment, that is an effective, desirable increase in the money supply (again because it corresponds to a real resource increase, though in the form of increased certainty about the stability of economic relationships i.e. coordination).
However, the cases you bring up in 1-4) are beyond the scope of the topic we were discussing, which was whether monetary injections can be helpful as a way to combat a recession, predicated on the belief that the fundamental problem is monetary disequilibrium.
The capacity of the economy to undertake all sorts of activities is reduced, not just investment
Yes, so consumption would have to be curtailed in the short term — that’s good. When the economy has suffered a productivity shock (which in a recession is in the form of greater discoordination), it is less capable of satisfying consumer demand, and people should be cutting back on consumption. Monetary policy can shift this around and try to hide it, but it can’t eliminate the decline in real productive capacity.
The costs you describe are already reflected in my example. When people know that certain investment activities are a bad idea, the market *already* has a great way of discouraging those kinds of investments …
No, people don’t know that certain investment activities are a bad idea. They estimate that certain activities are a good or bad idea based on their expectations, and current resource costs. If resource costs are artificially manipulated, they will undertake what they believe to be a good investment, but (more than usual) will not turn out to be a good investment.
And in an economy with rapidly-shifting modes of production (a term I keep using, and a concept which is pretty much absent — and disconnected — from MME models[1]), fewer of the economic “ecosystems” will look like they do now in the future, meaning investors and businesses need to make more conservative assumptions about what will still be valuable — but this information is not communicated to them if the “store of value” premium on money is squelched!
[1] I chose my words carefully in this parenthetical: it’s fine if a specific economic model EM leaves out the concept of heterogenous, shifting modes of production. However, it should at least be used with an understanding that can *connect* this model to other models of different scenarios, and thus allows it to identify when such a concept renders EM inapplicable. In other words, it must be used with a Level 2 understanding (by my hiearchy that I can link again), and it is this understanding that I claim is lacking among MMEs, and which is revealed in Sumner’s inability to explain what should be done with NGDP when unusual situations (like more barter and saving) cause it to fall).
September 22, 2010 at 10:46 am
jsalvati
Haha, I don’t mean to be hostile. I can get frustrated at times, but I work hard to remind myself that this usually stems from mutual incomprehension/miscommunication.
September 22, 2010 at 1:21 pm
jsalvati
I wanted to focus on these limited changes in the demand for money because I think they highlight our disagreement. I want to avoid talking about productivity shocks and any related changes in the demand for money and stick to talking about changes in the demand for money with other causes until one of the following happens: you agree that changing the supply of money with the demand for money in the second case is a good idea, I am convinced me they don’t occur, or I am convinced they are a bad idea (and why). I think agreement on this topic will greatly facilitate agreement on productivity shock related changes in the demand for money.
Do you agree or disagree that in the bus example or in a Scrooge-Money-Swimming case and other similar examples, a counterbalancing decrease/increase in the money supply would be a good thing, and that these kinds of shocks can happen? If you want to claim that these kinds of shocks are infrequent, that’s fine.
It may be useful to note that I think in a commodity currency regime, changes in the money supply due to new reserves of gold being found etc. are not necessarily good, though they may tend to come about when there is an increase in the demand for money (people look harder for gold) so they might be biased in the correct direction.
September 22, 2010 at 1:31 pm
jsalvati
Sorry for being dense about this, but I was wrong about this before, and it made for quite a bit of confusion I think, so I want to be super sure we are on the same page on this topic.
I should say that it is of course also fine if you claim that while it might be good to counter balance these changes in the demand for money, there are productivity shock related changes in the demand for money which are different and should not be counter balanced. In fact, I assume you do claim this.
September 22, 2010 at 1:57 pm
Silas Barta
Wow, you almost read my mind with the last comment. I was going to say that I didn’t see why you want to focus on those narrow cases. Even if I agree that increasing the money supply is desirable in those cases [1], it’s orthogonal to whether doing so is the appropriate response to money hoarding that results from the economic uncertainty inherent in a discoordination scenario.
Of course, they’re not completely independent because if I agree that increasing the money supply for one reason is okay, while doing it to counterbalance uncertainty-based hoarding is not, then I need a theory for how the central bank is supposed to recognize the difference between the two.
At the same time, if you’re trying to justify anti-recession monetary stimulus by pointing to all of these pathological cases where the money supply should increase, that sounds like a “not your real rejection” failure mode. (Please let me know if I’ve using unfamiliar LessWrong jargon.)
Since I’ve spent so much time accusing you of biases, perhaps I should elaborate more on _my_ “real rejection”, so you can better understand my intuitions: I believe that over the past 10+ years, the economy has become increasingly decoupled from the “fundamentals”, and economists have not quite appreciated this. That is, they have operated on this assumption that the big banks have to be saved “because we need banks”, despite the fact that (like I said about spending) banking is not good — rather, good banking is good. And banking, like everything else in the economy, should be judged by whether it’s really adding value, not whether it’s copying its traditional method of operation.
And similarly, I’ve been bothered by this insistence that consumer credit card borrowing, home ownership rates, savings rates, etc. MUST go back to their circa 2005 levels, even though there’s no logical reason for that level to be some optimum.
So I think the economy is due for some fundamental readjustment, as HUGE sectors of it no longer provide the value they used to (higher education, traditional banking/finance, law, newspapers), yet people are slow to recognize this. Furthermore, I see these fiscal/monetary stimulus/stabilization policies as ultimately just propping them up and avoiding necessary change, making me skeptical of the merit of injecting this money, and skeptical of explanations for the problem that suggest it’s just a matter of fixing some nominal variables, when there are fundamental changes that need to happen.
I hope that makes clear why I see MMEs as having a black-box view of the economy that lacks and understanding of why businesses are failing or succeeding.
[1] And I almost want to agree, but I’m not quite there yet, either. If e.g. I store my gold in a warehouse and get a receipt that entitles me to that quantity of gold, but then forget I have it, should the warehouse owner print up an extra receipt for that amount and buy stuff for himself with it? My reaction is “not cool”.
September 22, 2010 at 3:06 pm
jsalvati
I feel like we are getting somewhere. We disagree about whether agreement on these two topics should be orthogonal; I think the two cases are fundamentally the same, and you obviously disagree; that’s what the next part of our discussion will be about. Unless you tell me otherwise, I am going to assume you agree with me about these two topics, that adjustments in the money supply for the types of cases I brought up are desirable in principle. I won’t count disagreement about the desirability in practice as a disagreement relevant to this discussion. You could think that such monetary policy is handled so poorly that it is undesirable or that it is too difficult to tell the two cases apart in practice to be worth it, without making me complain.
I think I understand what you’re getting at in your footnote, but I am not sure. When the demand for money falls (i.e. Scrooge McDuck decides he sucks at money swimming and gets rid of his money pool by buying financial assets with the money), the Supplier of Money must undo the process of money creation by selling financial assets and destroying the money they receive in the process, reducing the money supply. While McDuck’s money pool exists, the Supplier of Money keeps that created money in the economy. Tell me if this was not your issue or if you disagree with my response, and we can continue this part of the discussion. Does this answer your question/skepticism?
My goal in reaching agreement on this issue is to find a little edge of agreement that we can use to build on. I am semi-comfortable with LW jargon. I don’t remember this particular piece well, but I think understand its meaning (disagreement about some topic which is not at the heart of the issue and not the real cause of disagreement). Fundamentally, I don’t think the cases I brought up and the cases you brought up are different in a relevant sense. I will seek to understand why you think they are different, and seek to explain why I do not think they are different, and then resolve this disagreement using our little region of agreement as a starting point.
Assuming we agree on the cases I brought up (perhaps these need a name so they are easy to talk about?), my first request is: explain the case that you see as different as fully as possible. It should answer questions such as: When there is an important productivity shock (or potential future productivity shock or preference change), why and how does the demand for money go up? Is it temporary or permanent? Who are the people who increase their demand for money and what is the incentive that causes them to do so?
As a side note, I am extremely skeptical of such “big imbalance” theories. My general reason is that it is difficult to trick people into doing things which are very stupid for themselves especially over long periods; you can manipulate them into doing things which are bad in general, but you generally have to compensate them so that it is not bad for themselves. I won’t reject the possibility that my view stems from a personality trait, since theories of this type are very popular.
September 22, 2010 at 3:36 pm
jsalvati
Another thing to note: I have realized recently that, while I don’t think mainstream macroeconomists disagree with Monetary Equilibrium theory, it is Austrian economists who emphasize its use far more. I also think there are many mainstream (non macro) economists who really just don’t have a clue about macroeconomics; there are probably some macroeconomists for which this is also true and spew nonsense because they don’t understand the foundations of macro.
September 23, 2010 at 11:01 am
Silas Barta
There is a lot to discuss here, and I’m going to need some more time than usual to write up a full reply, but I’ll try to leave you with my initial impressions.
1) Do you really see no fundamental difference between:
a) Change in demand for currency C because of a fundamental discoordination between factors of production and what people actually want (e.g. most factories are making buggy whips, while no manufacturing process uses buggy whips as an input anymore)
vs.
b) Change in demand for currency C because a nation is collectively, coherently transitioning out of C?
2) How the monetary injection policies you advocate differ from counterfeiting?
3) I differ from Austrians in that, when it comes to effects of money supply increase, I’m more concerned with the “store of value” role of money, than on the manipulation of the interest rates
4) I’m not claiming people are doing stupid things for themselves — these things that persist because of accomodative monetary policy are certainly individually rational, but collectively poison (like a prisoner’s dilemma situation), and I want to a transitition out of that. The university education system, for example, can’t persist because it’s been draining more and more resources for less and less benefit — though individuals still benefit, barely, from going to college.
(Then again, maybe that’s a sufficient reply to establish where and why our views differ?)
Oh, and:
5) If the past is any guide, you’re a little optimistic about how much progress we’re making 😉
September 23, 2010 at 11:58 am
jsalvati
1) Yes, and you’ll have to convince me of this distinction. Part of it is that I don’t really agree that an increased demand for cash per se is a necessary response to an increase in uncertainty. I am sure we’ll discuss this when you post your in depth reply.
2) In some sense they don’t much. You could think of counterfeiter as equivalent to poor monetary policy: counterfeiting is not done at the correct times/directions and done less efficiently than the Supplier of Money. If the Supplier of Money responds correctly (counterbalancing the counterfeiting), they lose money, which may or may not have social costs. If the Supplier Of Money does not respond correctly, then counterfeiting has the obvious costs of monetary disequilibrium.
3) OK. I don’t think focusing on interest rates is a good idea either. On a related note, a lot of our disagreement comes from the fact that you focus on money as a store of value, and I think this focus is misplaced.
4) OK. I’d probably disagree, but I think that’s a discussion for another day.
5) Haha, that’s unfortunately probably true.
September 23, 2010 at 1:01 pm
jsalvati
I think it might be useful as an intuition pump for how I think of things to explain how I would imagine a stable private money system. I don’t claim that it would be easy to get into a stable private money system; I think it would be pretty difficult indeed.
Consider a private issuer of money, they are well trusted for one reason or another. They issue bank notes by exchanging financial assets for their bank notes at the market rate (in terms of their bank notes) in the market. People hold these bank notes because other people accept them for payment in transactions, which lowers their transaction costs and because they can trade their notes to the issuer for financial assets at a “fair price” (and they trust the issuer to be honest about this). These notes pay a certain amount of interest to their holders (interest in terms of the bank notes, but not from new money, but from the interest they collect on the assets they hold). The issuer makes a profit from the difference in interest rates on the financial assets and the interest they pay on the bank notes (could conceivably be negative). It is natural for the issuer to adjust the quantity of bank notes to maximize the benefits of holding their notes. Of course my contention is that adjusting the quantity of notes to the demand for the notes will maximize these benefits, and naturally you could have a different view).
September 24, 2010 at 8:00 am
jsalvati
In hopes of convincing you to think of the examples of a change in the demand for money that I gave earlier up as non-pathological, I have come up with couple more.
Consider the case of a competitive currency regime with several private issuers of (different) currencies. There are also differences in the characteristics of the currencies. Perhaps some of them are all electronic, some of them look really pretty, some of them have anti theft devices measures, some have better monetary policy ect. . Some people and businesses use just one currency, some use multiple currencies. As people shift their relative usage of currencies (as will inevitably happen) they will want to exchange one kind of currency for another. However, because the different currencies have different properties, if you switch from using one currency to another you will not want to hold equivalent amounts, so the change in demand is uncertain. This means a pure exchange of one kind of currency for another will not suffice if you want to maintain monetary equilibrium.
You can also get a similar situation with a single supplier of money, if there is a change in the quality of their money (perhaps they go to all electronic currency), or if they change the rate of interest which they pay on their currency (perhaps the cost of currency paper increases so the cost of making money increases, so they have to recover their costs). People will change the quantity they want to hold.
I hope these seem like things which could actually happen, and thus situations which any monetary theory should know how to talk about.
September 27, 2010 at 7:46 am
Silas Barta
Please pardon the delay in response.
At this point, I think these are the major points of disagreement between us:
1) The role of money as a store of value. I think this adds a significant, time-varying premium to money’s value, and this premium reflects a fundamental scarcity that cannot be removed through changes in nominal variables. The relevant scarcity is the “level of coordination”, i.e. how well the factors of production match up to satisfying consumer demand. As it becomes less certain that you will be able to supply a valued input into the production chain, you be willing to “pay more” for money.
You disagree, as far as I can tell, on the grounds that discoordination need not be reflected in the “store of value” premium on money. My challenge to you, to resolve the disagreement, is this: how would a central bank distinguish between a) an inefficient curtailing of spending to increase cash balances, vs. b) people not buying as much because they don’t want the crap businesses are currently selling (and don’t want to invest because they believe current plans are stupid or too risky)? That is, what stops the Fed from unknowingly propping up failed businesses by mistakenly identifying them as being illiquid instead of insolvent?
Also, I submit that you cannot account for the use of money purely by its facilitation of indirect exchange, as this desire (for indirect exchange) would lead to seeking “trade cycles” (A gives ‘a’ to B … who gives ‘y’ to Z who gives ‘z’ to A), not to time-delayed indirect exchange and savings involving money. Only by factoring in the “store of value” premium into models of money usage can you have the “level 2” understanding of money that allows you to account for why people will shift into and out of barter, and avoid “holes” in your model that have you advocating increased NGDP (even from bans on self-barter i.e. homecooking), despite the economy’s being … economical.
2) Counterfeiting: I say that the reason counterfeiting is harmful is that it, in effect, extracts value from the rest of the economy without supplying equivalent value in return (i.e. the counterfeiter gets valuable production in exchange for performing the worthless service of printing bills). You say that counterfeiting is harmful because might be done at the wrong times. (I had previously thought you would draw a distinction based on what is bought with the new money, referencing the Real Bills Doctrine, which was discussed recently on mises.org, but I was wrong.)
Your question about private banking is related to this. You say that under a private banking system with different currency issuers, there will be permanent monetary disequilibrium, if I understand you correctly. I don’t understand why that’s a problem. If demand for different currencies fluctuates, it’s because those currencies have different properties, just as demand for cheese and inputs to making cheese fluctuate in response to prevailing conditions. In a sense, they will never be in equilibrium because preferences and conditions will always be shifting. If people want money as a store of value, they will prefer the issuers that don’t dilute the currency; therefore, competitive pressures would not lead private banks to all gravitate toward the kind of monetary equilibrium maintenance that you want out of the Fed.
September 27, 2010 at 9:22 am
jsalvati
1) You still have not explained this. I will repeat my request:
“[E]xplain the case that you see as different as fully as possible. It should answer questions such as: When there is an important productivity shock (or potential future productivity shock or preference change), why and how does the demand for money go up? Is it temporary or permanent? Who are the people who increase their demand for money and what is the incentive that causes them to do so?”
1.1) Here’s my answer; I’ve tried to express this before, but I haven’t known how to do a good job (not sure this is either): You cannot store value without investing or trading with someone for a promise to give you something valuable later, and they can either consume or do the same thing you did. If there is to be net storing of value, someone must be investing (in the real sense). Those people who think that current stuff is crap and real ways of transferring value into the future are crap shouldn’t have a 3rd alternative. If ways of transferring value into the future are crap, people must live with it, their future purchasing power should also be crap. They should have to make the same decision between current crappy consumption options and crappy real investment options they would have to in a non monetary economy; allowing people to think they can do otherwise will cause problems.
Money is not magic; it does not allow anyone to do anything they could not in principle do in a non-monetary economy. Money does not magically allow you to transfer resources into the future, in order to do so, you must find a real way to do the transfer, and if you can’t you must live with that reality.
1.2) Yes, you can. If trade the cycles involve financial instruments then you have trade across time.
2) OK. I will only point out that ‘it happens at the wrong times’ is not a full characterization of my views.
4) I didn’t ask a question about a competitive currency regime; I made some statements about competitive currency regimes because I thought it was interesting and because I thought it might help you understand my framework. A competitive currency regime will not be in permanent major monetary disequilibrium (unless people unexpectedly and cyclically want to change their money holdings in different securities and the issuers are unwilling to accommodate them). In equilibrium, people will want to hold a currency which has the biggest benefits for them; we disagree about what policy regime will produce the biggest benefits.
September 27, 2010 at 10:32 am
Silas Barta
1) I’ll take the first case then, though it’s the opposite: the case where Scrooge stops using cash for swimming and starts spending it (demand for money goes down). In this case, the economy has shifted to regard a set of production methods as being relatively better at satisfying wants than they did before (factories serving the non-money-swimming crowd are more valuable). Because the change in preferences is permanent, so is the change in demand for money. Scrooge is the one who changes his demand for money, and the incentive is his shifting preference schedule.
I still don’t see what is gained by having a money issuer suck out that extra money — those price changes reflect a real change in preferences.
1.1) Okay, well that’s a pretty big difference between our views then. You’re saying that if I throw some gold into my safe, wait a few years, and then trade it for newly produced goods, that’s somehow impossible? I didn’t get a promise for anything later in return, I didn’t use the gold to invest (since all the projects looked like crap), I didn’t buy goods I regarded as crap, and yet I was able to buy non-crap goods with the gold later. That’s impossible?
If you think that’s impossible, then you’re saying that gains in productivity are in general impossible — that, e.g., it’s impossible even for a single factory to stop producing crap, and start producing something of value. And that I can’t hold off on buying until that factory has started producing something else. That’s pretty clearly false.
It would seem to further imply that the Fed needn’t distinguish liquidity problems from solvency problems among businesses, because there is no difference? (Or at least, the transition to different, solvent businesses involves making loans to insolvent, illiquid businesses in the interim?)
These are pretty shocking inferences are your part.
1.2) That would explain a time-delayed exchange, but not saving in terms of money, in which one person produces *without* knowing what real good he will ultimately trade.
2) I’m sorry for not giving a fuller characterization, but what would you say is harmful about private counterfeiters, *other* than the fact that they sometimes counterfeit at the wrong times?
4) A competitive currency regime will not be in permanent major monetary disequilibrium
So the international, “competing” currency regime that we’re currently in is not in a permanent disequilibrium?
September 27, 2010 at 10:33 am
jsalvati
You want to keep the quantity of saving through money constant, so if one person saves more through money, another must save less through money, but why is this desirable?
If people want to save through money, why should the Supplier Of Money (SOM) not accommodate them? They collect financial instruments, and hand out money in return. The gains or losses the SOM makes off its assets minus the float is passed through in the form of positive or negative interest on money.
September 27, 2010 at 10:34 am
Silas Barta
Sorry, 1.2) should be “… what real good he will ultimately trade it for”.
September 27, 2010 at 11:22 am
Silas Barta
You want to keep the quantity of saving through money constant,
No, I don’t, and what is this in response to?
September 27, 2010 at 11:52 am
jsalvati
I think you do, but I think maybe it’s best to ignore this for now because our discussion is growing disorganized. It was mostly an elaboration on my 1.1) point, I apologize for doing this in a confusing way.
September 27, 2010 at 11:30 am
jsalvati
I am going to focus on 1.1, I don’t think I am saying anything controversial. I don’t think it will be productive to discuss a lot of other issues without coming to agreement on this.
I am going to assume that you mean gold as a commodity, we can discuss gold as a currency and commodity later if you like.
You claim “I didn’t use the gold to invest”, but you did; you have stored gold, and how much value you get in the future will depend on how valuable gold is in the future. Gold today is a fundamentally different thing than gold tomorrow. Perhaps you would prefer it if I said you couldn’t save without getting a promise or doing “real saving” meaning undertaking some real activity to transfer value into the future. Do you agree with this claim?
September 27, 2010 at 11:54 am
Silas Barta
Okay, if you’re going to count storing gold as a kind of “real investment”, that’s different from some other people I’ve argued with, but now that I’m clear on that definition, we can proceed.
Gold today is a fundamentally different thing than gold tomorrow. Perhaps you would prefer it if I said you couldn’t save without getting a promise or doing “real saving” meaning undertaking some real activity to transfer value into the future. Do you agree with this claim?
No, the claim still doesn’t work unless you stretch the definition of “undertaking some real activity to transfer value into the future” to include “not buying and consuming present output”. But at that point, you’re agreeing that saving money (qua money) can count as real saving, which defeats the purpose of the distinction!
How about another example, in reference to this:
Those people who think that current stuff is crap and real ways of transferring value into the future are crap shouldn’t have a 3rd alternative. If ways of transferring value into the future are crap, people must live with it, their future purchasing power should also be crap.
Say someone tries to sell me a birdhouse for $X. I think it’s ugly. The next day he comes to me, having made some modifications. I decide I like it, so I buy it for $X. Have I not taken the “third alternative” of “regarding current products as crap, current investment instruments as crap”, and yet enjoyed higher future purchasing power?
Is this something that you think shouldn’t be possible, or …?
September 28, 2010 at 9:04 pm
jsalvati
The real saving is not the “refraining from buying” it is accumulation of inventory. Accumulation of inventory is also investment (even government statistics count it as such). It may be effective investment or really poor investment, depending on the good. Some goods store pretty well (bird houses can be kept around for a while); some store very poorly (manual labor doesn’t store well at all). You are right to point out that accumulation of money in this way does cause a reduction in the usage of resources, but it doesn’t necessarily cause this in a way that makes much logical sense.
You seem to have it stuck in your mind that money is “about” storing value, and I think this is leading you astray. The real “about” of money is facilitating exchange; making trade easier. The big difference between a barter economy and a monetary economy is that exchange is much easier in a monetary economy. The ability to store value is a necessary feature for facilitating exchange, but it doesn’t even have to be especially good at it in order to facilitate exchange (though it’s certainly better if it does do a good job). Even when the monetary authority does a really poor job, and people are very reluctant to use money as a store of value people still use it as the medium of exchange. Money is useful for many things, but when is it *really* important to have? When you need to do exchange. There are many common substitutes for money as a store of value, but very few non-money substitutes for money as a medium of exchange.
I didn’t realize you had attempted to answer my request:
I’ll take the first case then, though it’s the opposite: the case where Scrooge stops using cash for swimming and starts spending it (demand for money goes down). In this case, the economy has shifted to regard a set of production methods as being relatively better at satisfying wants than they did before (factories serving the non-money-swimming crowd are more valuable). Because the change in preferences is permanent, so is the change in demand for money. Scrooge is the one who changes his demand for money, and the incentive is his shifting preference schedule. There are many kinds of stores of value other than money that we don’t call money, but if an asset starts to be used to facilitate exchange we begin to want to call it money.
What you have described is wrong for two reasons. To make this clearer, lets consider the time when Scrooge first decided to save. I am going to assume that you described Scrooge’s perception that future consumption would be more valuable to him than current consumption (the other possible interpretation does not make sense since only relative differences in expected utility affect one’s actions). The first error is that Scrooge’s perception will lead to an increase in the demand for *savings* (any kind of store of value), and not necessarily an increase in the demand for *money*. Such savings could be in the form of money, but need not be. The second error is that this kind of savings is an attempt to reduce consumption in one state of the world (present) in order to increase consumption in a different state of the world (future). Since the states of the world we are discussing are time-states, these are temporary changes. There might be some reason why Scrooge’s demand for money might go up permanently (I have mentioned one), but what you have described is not one of them.
There’s one other critical feature of your story which you haven’t done a good job of explaining: why is an additional incentive necessary to get correct investment and consumption decisions beyond merely bearing the costs of those decisions and why do correct incentives require monetary disequilibrium? Consider the following example:
Lets say aliens come down and secretly change a large fraction of the population’s preferences; they now value different things than they have in the past. Some currently produced goods satisfy these new preferences better than others. Future produced goods will satisfy preferences better than currently produced goods; the people who’s preferences have been changed (call them Alties) know this but don’t know what products will satisfy their new preferences, just that since their preferences have just changed someone will probably invent some products that conform to their preferences better quickly. The people who are unaltered (call them Unalts) do not know this to start out with. Thus, Alties expect that projects that people come up with in the medium future will be more profitable than in the very near future.
Inputs vary in how amenable they are to storing; some goods can be stored very easily (gold) and some that cannot be stored very easily (labor) and some in the middle (potatoes). Goods that can be stored easily have higher option value than goods that cannot be stored easily.
It seems to me that people have the correct incentives to choose between consumption of goods produced now, starting investment projects (both actual projects and storage investment) now and starting investment projects later without the additional hardship of holding more or less money than they wish to hold. People bear the costs and benefits of their consumption and investment decisions, so goods will be stored until their marginal option value equals their marginal current value, and investment projects which make sense will be undertaken and those which do not will not.
You have mentioned that you think people do not understand that some projects will be unprofitable. Alties know that their preferences have shifted; do they not have the correct incentives to shift their current consumption to products which are relatively more attractive, to short financial instruments which are based projects they know will do poorly and go long on financial instruments which have greater option value (storing easily storable inputs that may be useful for future more profitable production) ect.? Why are these normal incentives not enough, how do the incentives caused by holding less than the preferred quantity of money improve on these incentives, and why? This process seems like it would work equally well in the case where the demand for money did not increase and in the case where the demand for money increased but the central bank accommodated the increase with an increase in the supply of money.
September 29, 2010 at 9:10 am
Silas Barta
Long reply below. Please let me know if you think our disagreement reduces to a subset of the issues I discuss below so as to save us some time if possible.
I’ll begin a new numbering in a new namespace.
1A) First, I’m going to sort of mimic you and point out that you didn’t answer my critical (well, only) point: you seem to ground your position in the belief that, if people think current products are crap and current investment opportunities are crap, then it isn’t (shouldn’t be?) possible for someone to save their cash and have a higher purchasing power later. Yet this would imply that producitivty gains are impossible. And I gave an example of someone doing exactly that — holding onto money and getting higher purchasing power because of an efficient shift in production methods.
So what dynamic have I missed that made this example possible? Or why do you think that kind of thing (saving until birdhouses aren’t crap) *should* be impossible?
2A) You seem to have it stuck in your mind that money is “about” storing value, and I think this is leading you astray. The real “about” of money is facilitating exchange; making trade easier. … Money is useful for many things, but when is it *really* important to have? When you need to do exchange. There are many common substitutes for money as a store of value, but very few non-money substitutes for money as a medium of exchange.
But if money *lacks* the function of “store of value”, then it *isn’t* valuable as money. To see this, imagine that every time your produced something and sold it to someone else, you had to know *exactly* what you would spend the money on. Would you be willing to accept such an asset as payment? It’s probably better than barter (in which *that specific person* must have the good you want), but eliminates most of the reason people are so willing to accept money as payment.
If you did accept such a money as payment, the best thing you could do is decide that good you’ll trade it for is something … um, extremely liquid and which holds its value: in other words, something more money-like by the very fact that it stores its value! So you see that this role, as a store of value, is deeply intertwined with money’s role as an exchange facilitator. Without that function, that money is little better than a complication to bartering.
And (to bring this back to my challenge about the level of understanding of economists), if you dismiss this aspect of money, you don’t understand why there would be a transition out of barter toward the use of any money-like instrument, and you will fail to appreciate under which circumstances people will switch out of and into barter. And like Sumner, your obsession with the NGDP metric will blind you to this dynamic.
3A) I think we’re talking past each other on the Scrooge example. I thought what was going on was this: Scrooge saved that money because he likes swimming in money (specifically in money), and that this is only satisfying to him because the money is valuable; and monetary policy did *not* accomodate him. In that case, Scrooge is the increased demand for money, so yes, money demand goes up. When he gives up money-swimming and spends the money, prices for most goods go up — as they should, since Scrooge has chosen now to consume this output, involving a massive change in consumer demand (though due to one person). So what’s the problem? This is exactly how the economy should react to a contraction and expansion of consumption.
4A) There’s one other critical feature of your story which you haven’t done a good job of explaining: why is an additional incentive necessary to get correct investment and consumption decisions beyond merely bearing the costs of those decisions and why do correct incentives require monetary disequilibrium? [Alien example]
I thought I answered this by comparison to the wheat examples. Short answer: efficient allocation of resources requires that people receive signals informing them of all the costs. If this doesn’t happen, then the profit/loss test ceases to be indicative of allocative efficiency (and probably of productive efficiency too). For example, if the government promises to purchase *all* unsold output (after a certain period), then everyone will show profits — but it would *still* be ridiculously inefficient, because it’s not incorporating all preference schedules into the production structures — a sort of fake, Potempkin Village economy, if you will.
Long answer, involving wheat analogy: What if there’s a shortage of wheat because of a failed crop [1]? Should the government (or some governmental “central wheat bank”) issue more wheat to alleviate the extreme inconvenience of wheat trading at a premium? No, because the government must get that wheat from somewhere — and in doing so, introduces more inefficiency somewhere else, even if it can protect people’s inalienable right to low wheat prices.
You might say that the analogy breaks down in that government creation of money is nearly costless, while its creation of wheat is not. I believe I have refuted this: The higher money prices [2] reflect (per 2A above) greater uncertainty about the future ability to have sustainable production modes that satisfy consumer demand. This is a real cost, just as surely as the destroyed crop is a real cost. And just as government can only shift around the cost of the failed wheat crop (e.g. through lower production/consumption in other areas), it too can only shift around the cost of discoordination.
When the UnAlt products stop selling, the central bank sees people being morons who aren’t keeping NGDP high enough and gladly issues them “short-term working” capital (effectively, free money) so that their business appears profitable even despite this “temporary” slump. As this fails to prop up NGDP, the CB resorts to QE. The Alties soon realize that it’s better to buy these unwanted products because they still give some utility, which is better than losing the value of their money to inflation.
So this new issuance of money did come with a cost: it caused horrible inefficiencies by keeping the economy from adapting to a new preference schedule, even as the conventional measures show that the CB “succeeded” at keeping up aggregate demand, and NGDP at trend. But this has only hid the cost of the new money issuance, not eliminated it. Which is why I disagree with you that the process of adjustment has not worked equally well.
And this is exactly what I claim that MME/Sumner policies will do to the economy, and exactly the dangers I warn about when you promote NGDP to an end, rather than making sure your policies connect to “satisfying human wants”, which is what an economy should do.
[1] In the loose sense of “shortage” — high prices — rather than the technical sense of more willing buyers at the current price than sellers.
[2] Yes, this really means “lower price level”.
September 29, 2010 at 9:17 am
Silas Barta
Sorta-Errata:
On 2A), for clarity, the sentence should read, “…the best thing you could do is decide that the good you’ll trade it for is something”
On 3A), I forgot to mention that in my previous reply I had accidentally reversed my claim about what happens when Scrooge starts spending the money from his momey-swimming pool.
September 29, 2010 at 11:21 am
jsalvati
1A) My point is not that saving cash cannot transfer value into the future for any individual. My point is that accumulating money by itself does not guarantee that value is transferred into the future. To transfer value into the future, someone must effectively agree to a trade a cross time with you or someone must actually arrange for goods to be transported into the future (accumulating inventory is certainly one way to try to do this, but for many goods it’s not an effective way).
2A) Yes, I agree that it is necessary for money to be a store of value, and I said as such the first time around. However, money doesn’t necessarily have to be a *good* store of value for it to function pretty well to be very useful, and there are lots of other goods that also have this feature and substitute for money in this regard. Let me try to put this another way: the features that distinguish money from other goods are its roles as a medium of exchange and as a unit of account; if it has a special role in the economy it is because of these two features. This does not deny that being a store of value is necessary to have those features or that this feature influences the workings of the economy.
2B) I wish you would stop accusing me of having an inadequate level understanding. I understand you think I do, but you must understand that this is an extremely unpersuasive argument by itself because that the reverse is also true: you appear to me to have an inadequate level of understanding. This should be obvious. I think I can even claim a bit of an advantage in this respect: I have actually managed to explain monetary disequilibrium to you well enough to convince you it is an important phenomena in at least some situations, but you have not managed to even explain your other phenomena so that I believe it (to be clear: Recalculation certainly exists in the sense that people must sometimes struggle to find new methods of production and trade, but this is not something you introduced me to or changed my mind about, and the existence of Recalculation is not fundamentally what the argument is about).
3A) Perhaps we should pick a different but similar example; I think perhaps Scrooge’s desire to swim in money as a form of consumption is causing us to talk about different things. I think the original Scrooge example that I gave is basically a simple version of the bus example (increased money demand version) and basically purely supports the importance of monetary (dis)equilibrium. Think about the following slightly different alternative to the Scrooge example: instead of using actual money for swimming, he contracts with the kid down the block to print up an equal quantity of bills that look basically indistinguishable from cash (but others won’t accept them; think of them as financial instruments) and pays the kid the same amount of money (he likes the feeling that the bills have real value) with the agreement that the kid will reverse the trade at any time if Scrooge wants to. The kid invests the money in government bonds. These two situations are pretty similar for Scrooge, but quite different for the economy as a whole since in the first case there will be some amount of monetary disequilibrium and in the second case there will not be.
4A) Perhaps I haven’t been specific enough about what kind of explanation I expect to consider satisfying. An explanation I would consider satisfying would start from agents, their resources, incentives, capabilities, information and expectations; in other words a microeconomic explanation. I know I let you know that I saw the wheat analogies as wildly non-analagous. Let me give you a different analogy: if people suddenly want to hold more bonds is it bad for companies to businesses supply more bonds (one could call this “printing bonds”)? I doubt you will say it is bad. I see these situations as far more analagous than the wheat example (meaning at least slightly analagous). Analogies are not going to convince me (and I doubt they would convince you); you’re going to have to actually explain the cost you’re talking about (lets call it the Silas Cost) at the microeconomic level.
4B) I am a little confused why you like to conflate concerns about monetary disequilibrium with NGDP targeting. I try not to misrepresent your views, and you should extend me the same courtesy; it’s hard to have a useful debate otherwise. I have told you before that I have no doubt NGDP targeting has some flaws. I only claim that I think NGDP targeting would do a better job of avoiding major monetary disequilibrium than other rules like inflation targeting. After we come to agreement on the importance of monetary equilibrium, we can discuss NGDP targeting if you like. I am far more likely to be convinced that other monetary policy rules would work better than NGDP targeting than that monetary disequilibrium is not important.
September 29, 2010 at 3:43 pm
Silas Barta
Longer response coming, but I first wanted to apologize for accusing you of lacking understanding. There is a big inferential gap between us (closing by the day) which for now looks like poor understanding on each other’s part, and so I should avoid insinuating that the problem is on your end. My main complaint is with certain MMEs, whom you have already shown significant difference from, and I only really intended to highlight your previous uncertainty about where barter interfaces with NGDP, which you had agreed to, and which you’ve since adjusted for.
Again, I’m sorry for throwing that accusation around, and I will focus on closing our gap, and reserve further comments about understanding (on anyone’s part) for when we reach a conclusion.
September 29, 2010 at 6:39 pm
jsalvati
Glad to hear it!
October 3, 2010 at 7:47 am
Silas Barta
Sorry, I know I’d been planning to have a reply yesterday or the day before, but better late than never, right?
1A) Previously, you had said that it’s not possible for someone to regard current consumption goods *and* investment opportunities as crap, and yet enjoy increased purchasing power, or that, for various reasons, this should not be possible. I showed that this is just a generalization of the phenomenon of increased productivity — of producers adapting to meet consumers. Therefore, it does not require any explicit agreement, or even that it be “guaranteed” (for example, consumers can stupidly hold out for flying cars). So this doesn’t seem like a relevant point — “not buying or investing in anything” is itself a market signal, and the reason productivity can increase in the first place (not to mention what an economy is _supposed_ to be for).
2A) So we agree that money must be a store a value in order to function as money, but you dispute that a) it must be a “good” store of value, and b)whether this is “the” defining property of money.
But I think the more important issue is whether *changes* in the “store of value” premium on money reflect real economic scarcities. I claim that it does, and so they cannot be removed without moving the cost elsewhere. See 4A and 4B.
3A) It still makes a huge difference whether Scrooge’s enjoyment of money swimming depends on whether the money in the money-swimming-pool itself is the same money used in the general population, so you haven’t removed the consumption aspect of the scenario. (And at this point I don’t think you can revise the example to remove these confounding aspects.)
4A) An explanation I would consider satisfying would start from agents, their resources, incentives, capabilities, information and expectations; in other words a microeconomic explanation.
I thought I provided enough that an answer to these was implicit. But I think your real objection is that I did not provide enough to show you that (what you call) the “Silas cost” is a real cost. Do you agree that’s where our disagreement lies? If so we can go more in-depth on that, though I think you can see it by reviewing a barter economy and why they shift to using money — would their choice of money-good (and the price of that money) depend on its value as a store of value, and does that dependence not reflect a real phenomenon?
I know I let you know that I saw the wheat analogies as wildly non-analagous.
Did I not refute the claim that the analogy was broken? I pointed to the corresponding real scarcities in each case.
Let me give you a different analogy: if people suddenly want to hold more bonds is it bad for companies to businesses supply more bonds (one could call this “printing bonds”)? I doubt you will say it is bad.
There’s a difference between that and the wheat/money examples I gave: in those cases, the government was rescuing people from wheat/money being *expensive*, while in this case, the companies are rescuing people from bonds having a low *quantity*. But you are correct that in both cases the adaptation method will have real costs — in the example of printing new bonds, the cost is that (assuming the demand for more bonds replaces demand for equity instruments) there will have to be less production that depends on equity financing.
I claim, however, that such a shift would be Kaldor-Hicks efficient, because the cost of shifting production methods is less than the benefit. This is certainly not the case when the government tries to suppress an increase in the price of wheat.
4B) Then I don’t know what jsalvati I’m responding to — I thought you came into this discussion as a Scott Sumner supporter who agreed with NGDP targeting. In any case, you seem to still be claiming that the money issuer should do something to keep monetary equilibrium as the Alts contract their spending, and UnAlt-serving businesses fail. Do you think I’ve shown how this prevents exactly the kind of adaptation necessary to satisfy the new consumer demands?
October 3, 2010 at 11:00 am
jsalvati
6A) I think *maybe* I know what is at the core of out disagreement: I think you think of money as representing some sort of real option to save without investing or trading a promise; that it represents some kind of thing which other goods do not represent. If money represents something real like this, then of course printing more sounds like an attempt to cheat. This belief is false. I’ve focused some of my discussion, 1A and 5A, on this in an attempt to convince you this belief is false. I ask you to be willing to examine and reconsider this belief (if it’s something you believe anyway).
1A) OK, I understand your comments about “but this rules out productivity increases!” a little bit better now. I didn’t mean to say that people shouldn’t be able to believe that both current consumption and investment opportunities are crap as obviously that can reflect actual states of the world. I meant to say that money should not give people any options that don’t exist in a barter economy; money doesn’t represent any special real investment opportunity different from ones that exist in a barter economy (though they might be more accessible). If both consumption and investment opportunities are crap, people still should have to choose some mix of them (including inventory accumulation, gold hoarding ect., though we generally futures markets for many goods which could be a problem).
I realized that the real problem here is that money does not bear interest. In a monetary economy where money bears interest, in times where the only marginal investment opportunities are rather bad interest rates (including the interest rate on money) should be low or negative. More on this in 5A).
2A) This is a fair description.
3A) OK, I don’t see it, but we can drop this, since it doesn’t look like it’s a productive discussion.
4A) I was thinking this is correct, but I think perhaps it is just a roundabout way of disagreeing as in 6A. If you think the approach out outlined here is easier than arguing about 6A, or you think it’s a separate disagreement than in 6A, go ahead.
Just to be clear, I do think other things equal, a poor store of value makes for a worse money. However, money can still be useful even if it’s a pretty poor store of value. People still use money even during times of high inflation.
You did not refute the notion that the analogy was broken in a way that made any sense to me. I think you are trying to find differences in those examples without really thinking about it: in the wheat case price is high (lets say flexible price), the money case people hold less than they want, in the bond case the price is high (assuming flexible prices).
4B) This is partially fair. I do still think that NGDP is a good rule, but I have explained *why* I think so, so I can’t be accused of “focusing on spending without understanding why”. I simply view deviations of expected NGDP as a very strong indicator of monetary disequilibrium, and I think scott sees this the same way; though he talks about it in this way less than he used to. Reading his writings is how I came to know about this view in the first place.
I do not think you have shown this.
5A) I realized that my bond example is a little more fruitful than I thought. Lets start with the fact that fiat money is a financial instrument: Supplier Of Money gives it value by promising “the ability of this promise to be traded for something of value will be stable (in some sense) over time”. The promise money represents is much fuzzier than the promises of other financial goods. Money is kind of like a standardized financial product; standardized in such a way that 1 unit has a value stable in some way. The Supplier Of Money takes a number of financial instruments and repackages them so they are standardized and can be used as a unit of account and a medium of exchange.
The case of an increased demand for money is like an increased demand for a particular company’s bond (money is but one financial instrument; there are many others). They can choose to shift their debt-equity structure, yes, but they can also increase the extent of their activities and turn the proceeds of their activities into bonds. In the same way, faced with an increased demand for their product, the Supplier Of Money can get some bonds and repackage them as a new financial instrument, “money”. This instrument may pay interest (possibly negative) to reflect the rate of the underlying assets (relates to 1A).
October 4, 2010 at 4:09 pm
Silas Barta
Okay, I tried to identify a few key points and focus and those, but couldn’t do it, so I’m just going to respond to your latest points individually.
6A) I partially agree with your point money not (discontinuously) conferring powers that barter cannot. For example, if people produce valued, illiquid good X and trade it for a money-like good (i.e. money in a money economy or highly-liquid, durable good like grain in a barter economy), call it MLG, this does not guarantee that MLG will be worth anything at a later date, and so does not guarantee that the sellers of X will ever be able to trade it for something they want. And so it follows that complete stoppage of MLG-spending will not function as a signal to better adapt to the demands of the X-sellers.
(During such a complete stoppage, the economy would probably not respond, as indeed I seem to want, by saying “How can we better serve you, o previous sellers of X?” Rather, it could just respond by abandoning MLG as an intermediate good, destroying any premium it commands above its intrinsic value, and then completely ignore what sellers of X want.)
But I (predictably, and perhaps fanatically) do not believe this affects my broader point that a (non-total) contraction of spending can serve as a signal that most goods are not wanted, and the economy needs to “retool” around the fundamentals: it should recognize that these recession-resistant goods still need to be produced, then seek out better ways of producing these, and restructure “ecosystems” around them. And this is why I see it as so important to clear out the deadwood. And why, in economic analysis, it’s important to keep the focus on whether the economy is satisfying consumer demands, not on whether variables are returned to their circa-2005 levels (as much discussion implicitly assumes).
So as long as there is enough money-trade in *some* large class of goods, hoarding money does provide a kind of guarantee of future value, by virtue of its signalling that different goods should be produced; therefore, a premium on money resulting from this, despite inconveniencing people who like to have certain cash balances, should not be suppressed. (If you find it problematic that people can hold on to money until the economy makes something they want, you should find it problematic when people spend money on things that weren’t even produced at the time they earned the money.) This brings us to:
4B) Basically, you don’t think I’ve shown that fighting monetary disequilibrium prevents necessary adaptation. So let me go into more detail on that.
In the Alt/UnAlt scenario you’ve described, how would a Sumner (or a superior Salvatier) Fed respond, assuming they don’t know about the aliens? They have to infer what has happened from the general macroeconomic data they gather, and they would see a uniform contraction of spending across all sectors (because previously, everyone liked the output, being UnAlts; while now, half the people are Alts). Per your or Sumner’s policy, this suggests loosening monetary policy somehow.
But what does that accomplish? In order to be an efficient CB and restore monetary equilibrium _quickly_, you would need to cut the value of money in half so that current businesses can have the same nominal revenues, which permits them to stay alive, even despite selling less. What’s worse, if the Sumner supporters are arguing about how great the central bank’s policy’s are, they will use this as an example, despite it being extremely deceptive: while a contraction was prevented “on paper”, the economy is providing goods half as valuable as it previously was — by leaving an entire class unserved. So even though the CB could argue that “hey, it was just a liquidity issue — our loans got repaid!”, it would only be true because the economy made less valuable goods, something hard to see in the usual data.
(And this is exactly what I think is going on with the Fed’s purchases of MBSes. But I digress.)
And if you accept that a contraction in spending can be due to people not liking current output, like in this example (or not liking current production methods) you should at least accept that the Fed needs to distinguish between the causes of monetary disequilibrium so that it doesn’t prop up businesses that need to die or seriously retool.
(By the way, I already took back the stuff about not understanding — and I would add that, in light of your comments since the beginning of this debate, you have a better understanding of the “why” of looking at spending.)
4A) We’re disagreeing on a lot of points here, so let me fall back and try to find where our point of departure is. Do you agree that the government cannot generally make a Kaldor-Hicks efficient improvement by trying to suppress the price of wheat? If so, briefly state what your own reason is for thinking so.
5A) Brief reply: Increasing the money supply — and indeed, the supply of any financial instrument — is unproblematic as long as the increase in that supply is matched by increased value at its point of entry. With gold as money, the increase in money supply comes with the option to use that new gold for non-money purposes. With gift certificates, some of which are as good as money, that new money is matched by the store’s merchandise.
And when corporations (are able to) issue new bonds, it is because they have created a kind of value through increased coordination — by identifying a value-increasing pattern of activity. (Note how this aligns with my point about the premium on money indicating the economy’s general level of coordination.)
These processes match couple new financial instruments to new value. In contrast, the Fed does not provide any kind of value with its increase in the money supply — it has not added something that can be put to many possible uses (like in the case of gold), nor do its actions identify a new, viable production structure (as when a venture can obtain new financing).
October 4, 2010 at 9:20 pm
jsalvati
7(!)A) Let me start off by saying that I have noticed that you tend to at least *talk* in terms of absolute levels (not sure if this mirrors your thinking as well), instead of relative levels. In economics it is the case that utility functions are the same under affine transformations (scaling + translation). People’s decisions are not affected if you somehow subtract 100 from their utility function, or multiply it by 1/2 (not that this couldn’t be a bad thing, just people’s decisions won’t be affected). I mention this because you frequently say things like “…most goods are not wanted”; it’s not that this doesn’t have a meaningful interpretation, but it’s not necessarily clear how valuations of world-states are moving around, and the obvious interpretation (utility functions are multiplied by 1/2) is not interesting, so it’s kind of a weird way to pose the scenario. You can construct a clearer scenario similar to what I think you mean, by saying “someone invents some really great entertainment technology that everyone expects to want and thinks will replace a lot of current business activity, but there aren’t any commercial projects for it yet”. Not sure if this is relevant, but I want to make sure you’re not making the mistake of thinking that this kind of change should have real effects.
6A)
Your point about keeping your eye on actually having a working economy is somewhat fair, but I think the mainstream view is more reasonable than you think. If one thinks huge changes in expected sector productivity and consumer preferences are relatively uncommon and that there is another relatively likely and fixable source of fluctuation in economic activity besides, then seeing huge changes in economic activity is strong evidence for the latter source.
I think you are wrong about the implications of what I have convinced you of. Let me try to be more explicit:
A reduction in spending for a particular good can serve as an indication of reduced demand for that good and thus an increased demand for other goods. A reduction in spending on consumer goods can be an indication of reduced demand for consumer goods and an increased demand for transfering purchasing power into the future. What I think I have shown to you is that in order to transfer purchasing power into the future, you have to arrange for a physical way of doing this (or getting a promise to trade across time). If the increased saving takes the form of holding money, this should actually arrange for real investment to be increased or arrange for an effective promise to be transferred. The way to do this is to create the quantity of money that has been incrementally held and spend it on something. If it it used to buy a financial instrument, either the person selling the instrument will purchase a newly created investment financial instrument, meaning real investment has increased, or if the person spends it on consumer goods then they have essentially agreed to trade away future consumption for in return for present consumption. If (not my favorite way of doing this) the government uses the newly created money to buy goods itself then it has essentially agreed to trade away future consumption for in return for present consumption. Cutbacks in consumption due to monetary disequilibrium might to this to some extent, but it won’t do it well, some of the “inventory accumulation” will be in products that do not save well (like labor) and this can be very inefficient it will additionally lead to welfare loss because it will produce less smooth consumption paths which give lower utility.
4B) I am a little confused about 4A)/4B) here. Your sentence “Basically, you don’t think I’ve shown that fighting monetary disequilibrium prevents necessary adaptation. So let me go into more detail on that.” sounds like something in response to 4A, but the ensuing elaboration is a discussion of monetary policy rules, and as I have said before, I don’t think it is the right time to discuss monetary policy rules yet; it’s better to separate the issues and discuss this issue after we have come to a consensus about monetary equilibrium, meaning I want to drop 4B until we come to a consensus. Discussing it now is just going to be confusing and unproductive.
It’s also important to note that “a drop in spending” and “monetary disequilibrium” are not the same thing. Earlier, I brought up the case of an exogenous increase in home or within firm production. In these cases the the demand for money would have dropped and spending would also drop; the correct course of action is to reduce the money supply.
I do agree that if you basically accept the point in 6A) but still disagree with adjusting the money supply to the demand for money you probably need to elaborate why non monetary-disequilibrium related incentives are not enough to get you to an efficient allocation of resources as we discussed in 4A), but this is not a good route to that.
4A) For simplicity, let’s assume that the domestic quantity supplied of wheat is fixed, so any incremental supply comes from other countries. The government’s option for reducing the cost of wheat is to subsidize it, encouraging importers to import extra wheat.
The inefficiency is that people in the private sector would do it to the correct quantity if people’s demand curves reflect their actual desires, so any incremental amount is inefficient, the benefits of increased wheat consumption do not outweigh the increased transport, foreign growing, and decreased foreign wheat consumption costs. The government trying to buy wheat in the domestic economy (in a closed economy) and reselling it does not affect the price of wheat because it changes the net supply and demand curves in equal and opposite ways, so the changes cancel.
5A) Is it problematic to issue new financial instruments on the basis of other financial instruments? If you borrow money from the bank (or in the form of a bond) and then buy some financial instruments say some stocks you think are undervalued (or just so you can tell your kids you own stocks), have you done something fundamentally illegitimate? You’ve just originated a new financial instrument (either a loan in the bank’s portfolio or a bond) without creating any value. Holding companies do this all the time, and many important activities throughout the economy rely on this kind of process; debt related processes like this are often called “leverage”. It’s important to realize that no one is asserting that these new instruments represent new wealth, either in the loan case or in the case of money.
I should clarify that some new value is likely created in both cases, otherwise people wouldn’t do it, but it’s not necessarily proportional to the book value of new instruments. This is also true in the case of money; new money helps relieve the demand for the medium of exchange which allows people to be less constrained than they were before.
October 5, 2010 at 7:42 am
Silas Barta
7A) I’m familiar with utility function affine invariance, and I don’t see why you believe the things I’m saying have some interpretation obviated by that concept. At a basic level, you can understand what it means for someone to not want to spend much of their money, reasoning that most things aren’t worth buying — and how the fraction of things they regard as worth buying changes. Indeed, this happens constantly at the fine-grained levels: when you’re at home relaxing, you’re not spending that very second — at that specific moment, all goods on the monetary economy are crap to you. I’m just pointing out circumstances where this extends longer.
6A) If one thinks huge changes in expected sector productivity and consumer preferences are relatively uncommon and that there is another relatively likely and fixable source of fluctuation in economic activity besides, then seeing huge changes in economic activity is strong evidence for the latter source.
Perhaps so, but your models must at least be *sensitive* (in the statistical/ information-theoretic sense)to the possibility of major “recalculations”. If they’re not, then you will eventually hit upon a situation where the economy needs to readjust and your monetary policy treats it as people making pesky preference-schedule shifts that you have to suppress. And in that case, the economy could be straining to do that *the whole time* and yet your statistics would never show it.
So if that’s your assumption, it’s a much stronger one than “the economy doesn’t currenty have major discoordination”; it’s “the economy has never had major discoordination anywhere as long as the monetary authority has been implementing this policy or one with a similar blind spot”.
What I think I have shown to you is that in order to transfer purchasing power into the future, you have to arrange for a physical way of doing this (or getting a promise to trade across time). If the increased saving takes the form of holding money, this should actually arrange for real investment to be increased or arrange for an effective promise to be transferred.
And I thought I’ve shown how this is wrong, or trivially (and unconventionally) satisfied: I don’t like current birdhouses. The birdhouse makers try something different, which makes the birdhouses non-crappy, and so I guy them. This happens all the time — it’s just a matter of whether you want to re-label that as “arranging for real investment”, since the saving caused the production shift (and productivity increase), but it was through price signals and not direct investment planning.
If (not my favorite way of doing this) the government uses the newly created money to buy goods itself then it has essentially agreed to trade away future consumption for in return for present consumption. Cutbacks in consumption due to monetary disequilibrium might to this to some extent, but it won’t do it well, some of the “inventory accumulation” will be in products that do not save well (like labor) …
This highlights a problem with your view I’ve been trying to point out: when government buys something, it’s less likely to be encouraging the production of things people actually want. Government has to spend that money on *something* — it can’t buy “goods in general”. And so it will distort the market in favor of those goods and their production methods; if those goods/production methods need to go away to make room for sustainable production, this is a giant step backward.
4B) It’s also important to note that “a drop in spending” and “monetary disequilibrium” are not the same thing. Earlier, I brought up the case of an exogenous increase in home or within firm production. In these cases the the demand for money would have dropped and spending would also drop; the correct course of action is to reduce the money supply.
But now I’m confused. Didn’t you earlier say that you don’t know how monetary policy should respond to this kind of “good” collapse in spending? Or at least that Sumner’s monetary policy doesn’t correctly respond to this due to insufficient understanding of the role of money? I had previously asked whether Sumner would permit a deviation from 5% NGDP trend if it were for a “good reason”, and people’s wants were still being satisfied (again, _supposedly_ the purpose of economic policy), and at that point you had to break from Sumner.
4A) Okay, then here’s my extension of that reasoning to monetary policy: make the same assumptions (domestically “produced” money is fixed — whatever that would require). So government sees money trading at a premium and wants to suppress that premium. So it “imports” money in a way that causes the private sector to do things differently than if it had to adapt to the higher price of money. That can mean continuing to sell things people would rather not buy at current prices, or selling investment instruments with insufficient risk premium.
Are you saying the analogy breaks at this point?
5A) It’s problematic to “transfer” liquidity from party A to B, while retaining A’s liquidity. That’s what happens when a bank takes A’s money, loans it to B, and then promises A that A can still access the money at any time. I know that a lot of people claim that bank customers “know” that they don’t “really” have full access to their accounts, but then why did the liquidity crisis of ’08 happen, and why does the Fed act as lender of last resort? If someone saves by buying a bond, they don’t expect to be able to redeem it instantly for its face value at any time — they bear the risk of it being liquid, and have incorporated this into future planning (as do people who save in IRAs or CD, which have withdrawl penalties, accomplishing basically the same thing). Nobody has to intervene to fulfill liquidity expectations on early withdrawl of CDs or IRAs.
So when a standard banking transaction takes place, the bank is saying, “Mr. A, you get $X worth of anything, and Mr. B, you also keep your claim on $X worth of anything.” If money’s value comes, in part, from its function as a (liquid) store of value, and if that function is valued inversely of the level of general coordination (as I claim), then something’s wrong here. By copying this money, they’re signalling to the economy that there is a greater level of coordination than there really is, allowing more businesses to commit to a mode of production than can really be sustained.
The same value is not created in the case of a bond vs. a savings account, as the former requires Mr. A to forgo his “anything” claim, which allows the economy to not have to (expensively) redeem an “anything” claim, while in the latter, it does.
October 5, 2010 at 9:47 am
jsalvati
7A) OK, great; I just wanted to make sure. Clearly my misunderstanding. Topic dropped.
6A1) Lets clarify something: Short run changes in nominal spending < 2 years may be appropriate or they might not be; I don't really have a lot to say about them. Changes in expected nominal spending over periods where mostly flexible are much more likely to be caused by changes in the demand for money. There are exceptions to the later, but "recalculation" is not one of them unless it also comes with a shift to non-market economic activity. In a "recalculation" people will either need invest into current investments (including accumulating and storing resources) or consuming the subpar goods that currently exist.
If a time comes when there's a major shift to non-market activity, it won't look anything like what we're experiencing now, and people will recognize it. People won't be unemployed and looking for other work, people will be working in their homes to produce whatever it is that's so great or simply enjoying leisure. NGDP targeting won't identify this, but it's not unreasonable to think the central bank would see that the cause of reduced spending is because people want to enjoy non-market goods.
6A2) I talked about this issue *right in the relevant paragraph*. Reducing spending by itself can reduce usage of resources, yes, and you can call this saving, yes, but it will do it very very poorly. For example, for most kinds of labor "saving" it has a very very negative interest rate, if you save 1x labor in period 1 and again period 2, you don't have anything like 3x labor in period 3; you have something like 1.1x labor.
If you save in the form of money, and there's no corresponding increase in the supply of money you are basically saying "here you guys can have these goods now, and I will require some goods at some point in the future", but without actually ensuring that anyone is willing to actually make this trade. This is especially problematic since you are trying to arrange this through consumer goods markets which are not designed for this, and not through the markets which are designed for this, financial markets.
I am having a hard time understanding why this is difficult for you to accept.
6A3) The choice of kind of asset for money to represent certainly can have real effects, and the optimal asset is probably something like "the market portfolio". If the kind of asset is chosen poorly (choosing something that people don't want to hold on the margin) this will obviously be inefficient. If you think that the Supplier of Money did such a bad job of choosing the asset that it counterbalanced the benefits of its money, then this would be an argument against having money (at least from that SOM), otherwise it's just an argument for choosing the asset better (not something I disagree with).
4A) If we were talking about a shock to the ability of the SOM to produce money that might make sense, but we're not, so it doesn't. It doesn't make sense to say the quantity of money supplied is fixed because it never is (well it could be but that would be a very unusual scenario, and in any case not what we're talking about). It's just not analogous. The analog of the producer's of wheat is the Supplier Of Money; they are the producer's of money, and it's their task to figure out the efficient quantity of money to supply.
5A) Liquidity is not a zero sum thing; it can be created or destroyed (might be difficult to do so, but it's possible in any case); it's frequently contingent on circumstances, and it's not 1 dimensional (liquidity is generally regarded as a two variable function, a surface), so I wouldn't put a lot of energy into such arguments.
Let's take money out of the argument here. Is it problematic to borrow corn from someone and use that corn to buy corn securities (which resolve in terms of corn)? You've just created a new financial instrument that did not correspond to an increase in value.
Maybe this is a better way to address the mistaken idea you hold: Financial instruments do not represent net wealth for the whole of their book value. Financial instruments are an asset to someone and a liability to someone else. If the first person considers it a better asset than the other person considers it a liability then there is net wealth. Fiat money is no different.
You can sort of say some financial instrument used to directly finance a valuable project represents net wealth if you draw a box around both the actual project and the financial instrument, but only because you have included the actual project in your box.
October 5, 2010 at 9:59 pm
jsalvati
Perhaps I have a more intuitive way of explaining 6A2:
A shortage of the medium of exchange causes non-market saving, perhaps you could call this “internal saving”. Agents save by attempting to accumulate the resources they use to produce goods other agents want instead of selling those goods and purchasing financial instruments from people who have a comparative advantage in conducting projects that transfer resources into the future. Since these methods are chosen because they are internal and thus do not require the medium of exchange, they are bound to be very inefficient ways of transferring resources into the future. Workers save their labor; manufacturers save machine-time.
Does that make more sense?
October 6, 2010 at 10:28 am
Silas Barta
I’m going to try to “renormalize” over our crucial points of dispute. As usual, if you think something is more importan than these, say so.
1C) [relates to 6A1 and 2]: I think our dispute here is basically over what kinds of adjustments to economic conditions are appropriate, and whether the CB can hinder them; I claim it cannot, because positive changes will happen on their own as prices shift, and the CB will not be able to correctly identify good vs. bad changes in e.g. NGDP (or whatever metric — that was just the one that bothered me when I read Sumner).
So I want to take an example of an economy that is recalculating, what I expect would happen, what I expect the CB would see, and why it would respond inappropriately.
Let’s say that people have been borrowing enormous amounts of equity from their homes, stupidly thinking that the homes would continue to skyrocket in value. Then, the housing market crashes (well, has a mild decline, but it’s as much a shock). People feel poorer, which is good, because they are poorer. They mistakenly thought that their homes would continue to produce more of what people valued, and began consuming this ephemeral value. That value did not turn out to exist. They no longer are capable of providing that which others value.
Therefore, they signifcantly curtain their spending. The businesses that were serving these fake value producers fall on hard times, and think they will need to shut down. The most marginal ones are bought up on the cheap, and the new owners repurpose them to serve a different class of people. Because people feel — and are — generally poorer, they are repurposed into satisfying things generally higher on the human preference scale (i.e. more in the direction of food and home production, less in the direction of luxury). They start looking for workers who can provide that function, drawing them
The Fed sees a significant contraction in GDP (real and nominal). Though resources are being appropriately repurposed, it notices that people don’t spend as much in, say, Las Vegas as they did five years ago. Rather than blow their money at Vegas, or invest it in a new Vegas casino while the ones in construction are on hold, people save a lot more of their money. Commentators warn that this is a sign of serious economic times, and we need to get out of this with policies that restore the business these business are losing, so they will be back at their levels as of five years ago.
Now, which financial instruments should the Fed create/buy, how would it determine this (noting how its metric are sensitive to this being a recalculation rather than something else), and why would it make that repurposing happen faster?
2C) [relates to 6A3, 4A, and 5A]: This dispute is about under what conditions I money can be costlessly created. My arguments:
I don’t claim that the supply of liquidity is fixed, just like I don’t claim that the supply of gold is fixed. But there are things that produce new gold, and there are things that simply shift around the cost of existing gold. Likewise, there are things that really produce liquidity, and there are things that simply shift its cost around. Because (I claim) the premium on money exists in large part because of discoordination, then, to produce liquidity instead of simply shifting its cost around, you must increase coordination.
If money is increased in correspondence with the increase in another resource (whether gold or coordination), that does not take resources from other people, and vice versa. If you mine gold, then yes, you’ve reduced the price of gold, but you’ve also increased the amount of gold that can be used for industry, etc. If you produce fake gold, you simply redirect resources to yourself. This is why I asked you about counterfeiting, and why I’m unsatisfied that you (and, IIRC, Scott Sumner and Steve Landsburg) conclude it *isn’t* harmful, so long as it’s done at the right time, which is a time of “tight money”. (You objected to a similar characterization before, but I honestly don’t know what a correct one would be, short of morphing it into my position.)
When a money arises in a barter economy (i.e., as people shift toward accepting salt for most sales, not to use themselves, but because others will accept it), they are, in a sense, producing money. I claim that the money is produced in a positive sum way because salt becomes moneylike to the extent that the coordination in the economy has increased. Once people recognize that they can reliably convert salt into anything, within a stable exchange rate, and this recognition is accurate, the patterns of economic activity are better coordinated. This is because people no longer have to make their production “fit” a complex exchange of goods, but can produce anything that will convert into salt.
Likewise, in a modern economy, when you save less of your money because you identify a way that you can satisfy some of your wants through barter or mutualist arrangement (like friendship or marriage), or simply make more extensive use of a product thatn you would otherwise, you are also creating money, but without simply redirecting wealth to yourself. Rather, by satisfying a want without having to make an intermediate exchange for money, you are increasing the level of coordination in the economy — the *way* that (If you’ll recall, I had previously appeared to claim that barter opportunities do not cause cash balances, and you got me to clarify that it doesn’t usually substitute at a full 1-to-1 rate.)
You call it (equivalent to) destroying money when you hoard it, and in a sense, I agree. If you hoard money because of discoordination in the economy (basically what is meant by uncertainty), you are doing the reverse of what happened to salt in a barter economy: because existing productions *stop* “fitting” the rest of the economy, it only makes sense that money would be destroyed, signaling others to increase coordination (which can be by looking for barter opportunities, or making your resources less specialized). This, I claim, is simply another case of prices reflecting relevant signals, and the economy reacting appropriately.
Unless the new money increases coordination, then, it is no more productive than counterfeiting.
October 6, 2010 at 10:35 am
Silas Barta
Note: when I said “more in the direction of food and home production”, that should be taken to mean in-household production, not house production.
October 6, 2010 at 10:38 am
Silas Barta
Also, ignore the hanging “— the *way* that ” in 3rd to last paragraph.
October 6, 2010 at 12:40 pm
jsalvati
1C)
I have posed 2 different ways for you to give your thesis a firm theoretical basis (6A2 and explaining the additional cost in the Alties example). You have definitely failed 6A2 and haven’t responded to the Alties example in a way that’s not a request for me to explain some other case, and I suspect that you don’t know how to show the additional cost. I think you should treat these failures and the other instances where I have showed your understanding of money was flawed as reasonably strong evidence that your position is wrong, and should acknowledge this.
Given that you still haven’t change your mind (though I am having difficulty understanding why), and no other obvious methods of explanation, I will attempt to respond to your example. I am somewhat reluctant to respond to your example because it gets us into monetary policy rules, which is going to be difficult to discuss without first agreeing about how money works and what kinds of situations are improvable and which ones are as good as they can get.
Perhaps I should have mentioned this some time ago, but I didn’t because I figured it was obvious I thought this: “recalculation” by itself does *not* get you a reduction in aggregate spending; you need either a change in the demand for money.
Let me state some assumptions I would like to make for simplicity so you can say whether they assume away something you think is critical, and ask you some questions so I can avoid answering the wrong question.
I assume that the housing stock is constant (people aren’t investing in building new houses) if you really think this is important we can bring this back in.
I am also going to divide the world into two groups, homeowners and non home owners. I am going to assume that because homeowners believed their homes were going to appreciate in value they were taking out loans from the bank in order to smooth their consumption and consume more now and less than they would have otherwise later. The bank was offering pretty attractive savings rates because these loans looked attractive, thus the homeowners saw an opportunity for increased consumption later for somewhat reduced consumption now, so it was primarily the non home owners money that was channeled to the homeowners for present consumption.
You’ve left out a critical feature of your example; one that I think you should have been able to predict: who are the people with increased demand for money, and why? If it’s the homeowners, are they reducing their spending beyond not getting and spending new equity loans? Is it the non-home owners?
2C)
If you’re going to fall back on the position “well, creating money *must* have problems because I think the premium on money exists in large part because of discoordination” then this part of the discussion is a dead end. I hoped to prime your intuition by showing you how creating money was similar to the process of creating any other financial instrument and thus there is no “cheating” going on. Clearly this has failed. Keep in mind that your ability to reject this is not evidence against my position since you used the fact that you believe “premium on money exists in large part because of discoordination” as the reason for not changing your intuition, but this is the proposition you are trying to support in the first place.
October 6, 2010 at 12:56 pm
jsalvati
I should apologize for my statement ““recalculation” by itself does *not* get you a reduction in aggregate spending; you need either a change in the demand for money.” that’s not really what I want to say, and I’m not sure how to say it correctly, so scratch that for now. I suppose that indicates that it shouldn’t have been obvious.
October 6, 2010 at 1:11 pm
jsalvati
I should also ask: in your example, what is the unknown? What is the thing that is being recalculated? Is it what to do instead of building houses? (I suppose I need to add back housing production then) If not, what else is it?
October 6, 2010 at 1:17 pm
jsalvati
Lots of afterthoughts here. I should also point out that the Alties example was seems fairly similar to this, and I established that unless you could point out an additional cost a failure to change the quantity of money for a change in the demand for money was just going to produce additional difficulties than those coming from recalculation. You were unable to come up with the additional cost. If you do think this example is different in an important way, then let me know how.
October 6, 2010 at 2:26 pm
Silas Barta
Okay, a lot of your last response is based on the claim that I haven’t substantiated my own claim of a cost in the Alties example. But I thought I explained it very thoroughly: the cost is that the economy fails to retool to satisfy the wants of Alties because UnAlt-orientied businesses are propped up. (This is very similar to the cost that a counterfeiter imposes: resources are redirected toward him, despite him not providing resources that actually satisfy others’ wants.)
Your reply was to say, basically: No, your monetary policy wouldn’t do that, because you wouldn’t have the kind of MP that Sumner wants. If that’s the case, then I emphasize: I don’t know what position I’m criticizing anymore.
We started this exchance because I was frustrated with Sumner’s (IMHO ridiculous) belief that if we just get people spending, our problems will go away — even if it’s really stupid spending that requires us to arm-twist people, and rewards and props up deadwood businesses. If you’re not defending that anymore — and the discussion keeps regressing to points related to that — then it seems my original claim is vindicated: Getting nominal spending back up is not costless, and can be actively harmful, especially if the economy is not making what people want, and so has lost its purpose (in no longer valuing things based on how well they satisfy a want).
So, I think that, if we still have disagreement, it’s because you believe your favored monetary policy and Sumner’s are the same at a sufficient level of generality. That is, although you might advocate doing something different from Sumner, you both advocate doing the same *kind* of thing, because it costlessly prevents the same *kind* of problem with the economy. And we should be talking about that.
Furthermore, one of my prime concerns about MP is that it encourages production that needs to stop (because it has lost connection to the economy’s purpose) — so even if I were to agree that, in the abstract, nominal aggregate demand needs to increase, that would still not answer the question of how you know that the cost of low nominal AD is worse than the cost of these distortions introduced: if the CB buys government bonds, facilitating government purchases and programs, why will that lead to sustainable patterns of production, appropriate to actual preference schedules? If it buys crappy MBSes, how does that do likewise? If it makes loans to clumsy, incompetent banks that just failed at their primary role, how does that do likewise? And so on.
I think you’re dismissing these choices as mere details — but that’s exactly what monetary policy has been doing this whole time! If you are distancing yourself from these policies because you now recognize costs, then I’ve accomplished my goal of showing MP not to be costless like you thought.
Finally, some meta: I think both of us are frustrated, not just because we disagree, but because we think our positions are obviously right. You, because the things I’m disputing are widely-accepted basics. Me, because the monetary policy advocates cannot pass the basic test of connecting their concerns to an economy’s purpose.
There’s more I could say, but I think that clarifies where I think we are in this discussion.
October 6, 2010 at 2:56 pm
jsalvati
Let me pose you a different kind of example. Consider an economy with two closely related firms in it (lots of other firms too). One of these firms makes an intermediate good for the second firm (which doesn’t have other suppliers of this good). They have long been on the cusp of it making sense for them to merge into one firm in terms of shared activities etc. (lets say switching between being 1 firm and 2 firms are low). One day a slight tax change makes it slightly more efficient to be one firm. Their real activities don’t change much at all, but the sum of their normal cash balances is lower than it used to be because they don’t need money to conduct their internal trades, just their outside trades. So they try to get rid of their excess money; now there is monetary disequilibrium. Now there will be shortages of goods as people try to buy more goods than there are or production will be increased beyond its equilibrium level. This also means that after prices adjust the price level will be higher than it was before. Do you think in this case it is still appropriate for the money supply to stay the same? Or do you think, as I do that since the demand for money has decreased, the supply of money should also decrease? It may be relevant to know that this is a case that NGDP targeting will miss; spending will decrease because there has been a shift from trading to non-trading, but spending increases elsewhere because of the excess money, so total spending stays roughly constant even though a decrease in the money supply is appropriate.
October 6, 2010 at 3:23 pm
Silas Barta
In the example as specified (though unrealistic in irrelevant ways), coordination has increased, so money *should* fall in value to reflect this, and should not be resisted. Such a signal (falling value of money) correctly conveys to people that they should spend fewer resources searching for alternate modes of production, as the two frims have stabilized around a very good one.
October 6, 2010 at 6:17 pm
jsalvati
OK, at least you’re consistent.
October 6, 2010 at 3:25 pm
Silas Barta
Also, I dispute your claim about shortages — business just raise prices as the demand for the specific goods (that are now able to be purchased) increases.
October 7, 2010 at 7:42 am
jsalvati
This is just a denial of sticky prices. Is that really something you deny? Incidentally, output above equilibrium output is a much more likely outcome for most markets as most markets are monopolistically competitive so they have some degree of overcapacity.
October 6, 2010 at 3:30 pm
Silas Barta
One more thing: to head off a potential fruitless exchange: Yes, adapting to these changes (lower value of money) has costs. And in the abstract, it would be nice not to have those. But attempting to eliminate them (I claim) necessarily comes at the cost of impeding a much more important economic goal, of ensuring that businesses adjust to sustainable modes of production that satisfy actual consumer demand.
To prioritize elimination of these adaptation costs is to put the cart before the horse: if people don’t have the appropriate signals about whether they’re in the right line of business, the economy is unable to satisfy its “prime directive” of satisfying wants in the first place.
October 6, 2010 at 6:16 pm
jsalvati
Sorry, I hadn’t realized that you thought you had answered the Alties example.
If your argument really is that monetary policy “props up” or (presumably) “pushes down” specific kinds of activity, then your error is simply a failure to appreciate efficient financial markets. I hadn’t responded to these kinds of comments before because I thought that error was clear, and you were just saying such things as an annoying rhetorical device. If (roughly) efficient financial markets are true, then if you buy one financial instrument (and don’t have special knowledge or anything), then it will shift supply of funds for investment in general, not affect the price or quantity of that particular financial instrument. If you buy a financial instrument and push its price higher than estimates it was before then it’s marginal holders sell it and buy other assets. Now obviously financial markets are not 100% efficient, but they are say 99% efficient, so this is a pretty good approximation. The expected future profitability of those businesses and projects will be low, so the prices of those assets will be low and give a clear signal that starting new similar projects is unprofitable even at the new supply of investment funds. Now you can argue that the increase in the supply of investment funds is in efficient, but that’s different and something you have definitely not yet established.
I am pretty sure I have explained my position explicitly before (I think twice), so I am not sure what it is that I am not communicating. Let me try to make it more clear. My position is that 1) monetary equilibrium is the correct way to think about money. In addition to that, I think that 2) NGDP expectations targeting would be a significant improvement over current policies and certainly not “ridiculous”. This follows from monetary equilibrium theory, and my expectations about what kind of changes are important and common. I also more strongly believe that 3) NGDP targeting would have been a significant improvement in this crisis over current policy, this follows from my prior beliefs and my understanding about the evidence on what kind of shock we had during the crisis. My understanding of your position is that you do not believe monetary equilibrium is the correct way to think money. I don’t think it would be at all productive to talk about 2 or 3 until we have consensus about 1, which is why your attempts to talk about 2 and 3 annoy me, especially when you want to conflate the two or switch from talking about 1 to talking about 2 or 3.
I don’t talk about those choices because I think in terms of efficient financial markets, where it won’t make *that* much difference. It’s not unimportant, but it’s not the most important issue either. No doubt that the Fed buying MBS gives people who were in MBS’s a slightly cheaper way out and others an opportunity to skim off the top. The way the the Fed conducts open market operations isn’t the optimal way to conduct monetary policy either (I think there’s a good ol’ boys network of government bonds dealers). But neither of these points are new, they’re basically generic critiques of government that I naturally agree with, and the sums that rent seekers get are trivial in comparision to the problem of monetary disequilibrium. I didn’t mean to imply that I thought doing monetary policy was *literally costless*, I have never thought so, just that it was very low cost in comparison to the problem it can solve; even +1% unemployment is a HUGE cost.
October 7, 2010 at 9:55 am
Silas Barta
Thank you — the points made in your first paragraph do much to clarify where I disagree with you.
I dispute that financial market efficiency could undo the kinds of misallocations I’m describing. For one thing, you seem to be making a fallacy of composition, in that you’re saying, “Because one financial asset purchase can’t make a difference [by prompting a reverse act from other buyers], no amount of financial asset purchases can make a difference.” But obviously, the larger these purchases are relative to the entire market, the more they impact the equilibrium price. I can’t influence the price of a Treasury Bond, but the Fed sure can!
More importantly, though, the CB does not simply buy some fixed number of financial assets. Rather, it acts as the controller in a feedback control system — it buys however much is necessary to establish a new equilibrium. This necessarily distorts prices (since that’s the entire point of it!) and conveys information to the market, which is then used to make decisions. And since it cannot buy “all assets”, it will buy some rather than others, distorting investment in some arbitrary direction, disconnected from (i.e. has no mutual information with) actual consumer preferences.
(Since one of the things the CB does is ensure that consumption loans return to their c. 2005 level, it also promotes consumption at a time when it should be contracting because of the contraction in wealth and productivity.)
Furthermore, the MBS purchases in particular had an additional side effect: since they saved the financial lives of enormous, clumsy banks, they (in contravention of how a market system is supposed to work, if you value its ability to satisfy wants) kept them from having to liquidate, turn over their assets to non-idiots, and let new, innovative financial intermediaries (like Virgin Money), who aren’t stuck in 2005-appropriate intermediation methods, gain market share. You’re really claim that this didn’t distort the structure of production away from a sustainable one that satisfies consumer desires?
I’m also glad you listed the 1), 2) and 3) as the chain of reasoning you want to go through. I remind you that I believe the correct way to think about money is in terms of whether it satisfies wants across society. If that requires that money be abandoned entirely, well, then all the worse for money! (Avatar economy? Fine by me.) If it requires monetary disequilibrium, then all the worse for monetary equilibrium. I only care about these to the extent that they connect back to satisfying genuine individual wants. Every economic policy must be justifiable in these terms; I really don’t care about NGDP or GDP or spending or employment or inflation otherwise.
I hope that is more responsive to our fundamental disagreement.
October 7, 2010 at 9:57 am
jsalvati
It might also be useful to add a proposition I believe that I am more likely to disagree with Sumner on and more likely to agree with you on 4) it is important to think about and discuss policy in terms of monetary disequilibrium (and other microfoundational ideas) as opposed to “aggregate demand” and similar ideas as it is easier to get confuse yourself and others using the latter ideas. Since I think you would roughly agree with me on this (though you might disagree with me on what the microfoundational ideas are), I don’t think this idea is being debated.
October 7, 2010 at 11:01 am
jsalvati
If people hold instruments for something other than their discounted expected cash flows, then sure, but to a pretty good approximation, people hold financial assets for their expected discounted cash flows.
Yes, open market operations will change the equilibrium price level and *all* financial instrument prices will reflect this, but this was my point to begin with. Financial prices as a whole will be affected, not the prices of some specific sector. Even if people have different expectations of cash flows as long as shorting is allowed. If some sector becomes less profitable in expectations, related financial instruments will have lower relative prices, open market operations will not affect this. Such actions shift the supply curve of funds for investment, but don’t “prop up” any specific sector, but you were claiming (or seeming to) they did. You can see this when the Fed announces some new expansion, people talk about how “the market” rose, rather than some specific sector doing much better than it was.
This is pretty basic EMH stuff. If you really disagree with me here then you probably should spend time arguing with those crazy EMH advocates rather than arguing with me about monetary equilibrium, since your complaint is really that there’s no good way to conduct monetary policy and not that monetary disequilibrium is actually necessary.
In order to argue your case, you have to provide evidence that the shift in the supply curve of funds for investment is bad or that the change in the price level (back to the value it had before the change in the demand for money) is bad, and you haven’t done that and was what I was pointing out.
Perhaps people talk about how consumption loans should go back to 2005 (though I rather think you are exaggerating the extent of this), but the Fed is certainly not targeting this. The evidence is irrefutable: consumption loans are lower than they were before and the Fed has not continued to expand the money supply.
I don’t really discuss MBS a lot because it was not monetary policy in the sense that we are discussing. The Fed specifically “sterilized” their purchases of MBS by selling bonds in the same $ quantity that they were buying MBS in order to avoid making it monetary policy. The quantity of money was unaffected by this action. The Fed could have easily done as much monetary policy as they liked without doing this (probably saved themselves a lot work and a lot of criticism too), and another part of the government could have easily taken the same action. It was really more of a Fiscal action than a monetary action. I don’t know a lot about this topic, so I can’t say if it was a good idea or a bad idea, I would guess bad idea. You can criticize this action as much as you like, and I won’t have much to say about it. It’s not really all that relevant to the discussion.
I didn’t list “I care about people’s utility, not nominal variables” because it’s obvious, and you don’t disagree nor does anyone else of note; it’s not at debate. We’re discussing monetary equilibrium because we disagree about how it affects people’s utility. I don’t have a problem with abandoning money if that’s what’s efficient either. If it’s no longer necessary to use money then the demand for money will drop, and the central bank will decrease the money supply until it disappears. NGDP targeting wouldn’t handle this well, but the general policy of changing the money supply in response to changes in the demand for money doesn’t have a problem with this.
October 7, 2010 at 11:36 am
Silas Barta
If you’re going to endorse some blanket version of the EMH, then why don’t you also believe that the market will efficiently adapt to the money supply that currently exists without CB help? That door swings both ways.
In order for EMH to be relevant, we need to identify which formalism of it each of us agrees with, and how it favors our preferred policies.
Also, do you really think keeping incompetent banks alive through the MBS purchases did not introduce the major structural inefficiencies I worried about, for the same reason it would be ridiculous idea to bail out breadmakers that have long used ridiculously inefficient methods to make bread but have kept the losses hidded?
And yes, consumer loans aren’t back to their 2005 levels. I think that’s a good thing. But your intellectual comrades-in-arms seem to regard that as a failure and want to pump it back up. I can dig through Krugman, Sumner, Daniel Gross, DeLong, etc. if you want.
If you’re going to abandon your support for these major pillars of the “only monetary disequilibrium could possibly be the problem” crowd, it’s not clear what’s left for you to defend. If you already agree all of that is bad, then I’m not sure how you can really blame monetary policy for the economy not recovering (in the sense that we *care about*, not the raw GDP or employment numbers).
October 7, 2010 at 12:19 pm
jsalvati
I think you’re being silly. The difference should be obvious: financial markets are have a very short time scale on which they become efficient (information about fundamentals is reflected quite quickly) while product markets are efficient on a longer time scale (it takes time for prices to adjust since it takes time and effort to changes prices and for those changes to propagate throughout the economy).
Changes in the equilibrium price level fully show up very very quickly in financial market prices, but take a lot longer to fully show up in product markets (I’d guess ~ 1 year). Is it really hard to believe that non-financial markets have significantly sticky prices and financial markets (at least public financial ones) do not?
Incompetent Banks – I certainly don’t rule it out, but I don’t think it’s 100% obvious either. Keep in mind that several of the rescues are expected to actually make a profit. Not familiar with the magnitude of the premium the government paid to get these instruments, but I do know that they had already lost a huge fraction of their value, so it’s not like the holders suffered *no* significant losses. I am also think that poor monetary policy probably exacerbated the real problems with MBS and the banks, making the financial crisis worse than it would have otherwise been. A change in the equilibrium price level loan holders pretty directly in real terms; a big enough shock can make them insolvent even though the activities they were doing were profitable in real terms.
In any case, MBS are not going to be popular for a long time. If new MBS’s are not getting originated, it’s not the case that they’re being “propped” up. It may be the case that we will get more similar kinds of projects on the margin some years down the line because the government has shown it’s willing to cover some of the losses (moral hazard), and I agree that’s bad, and we should strive for better/less government because of it, but it’s not like similar problems don’t already exist, so it’s not going to be the end of the world.
I think in most cases people talk about reduced consumer loans as a *sign* of bad things, rather than a bad thing itself. You can disagree about the interpretation, but that’s a different argument than you’ve been making.
Did your really think that “poor monetary policy is the only government caused problem anywhere” was my position? I am sorry if I gave that impression, but it isn’t and it never was. The government does many stupid things (during this crisis and other times) and they are of varying degrees of importance, and I doubt many of the people you listed would disagree. That doesn’t stop poor monetary policy from being an important problem. Just because I think other things were poor policy doesn’t mean I think that monetary policy is not a major driver of current problems. There are lots of poor policy decisions that aren’t *disastrous*.
October 7, 2010 at 1:40 pm
jsalvati
I think that my previous answer to your first statement was not as clear as possible: financial markets can and do adjust quickly to changes in the demand for money, but this doesn’t stop a shortage of money from being a problem in the real economy. Consider the following: some agent wants to accumulate more money, so they spend less, other people receive less income so their money balances dwindle, if they still want to hold the same quantity of money, they sell off some assets in financial markets, this is kind signal that financial markets get that causes prices in financial markets to adjust to the new equilibrium price level. Meanwhile in the real economy, prices are sticky and the demand for money is an increasing function the the amount of money transacted, so the shortage of money is still a problem for conducting real transactions. The fact that prices are not sticky in financial markets is helpful since it means that the demand to hold money related to financial markets will generally adjust quickly reducing the costs of an increase in the demand for money, but that doesn’t eliminate the fact that in other parts of the economy demand to hold money does not adjust as quickly.
October 7, 2010 at 9:16 pm
jsalvati
When you say “coordination has increased (decreased)” what do you mean descriptively? Specifically, I am interested in how you would distinguish the concepts of coordination and productivity. Feel free to copy paste if you think you’ve defined this before (I think maybe I asked this before but I can’t find it).
October 8, 2010 at 10:08 am
Silas Barta
I’m not sure how the relative efficiency of financial instruments makes a difference here. If the problem is that the wrong non-financial changes are being made, and non-financial markets are less efficient, then the non-financial markets are limiting factor in the adjustment speed. This inefficiency will be inherited by the trading of financial instruments, because they must make judgments about the real viability of the real production processes. They can only learn this information as quickly
Furthermore, the distinction disappears if and when the part of the financial market that would fund the shift toward sustainable production modes is not part of the standard exchange. A venture capital group’s equity purchase of an innovative new production method is *not* traded on a liquid exchange, and so is not in the efficient part of the financial market. (I believe that most methods involved in getting an economy back to sustainable patterns of specialization and trade are of this type, which is why I find it so repugnant to make purchases and loans on preferential terms to established players.)
I think in most cases people talk about reduced consumer loans as a *sign* of bad things,
Oh, I wish. I really, really wish. Unfortunately, one thing I’ve noticed recently is that people have completely lost touch with the difference because _causes_ of good things vs. _indications_ of good things; for example, virtually every mainstream analysis I’ve read equates increasing inflation with an improving economy and suggests we need to cause this *so that* we can have a good economy, which blurs the distinction. I remarked on the issue here.
Did your really think that “poor monetary policy is the only government caused problem anywhere” was my position?
No, what I (thought I) said was that you, like Sumner, find tight money to be the definitive factor preventing a recovery, and so if this problem were solved, the economy could return to an equilibrium in which, of course, it’s actually satisfying genuine consumer wants. If you *don’t* think this is the definitive factor — if you just think that bad monetary policy is just causing suboptimal consumption here and there, but no _structural_ problems — then I still disagree with you, but we’re not having the fundamental disagreement I thought we were having, over whether pumping up spending would be a big move in the right direction.
Regarding coordination vs. productivity shock: A discoordination is a kind of productivity shock. What makes it different (and I touched on this briefly before, though it probably wan’t clear) from a non-discoorinative productivity shock is that a shift to a sustainable mode of production requires that the factors of production leave their local domain of attraction. That is, they must jump out of a local optimum and go to a (more) global optimum. This implies that a “hill-climbing” type algorithm will not find a new stable equilibrium, so simply adjusting local parameters (e.g. hours spent maintaining equilment) is insufficient.
This actually relates back to the question of efficient markets. You might have heard of a paper floating around claiming that markets are efficient iff P = NP (i.e. ease of verifying a solution implies ease of finding a solution), a claim that most relevant experts believe to be wildly implausible but cannot yet prove so. Without having read the paper, I think the way it gets this result is by definining efficient markets as markets which always find a (fully) global optimum (which is in NP), while markets are limited to using hill-climing algorithms (which are in P).
Whether or not you require a market to find global optima before you deem it efficient is therefore a crucial question.
October 8, 2010 at 10:10 am
Silas Barta
Sorry, poor proofreading again — the 1st paragraph should end “… as quickly as the non-financial sector learns them”.
October 8, 2010 at 10:14 am
Silas Barta
And a clarification: I was saying that experts find “P = NP” to be wildly implausible, not that experts find “markets are efficient iff P = NP” to be wildly implausible — although I think the relevant experts would believe that too.
October 8, 2010 at 10:29 am
Silas Barta
One more thought: There’s a background assumption I’ve been using throughout, and I want to state it explicitly.
A paradox exists whenever you have two conflicting intuitions, both of which seem to be solid and reliable. To resolve them (following Gary Drescher in Good and Real), it’s not enough to point to one of them and say, “Yep, that one’s right. Problem solved.” Rather, you must *also* show how the *other* intuition goes astray.
When I look at Sumner’s argument, I see a paradox. The paradox is that “basic”, “reasonable”, “mainstream” economic analysis implies that (as I ) we should make lots of ultra-underpriced loans to incompetent, dinosaur banks. (Yes, you can improve on this by saying, “Okay, fine, no privileges — *everyone* should get zero-interest loans”, but in my view that’s just another paradox.)
So to satisfy me, it’s not enough to say, “Yep, that’s our standard economic model, I makes sense and I can show you how.” You have to also show why the intuitions underpinning the belief that “we should make lots of ultra-underpriced loans to incompetent, dinosaur banks” is wrong — and that involves a different argument than substantiating the economic model that says otherwise.
And I know you think, upon reading this comment, that you think you’ve already been doing this. But I really think it will give you a better idea of what arguments I find satisfying.
October 8, 2010 at 8:23 pm
jsalvati
A) OK, so you agree that if financial markets are roughly efficient (and think financial markets are roughly efficient), open market operations aren’t going to “prop up” any particular asset or sector, yes? If you do agree, then I hope you will agree that this makes maintaining monetary disequilibrium *at least non-harmful*.
And now you have moved on to denying that prices are sticky in the sense I have been talking about, basically you’re claiming that in efficient product markets prices are only slow to change because it takes time for information about how shocks affect the efficient price to emerge. This is not what economists mean by price stickiness. Price stickiness means that prices are slow to change for additional reasons: menu costs, usage of heuristics in making prices changes etc (you might also consider differential knowledge theories: actual price setters don’t know how to interpret monetary policy actions and such). Of course, you can deny that such price stickiness exists if you like.
B) You seem to assume that people you disagree with to be as stupid as possible and still fitting the facts of what they say. A significant fraction of macroeconomists seeing inflation or spending as “good in itself” seems like absurdly implausible psychology. A far more plausible theory to me is that they have some model for the economy (could be a good or bad model), and increasing inflation or spending at this moment would improve people’s welfare. The obvious counterexample to your theory is that macroeconomists don’t think inflation is *always* good. Few macroeconomists would say that 15% normal inflation or spending growth is good, and if we had been having higher than usual inflation, many more macroeconomists (Sumner likely included) would be stating the opposite, that decreased inflation or spending would increase people’s welfare. It’s fine to criticize their model, but it’s absurd and unproductive to claim that they think that some nominal variable in itself is good. The only thing such arguments accomplish is making yourself feel smarter than all those backwards economists.
C) I don’t think I would say we would necessarily return to an equilibrium, since it could easily be the case that we are in a period of adjusting production and that needs time to work out (when are we not in such a period?). Sumner has talked about this a few times, and it seems plausible enough to me. My position is that monetary disequilibrium is an additional problem that has additional costs, and that a large fraction of apparent problems are due to this, and that reducing monetary disequilibrium (which currently involves things that would increase spending) would improve things rather than make things worse. So we do still disagree.
D) I do think I’ve been doing this. The fact that there are ways of attempting to address a problem which cause other problems does not imply a paradox for the theory describing that problem. I think I have noted this logic before, and you said that the supposed solutions are like subsidizing pollution. OK, but if those solutions don’t have benefits in standard theory, you shouldn’t consider them in the first place.
Would you say that the theory of externalities recommends banning cars (since they pollute), and that is obviously ridiculous so there must be a paradox? Similarly, one can conduct monetary policy without subsidizing banking companies. The Fed actions to that effect are not generally described as “monetary policy” they were described as “rescuing the financial system”, and for good reason: they don’t have a lot to do with monetary policy (except interest on reserves, which was acknowledge by the Fed to be contractionary so in that sense it’s disrecommended by standard theory during current times).
October 10, 2010 at 5:16 pm
jsalvati
I’ve been trying to figure out “coordination” as you use the term (I suspect it of being incoherent or irrelevant), and I have two question.
Coordination1) What you’ve described doesn’t obviously (to me) have anything to do with how people coordinate things or how coordinated they are. Do you have some link in mind? If you don’t then I urge to rename your idea (http://lesswrong.com/lw/ng/words_as_hidden_inferences etc.)
Coordination2) You described a coordination shock as a productivity shock that requires a change that won’t be found by local optimization. However, you seem to also use it in a different sense, as you talk as if coordination had a “level” (not necessarily an absolute level; perhaps a level in the same way the price level has a level). It’s not obvious to me that coordination as you described has a “level”. If you do mean to imply coordination has a “level”, can you elaborate on this point?
October 11, 2010 at 8:27 am
Silas Barta
October 11, 2010 at 8:28 am
Silas Barta
What the heck? My post went through as blank. Let’s try that again.
A) No, I don’t. I might agree that open market operations don’t prop up assets relative to the rest of the reference class CB operations choose from, but the act of these purchases promotes assets _in_ that class over those outside it. To put it bluntly, allowing a large, inefficient institution with a very poor history of directing resources to productive ends (say, it optimized for hiding the risk of investments that is only now being discovered) to have access to cheap loans will favor it over Dan the Disruptive Entrepreneur, whom the CB does not call when performing open market operations, and whom traders do not invest in as a result of being able to offload (temporarily) overpriced assets to the CB.
B) If I saw any evidence — any whatsoever — that MMEs were making the distinctions you claim, I would not be harping over this. Instead, we see things like Blinder’s article, which Sumner cited with glowing approval. Read it carefully: he wants banks to lend, period. That _any_ increase would be “better than nothing”. No caveat about “assuming this doesn’t delay the shift in business models that needs to happen”. Just that there’s more lending — in anything — is good. Lending at negative interest rates? Even better.
If you can show me where any MME has ever inserted any such caveat — anything related to how “actual satisfaction of consumer wants” screens off any of the measures they’re focusing on (or why such a caveat is implied *and* understood) — I will gladly admit my error.
And while MMEs might not deem 15% inflation good, the certainly consider 2% inflation good, and deflation or non-inflation to be necessarily bad. And why? Well, because that would allow people to just hold on to their money — they wouldn’t invest it. That argument only works if you regard investment as a terminal, rather than instrumental value.
C) Okay, so you’re saying that monetary policy can aid in a recalculation (if that is indeed going on), and that its failure to do this slows the process? So that MP will mean a faster response to changing consumer wants than otherwise?
D) The analogy still works — pollution subsidies *do* have benefits under current theory: they push out the margin of production and allow for previously non-viable production to take place (a product might appear profitable now because the maker can collect the pollution subsidy), and thus increased consumption. (Dare I mention the increased AD due to businesses “selling pollution” to the government?) It’s just that the costs exceed these benefits. The arguments for mainstream monetary policy, however, claim the costs *do not* exceed the benefits.
This would imply, as I claimed, that private counterfeiting can be good, so long as the counterfeiter works at a time of tight money. Hence, a paradox.
And now you look to be any defining monetary policy you don’t want to defend (or do you defend propping up the big banks?) as “not monetary policy”. But even so, wasn’t the point of these to restore liquidity, the very same thing you propose mainstream MP to maintain, even if it involves overpaying for assets (because you think efficient financial markes will correct for this)?
Do you think that taking a chance of being illiquid is not a real cost? That, when I buy a bond, no cost is incurred by someone guaranteeing that I’ll always be able to redeem it at face value?
E) About the level of coordination: I don’t think my choice of words differs from its normal use; it’s just that there will always be different levels of granularity at which you can apply a concept. (Compare: is “central planning” good? Depends on what kind of central planning.)
Continuing the energy landscape metaphor (where energy corresponds to total utility), you can take the “level of coordination” to mean the maximum of the local domain of attraction. If that is the same as the global optimum, then that economy is at the maximum level of coordination (even though it might be very far from the global optimum, and below numerous other local optima); the economy can proceed directly to that global optimum while only making incremental changes that involve looking at its immediate neighbors.
In contrast, a less coordinated economy is stuck at a local optimum which requires “jumps” to get out of. Policies favoring current production structures (e.g., what I claim MP is) amount to making this domain of attraction steeper and so harder to jump out of.
October 11, 2010 at 8:29 am
jsalvati
Something went wrong with your comment; I don’t see anything.
October 11, 2010 at 10:48 am
jsalvati
A) This is simply an argument you’re not going to win; the deck is stacked against you.
I assume by “cheap loans” you mean “buying bonds”. Basically all entrepenurial projects involve loan markets; when the Fed buys bonds, it increases the supply of funds available of investment. Bank loans, venture capital etc. are also affected by this. If you find it helpful to go back to the mechanism I mentioned before: the Fed buys bonds which raises their price and makes them look less attractive than other investments, people move out of bonds and into other investments until bonds/other-investments look equally attractive. This applies to funding for entrepreneurs too: when other investments are expensive relative to the entrepreneurial investment, the supply of funds for investment available to entrepreneurs increases until they look equally attractive. Now, the market is not *perfectly* efficient, so perhaps this gives whoever’s bonds the Fed bought a small advantage, but the effect I am describing definitely predominates by a large margin.
B) The article you describe is an article discussing the options open to the Fed given a goal. Car repair manuals don’t usually discuss why you might want to own a working car. You can hear Krugman et. al. talk about why they care about simulus all the time (in order to cast the opponents of stimulus in a negative light). They talk about the costs of unemployment, people being unemployed even though they would much rather have a job so they can buy food healthcare ect.. Now that may not be as fundamental a discussion as you would like, you might think they are unemployed for a good reason (benefits outweigh costs), but they don’t agree, and they are definitely not thinking that inflation is good for inflation’s sake.
MME’s do not think that 2% inflation is good byitself. They like it because they think price predictability is good and 2% is what we’ve had in the near past hence all that talk by the CB about “price stability”. They think more inflation is good *right now*, but certainly not always. They think 15% is probably always bad because of the 70s where they saw the costs of wasting a lot of time trying to figure out how to hold the least possible amount of cash, constantly adjusting to price changes, having to index all contracts ect.. They also think low trend inflation is bad because it exacerbates nominal stickiness (I don’t necessarily agree).
I find your interpretation absurdly implausible given that MMEs are generally moderately intelligent people.
Anyway, I don’t know that this is a productive part of the discussion. I can’t imagine what makes you interpret MME’s words this way, so I doubt I will convince you it’s wrong.
C) I am not necessarily; it might, I wouldn’t rule it out, but that wasn’t my argument. It’s more like having your options limited because of a shortage of cash specifically (as opposed to assets in general, i.e. monetary disequilibrium) is bad, even when you’re looking for a job (recalculation). Allowing you to trade your assets for cash is a good idea regardless of your job situation. It’s mostly orthogonal.
D) You’re mixing analogies. I was making an analogy between the theory of externalities’ relationship to pollution taxes subsidies and the theory of monetary equilibrium’s relationship to monetary policy. The existence of poor ways of addressing pollution externalities, say dropping all cars into the ocean, does not imply a paradox for the theory of externalities. This is analogous to monetary equilibrium and handing out money to various companies that gave you political support.
Counterfeiting can be good in the same way any theft can be good. If you steal from a a grain producer and sell the products to people who pay more for it more than they would have, that is a benefit to society. If you spend a lot of resources doing the theft, or cause the grain producer to spend a lot trying to stop you or you cause them to go out of business or raise their prices then those are costs and weigh against any benefits theft brought. Theft usually brings low or negative benefits and has high costs, but this is no different from counterfeiting. If counterfeiters literally do the Supplier Of Money’s job better than it does or robbers do a better job than grain producers at selling grain, it’s hardly a paradox to say either is a benefit.
Monetary equilibrium != liquidity. I call Supplier Of Money actions which affect the money supply or demand for money monetary policy.
There’s not necessarily a *net* cost. There could be, but there could just as easily be a net benefit. There’s a cost to the provider of such a contract and a corresponding benefit for the consumer of such a contract, but in the case of money, they are amply compensated for that when money pays no interest (the provider gets the difference between the interest paid by the bond and the interest paid by the bond; the central bank is generally quite profitable). The fact that people want a product (money) and the producer (CB) can make a profit producing it should make not producing it seem slightly ridiculous.
E) I am just not seeing the link; you’re going to have to spell this out for me. You could have a “coordination shock” as you described it inside a firm or inside a single person’s activities, which are not events I would relate to coordination. If the level is the maximum, then that seems like it is just productivity or something similar, and nothing to do with how people coordinate. One can’t add up “non-local global optima changes” and get anything meaningful, as far as I can see. You could add up all the global maxima changes, but that’s more like productivity not “coordination”. I am not seeing a coherent idea.
October 11, 2010 at 11:45 am
jsalvati
A) Notice that your argument here applies equally well to the case where people’s demand does not increase but instead they mindlessly (say the people who’s demand for money increases are not financially sophisticated; you cannot assume away financially unsophisticated investors) decide to buy the same asset that the central bank uses for open market operations. Do you expect the same inefficiency?
If A) is really your major objection to monetary policy, I think we should make a discussion thread about this on the LW discussion section asking people whether they think this is a serious problem with open market operations likely to lead to many bad things. I would honestly be very surprised if many people who consider markets reasonably efficient would say this.
October 12, 2010 at 10:13 am
Silas Barta
Long reply alert, hope it’s worth the delay:
A) If what you’re saying is right, then no amount of action by any CB, counterfeiter, or government could ever cause misallocation or stagnation because “the market would just route around it by taking the opposite actions”, even if the CB/Cf/G kept up these purchases until major macroeconomic variables moved to certain values. Nationalizing any sector of the economy (with compensation) wouldn’t matter, because the economy could just disinvest in whatever the government invested in. In short, socialism works. (Yes, I said it.)
Your follow-up comment is like saying, “You think an economy with a central planner dictating every production process is necessarily inefficient? Well, what if private investors would have done the exact same thing?” Indeed, what if I correctly guessed the road layout of Timbuktu from my basement?
B) Short answer by car analogy: First, car repair manuals recognize that you have the car so you can efficiently drive it *when and where you want to*, and that this metric screens off all other performance metrics having to do with the car’s functionality. It would never tell you to leave the engine running “because that’s what you bought this car for, right? So that fuel would combust and turn an engine? You don’t have to be *moving* for that to happen, just leave it on overnight! What’s the problem?” If it ever told you to do something that uses up resources and doesn’t appear to contribute to your purpose in having the car, that needs an explanation.
Long answer: Let’s say the economists writing them really do have the Level 2 understanding that I accuse them of lacking — they really can justify their policies all the way down to the level of “and so it satisfies human desires better than not doing so”. With that in mind, look at what they are advocating: as people judge that doing X is *not* in their interests — not satisfying their desires — the economist asks for a policy that fundamentally changes their incentives until they *do* deem X within their interests.
Well, hold on! Doesn’t that destroy the link between whether doing X is informative about whether X is welfare-enhancing and thus whether we gain from more X happening? For example, if I donate money to street thugs because I want them to have money, you can infer that this is welfare-enhancing for me. But if I do it simply because the alternative is death (say, I’m being mugged), the relation no longer holds. In causal diagrams, it’s the difference between:
A –> X
and
A –> X <–C
where:
A = Silas deems it an optimal use of resources to donate to street thugs?
X = Silas donates to street thugs?
C = Silas is being mugged?
In the general case of conversions from the first type (X suggests A) to the second (X suggests A or C), you must check, or at least know, the implications of your introduction of C on whether X still suggests A. Indeed, this is how I would characterize the difference between a Level 2 and a Level 1 understanding on the matter: whether you notice when the inference from X back to A is blurred by a second cause C.
So when you break the fundamental assumption (X only suggests A) and make an argument relying on that assumption, then yes, you do need to justify why X should still be taken to imply A (keep in mind it can imply both A and C).
Economists don’t do this, nor, as far as I can tell, recognize that something needs to be justified in the first place! It’s just that the variables have these meanings instead:
A’ = Economic efficiency would increase with more bank loans?
X’ = Banks make more loans?
C’ = Banks’ assets will be raided tomorrow of any unloaned funds?
C’ is an extreme version of what they actually advocate, of course, but the difference is in degree, not type. Like I said in the quote at the top of the page, loans are not good. Good loans are good. Loans that exist because incentives have been artificially manipulated to the point that they merely look good … are not good. And likewise for loans that exist only because of this jawboning.
C) Yes, just like giving me free money from the printing press is a good for me, regardless of my job situation. It’s just that this fundamentally alters the incentive structure of the economy — which reduces to the point I make below in D).
D) Point taken, I did change the analogy (from “they should say [blah] is 100% bad” to “they should say [blah] is partially bad”), so I admit the error and won’t try to rescue it.
Still, you seem to accept an equally important analogy: that large-scale theft will generally not be welfare-echancing — despite it “merely” being a transfer of wealth in the short term — since it fundamentally alters the incentives of those considering whether to produce that which people value; and, you accept that counterfeiting matches theft in a crucial way: namely, that it involves a diversion of wealth to the counterfeiter from those that don’t deem the exchange in their interest (I’m talking about the third parties, not the ones the counterfeiter buys from).
It follows, then, that we should infer a corresponding loss, even before we can precisely identify it, and thus
Monetary equilibrium != liquidity. I call Supplier Of Money actions which affect the money supply or demand for money monetary policy.
This looks like defining away the problem. You originally said the problem was that cash trades are too few, resulting in a premium on cash. That’s not a liquidity problem?
There’s not necessarily a *net* cost. There could be, but there could just as easily be a net benefit. There’s a cost to the provider of such a contract and a corresponding benefit for the consumer of such a contract, but in the case of money, they are amply compensated for that when money pays no interest …
I don’t know what aspect of the scenario that is replying to. The situation I have in mind is one in which I make a loan to a business. One inherent, unavoidable risk of such a loan — reflecting real factors — is that there may be times when *no one* will regard the bond as worth anything, and thus I will be unable to convert it into consumption.
If you deny that it’s necessarily a net cost to guarantee otherwise, then you’re saying I will make the optimal decision *whether or not* I see (have mutual information) with this risk at all, and thus the merit of my judgment is unaffected by any knowledge of any presence of risk.
E) I don’t see what the confusion is, but here goes: “upward” means “better” (KH efficient), and each point on the landscape corresponds to some allocation of resources and production methods. The landscape has mountains scattered across it, and the landscape will change over time reflecting preferences and physical limitations. At any given time, each mountain is a “domain of attraction” because, looking only at immediate neighbors and taking the locally-optimal path will lead the point to the peak of the mountain, with each peak representing a local optimum, some higher than others.
The level of coordination is related to productivity, but not exactly the same thing. If an economy moves from a short mountain to a lower point on a taller mountain, it has increased coordination, but is temporarily at lower productivity (capability of satisfying wants).
The general point is that moving to a higher point (which may be necessary for some people to have a job at all) may involve taking paths that are locally suboptimal and so painful; any subsidy that “reassures” agents that they can just keep making local changes (perhaps back to the c. 2005 optimum) is fundamentally misguided; and that overpaying for the assets of major financial players, who focus on making local changes, will grant them much more power to resist moves out of the local domain of attraction — a domain which involves unemployed people spending their time improving hole digging-and-filling skills to match the great job offers in that sector.
You may disagree with all of that — but does it have obvious incoherency?
October 12, 2010 at 12:22 pm
jsalvati
A) Ignoring the effect of changes in monetary equilibrium (which, as I understand, you are not talking about), to a large degree, yes. You are correct about the scenario you propose, but wrong to say it implies socialism in any important sense. The caveats:
1) Assuming the government buys less than the quantity people want to hold for reasons other than the expected cash flows (maybe they just like the name of the security or they have a family member who works there and it gives them warm fuzzies ect.).
2) If the government buys and holds this may make adverse selection in financial markets a much bigger fraction of trading (informed trading is a bigger fraction of trading than in a different scenario) which might make financial markets be less efficient.
3) Assuming the government does not exercise voting rights. If the government nationalized a particular sector by buying a lot of stock in that sector this would matter because owning stock gives you a right to influence that company’s decisions, not because the government owned the residual income from that company. It would be bad to the extent that the government negatively influences the decisions the company makes (i.e. very). If the government acquired the right to influence the company’s decisions in a similar way but had no right to the residual income from that company it would be equally bad. You should note that bonds do not give control rights, and that the Supplier Of Money could easily adopt the policy of being a passive stock holder that doesn’t exercise its voting rights (most mutual funds don’t).
All of these hold pretty darn well for the sizes of monetary policy actions we are discussing.
It’s important to distinguish between the idea that open market purchases have literally no effect on allocation (not via changes in the equilibrium price level) and the idea that open market purchases have only negligible effects on allocation (not via changes in the equilibrium price level). The former is obviously false, and the latter is very likely to be true.
As for your snark: the CB’s preferred asset is currently government bonds; I do not think it is implausible to suggest that for many people government bonds are the “default asset”. If people are saving by accumulating money (they want money as a form of increasing their saving rather than as a medium of exchange), it is not difficult to imagine that they might instead (in a slightly different world) buy government bonds instead. As a vehicle of savings government bonds and money are often not terrible different.
B) Just to clarify, I am not arguing that all economists or macroeconomists have the best possible understanding of money, or even as good a one as I do. What I am claiming is that macroeconomists are not making the particular mistake you were describing, thinking that levels or changes of nominal variables (spending, inflation etc.) where good by themselves rather than because they had some (possibly poor) model about how those things affect the people’s utility.
I think you make the mistake of not distinguishing between different kinds of mistakes. In particular between the mistake of having a poor causal model and the mistake of valuing surface features for themselves. You can make the former mistake without making the latter mistake. MMEs in general seem unlikely to me to have made the second mistake.
I agree that MMEs talk mostly about surface level features in popular articles, and I agree that this is a mistake, but I don’t agree that MMEs are making the second mistake. Some MMEs make the first mistake and some don’t.
There’s a difference between not typically explaining how one gets from changes in nominal variables to changes in welfare, and not having some model for how to do so (even if it’s a poor model).
Your impression of economists really does sound like a surface impression. As I have said many times, macroeconomists discussions are typically poor, but that doesn’t mean that there isn’t a deeper truth. I didn’t get my explanation for how monetary theory works from nowhere: I learned it from listening to MMEs, and it’s clear to me that even MMEs who I disagree with have *some* kind of model justifying their position, even if I think it’s a bad one.
D1) But you *can* identify the loss; it’s from the central bank (if they’re doing their job in any reasonable sense); the central bank has to remove money from the economy and this cuts into their profits or means they have to tax (if they are a government CB; if they are a private CB this won’t be an option).
D2) I’ve seen a trend in your thinking that monetary policy actions necessarily “dilute” the value of money (presumably they could also “undilute”). However, the monetary policy actions I describe (meaning taken to reduce monetary disequilibrium) are taken precisely avoid changes in the value of money (in terms of goods and for monetary reasons, the price level may still change for changes in productivity).
Let me try to persuade you that analogizing monetary policy to theft is not a useful analogy:
Remember that I explained how you could have private provision of money in pretty much exactly the same way as the government does (with good enough commitment technology). It could do its job equally well if it announces in advance it’s policy (when it expands/contracts the money supply). Now “theft” in this context doesn’t make much sense: people are free to use money or not, and if they don’t, there’s no possible way for the Supplier of Money to take from them. The questions is not whether “theft” is good or not, but whether a policy is good or not. I argue that keeping a stable value so that the value of a unit of currency does not change for monetary reasons is better than other policies. You argue otherwise, and that’s fine, but the answer is independent of whether monetary policy is or is not “like theft” in another context.
D3) “Too few cash trades” is a symptom rather than a cause. It was you who likes to think of it as a premium on cash; I don’t think that’s particularly valuable, but I suggested it because it was less incorrect than talking about liquidity. There’s probably a relationship between liquidity and monetary policy, but it’s not a simple topic, and I haven’t spent a lot of time thinking about the relationship so I can’t say exactly what it is. People confuse themselves a lot because they think liquidity is a simpler concept than it is. Liquidity is a surface, not a scalar and it is both context dependent and instrument specific (MBS liquidity is not the same thing as liquidity of other instruments). I want to discourage you from making this mistake.
E) I understand your idea a bit better now, but I don’t see how it relates to the laymen idea of coordination. Like I said, this seems like it could all take place within a firm or even a single person; can crusoe be more or less coordinated? Coordination seems like it should definitely involve multiple people, especially if it is supposed to be inherently tied up with money which exists because of exchange between people. It seems like “maximum local productivity/utility” or somesuch would be a better label. Can you elaborate? I suspect you of largely relying on the label you’ve given the idea (“coordination”) rather than the description for thinking about how this relates to money.
October 12, 2010 at 1:57 pm
jsalvati
B) An attempt to clarify: A lot of macroeconomic models are superficial in the sense that the way they are tied to the utility of agents in the model is not explicitly incorporated into the mechanics of the model. However, for those same models, there are explanations that tie the workings of the models to the utility of the agents in the models in a reasonable way (monetary-equilibrium like explanations). I would expect basically all MMEs have had some contact with these explanations. Many obviously don’t focus enough on these explanations. However, it’s not the case that MMEs suggestions about monetary policy are not at all grounded in an understanding about how to make the economy work better, and it’s not the case that MME’s understanding about when it’s good (for example) to increase inflation or decrease inflation are ad-hoc. Now it’s alright to say that some class of models is missing some phenomena and that would be more obvious it was missing if they focused more on the microfoundations of their models, or to say that MMEs should do a better job of communicating, but it’s wrong to say that MMEs in general have no understanding about how to tie their recommendations to actually making the economy work better.
October 12, 2010 at 2:10 pm
jsalvati
B) continued from above: As evidence that MMEs know about the microfoundations of macro as I described I present the following short exchange:
“I have been interested in macro micro-foundations for a little while, particularly the “monetary equilibrium” approach. Macroeconomists in general don’t seem to talk about micro-foundations very much. Are there other kinds of micro-foundations? Why don’t macroeconomists talk in terms of micro-foundations more?
Posted by: jsalvati | September 21, 2010 at 09:33 AM
jsalvati: quick and dirty answer: you don’t talk about something when everybody is doing it. You only talk about things that are new, and controversial. There’s a lot more microfoundations of macro today than 40 years ago, so there’s less talk about it.
Posted by: Nick Rowe | September 21, 2010 at 10:29 AM”
http://worthwhile.typepad.com/worthwhile_canadian_initi/2010/09/p-data-q-data-and-l-data.html#comments
Nick Rowe is definitely an MME.
You can say that MMEs (or me) are missing something important in their models, but I do not think you can meaningfully say that MMEs (or me) do not attempt to tie their recommendations to actually making the economy work as well as possible.
October 13, 2010 at 10:05 am
Silas Barta
This is a lot to respond to, so it may take a while to give a full reply. (If there’s something specific you think is characteristic of our general disagreement, please let me know so I can focus on that and answer it earlier.)
For now, I will just pass on the latest exhibit of “my theory implies that counterfeiters can be good, and I won’t consider that a strike against my theory”.
Also, I’m surprised you call Rowe a MME — everything you’ve linked to from him (including his posts on barter in recession from before), I’ve been in near total agreement with! Perhaps building from one of his posts could help close our inferential gap?
October 13, 2010 at 10:07 am
Silas Barta
Note: The link above is also an exhibit of, “It is sufficient justification for my policy to show that it will make people spend more.”
October 13, 2010 at 10:32 am
jsalvati
Rowe is definitely mainstream though I would say a better economist than average. If you read his blog he discusses money in much the same way I do. Reading his posts is a big part of how I learned to think about money in this way.
I think I explained pretty well how counterfeiters can be good in exactly the same way that robbers (from producers) can be good: if they do the job of the producer better than the producer; hardly a paradox.
In the particular case of the “the option value of money” post I think it was misleading for the reasons I explained when were discussing this topic (basically it ignores that other assets also have option value in the same way and to the extent that it option value in a special way this is a misleading signal by the producer of money). I have a half written post on this topic that I will complete sometime.
I will try to think about whether there’s something meta here and post if I come up with anything good.
Yesterday I did think of a good question: Are you at a point where you can say, “I agree with you if you assume x, y, z but I don’t believe x, y , z ” for some x, y and z? For example x = efficient financial markets in the way I have described. Another way to think about this is: do you think you can describe the extent of our disagreement? I’d guess no, but I wanted to check. I cannot.
October 13, 2010 at 10:53 am
jsalvati
Perhaps a better way to think of the counterfeiting thing is like this. Lets consider the case of a movie IP monopolist he owns the all the IP to a particular movie. If he does a poor job of supplying the optimal quantity of movie showings (maybe he’s dumb or maybe he just doesn’t pay enough attention; if he has some special preference for low showings (art film) that doesn’t count), then movie pirates can improve welfare by pirating his movie in the same way that counterfeiters can improve welfare. Both situations happen pretty rarely.
And just to show you that it’s not very useful to use this analogy to reason, I will show you that your theory also implies counterfeiters can be good. Lets say that the Fed liked contracting the money supply (some people like deflation). Then in order to maintain a stable money supply you would need to add back more money to the economy, so you should be saying “my theory implies that counterfeiters can be good, and I consider that a strike against my theory”. Same thing.
There’s an optimal monetary policy (could be keeping the money supply constant), and arguing about whether something is “like counterfeiting” is not going to help you figure out what it is. It’s not useful to focus on surface features here.
October 13, 2010 at 12:18 pm
Silas Barta
Brief reply (I promise to get back to your long reply to my long post and address anything that isn’t obviated in the interim):
1D) My point about counterfeiting never relied on surface features, but the functional equivalence: counterfeiting allows consumption of resources without advance production of resources; this is a net drain on the economy. Even without somehow discounting the gain for the counterfeiter, his actions undermine the very structure that production depends on, leading to a net loss. This has a functional isomorphism to theft, not a mere superficial similarity.
You’re raising the issue that you can look at “knife-edge” cases wherein some “benevolent criminal” can outsmart the economy and do something with the loot that more than pays for its costs — and perhaps even pay those costs back, plus all appropriate premia!
And I agree completely — I just don’t think that some 4th-party planner is capable of consistently identifying these cases, for the same reason a central planner can’t consistently spot them, and I can’t consistently guess the roadmaps of towns I haven’t been to. That — not the theoretical possibility of benevolent criminals — is where I disagree.
2D) I have never claimed that the money supply should be somehow kept constant (as a near-terminal value), so my position doesn’t imply that there’s a situation where counterfeiting would (on average, see 1D above) be good. I do accept that decreases and increases in the money supply can be good, just like the tricycle supply, so long as they’re for the right, not-lost-purposes reasons. For example, when a barter economy transitions to using a money-like instrument, like salt, they are creating money. I find that unproblematic, because as that happens, the economy produces the very same good which money is valued for *as* it produces that “money-premium” on salt.
And likewise, if people transition to a different currency (perhaps because of hyperinflation) — or even to modes of production that obviate money, to greater or lesser degrees — that destroys money, at least by some relevant measures. No objection from me!
Or if gold is the dominant money, and someone mines gold, no problem.
Or for a private central bank, if Wal-mart tries to become one by issuing lots of gift-cards that get traded nearly at par with USD (a situation I find plausible), that’s creating money, and that’s unproblematic (because every such liability is balanced with created value).
But when you create money, without this corresponding value creation process, simply so that people can return to the purchasing and production patterns of that “Golden Age” we apparently just had in the middle 00s, despite their greater poverty … well, then I’m going to have to call “lost purposes” on you.
3D) I will now attempt to state the extent of my agreement and disagreement with you:
I agree that under these conditions:
i) The money issuer is not acting as an arm of governmental policy or any special interest fundamentally in opposition to that which we really mean by a good economy
ii) The money issuer actually follows through with destroying the “extra money”
iii) The economy is not in a severe “lost purposes” trap, in which enormous production structures have been built up around processes that add no value and encourage people to do useless things in expectation of profit,
you have convinced me that printing money and buying liquid financial assets can remove some inconviences at negligible cost (and thus without foisting “lost purposes” onto the economy).
And no, I would not have agreed to even that much before.
Is there anything that I have convinced you of (other than that economists should better articulate an understanding that I don’t believe they have)? Please do not answer “yes” simply out of feeling pressured to “reciprocate”.
October 13, 2010 at 3:03 pm
jsalvati
I think we agree dramatically more than earlier.
1D) The comparison still strikes me as “surface level”. Counterfeiting can be beneficial in the same way that movie pirater’s can be beneficial in the movie example I gave. Saying the Fed printing money is like counterfeiting is saying that the movie monopolist selling more movies is like pirating.
More important to address: you implied that the Fed was like a 4th party. I don’t think this is the case, the Fed is the *producer* of US dollars. US dollars are the Fed’s product, it says so right on the bills “federal reserve note”. US dollars have always come from the US government, and the government has the Fed manage and produce that product. The government also limits competition (as it is won’t to do), and doesn’t make the best possible product, and I agree these things are bad bad.
I want to emphasize that there are many different producers of different monies. Walmart can produce its own currency and the Fed can produce its own different currency. These different currencies can be in different states of monetary equilibrium. You can have monetary equilibrium in one currency and not in another (though if these currencies are substitutes they will have effects on each other).
2D) I am glad you bring up a private Supplier Of Money, such as Walmart gift cards. I see the government central banks as basically doing the job that a private Supplier Of Money would do, albeit not as well. Open market operations are equivalent to creating new gift cards and selling them, just with a different kind of financial asset (some kind of bond vs dollars). I’ve tried to convey this, but perhaps I haven’t done a good job. I will try to expand on this in the future.
Perhaps part of our difference here is that I try hard to look at the government as not fundamentally different from other kinds of institutions. Yes, it’s generally preferable to do things without the threat of violence, but sometimes it’s necessary. The government is part of human institutions just as much as Walmart is, and it’s arrangement desirable or undesirable just as much as Walrmart’s can be. This leads me to see the government as not inherently undesirable, though certainly suspect (not fond of the incentive created by voters; you can see my proposal for an improved voting system listed on the left).
3D)
I am glad I have convinced you of so much.
I agree with i-iii. On iii, I am not totally sure I understand what you mean. I would agree that the Supplier Of Money should not attempt to maintain any ‘lost purposes’. I would not expect that if suddenly the economy had lots of lost purposes that normal monetary policy would encourage these to a significant degree. I am open to being convinced otherwise.
I have found our exchange very very productive as you have forced me to develop and articulate further my understanding of the economics of money starting from the basics of what money is ect. . I wouldn’t dream of making something up. There is at least one thing that you have made me realize (‘convinced’ maybe isn’t the right word): that shifts to from market to non-market activities could be important (though I don’t think this is currently the major problem) will be misinterpreted by spending targets. Price level targets won’t misinterpret such changes, but they also have other problems.
As a result of our exchange, I am writing up an introduction to monetary economics from a Monetary Equilibrium perspective. You’re welcome to help out, ask questions, criticize ect. I will post the first installments later today.
October 13, 2010 at 4:29 pm
Silas Barta
Looking forward to reading that! I will comment on your ideas as usual, and make sure to address anything important here that I haven’t already.
Sorry I didn’t get you a prompt reply to your longer comment from today — or all the other times 😉
Btw, you seem to regard my 3D as a bigger move in your direction that I do — what stands out the most?
October 13, 2010 at 4:43 pm
jsalvati
Part of it was over reading, I reread it and think I understand what you said better now. But I do think this is significant, because before I was confused about why you disagreed with me that monetary disequilibrium was something to be avoided. Now I get to make you try to come up with a different objection.
October 13, 2010 at 9:52 pm
On the option value of cash « Good Morning, Economics
[…] option to do anything. At first, I didn’t see a problem with this, but as part of my ongoing debate with Silas Barta, I’ve thought a bit more about this. I think Rowe’s point is […]
October 22, 2010 at 7:33 pm
Money as a product « Good Morning, Economics
[…] October 22, 2010 in economics | by jsalvati I want to make the case that thinking of money as a product specifically designed to be used as money and produced by a producer is often a useful perspective, and that this perspective remains useful for government created money. From this perspective, the Federal Reserve (a branch of the Federal government) is the producer of US dollars and the Chinese government is the producer of the yuán. This perspective grew out of my ongoing debate with Silas. […]
October 27, 2010 at 1:52 pm
Fake Silas
Cross posted to a more appropriate location:
First, some big-picture stuff: It’s clear from this post that you’re talking about a different topic than what originally got us into our disagreement. You’re basically saying that there are good ways to run a currency that can avoid inefficiencies. However, what originally got me on the topic of criticizing mainstream monetary economists (MMEs) was how they (and you) seemed to believe that this poor money policy the Fed has is preventing a recovery, and during the ’08 crisis it “hurt the economy” — properly defined, of course.
But from these comments, it seems like you diverge from them, in that you’re saying the issue of optimal money issuance is mostly orthogonal to our current economic problems: that yes, having a different monetary policy can improve efficiency; but unlike the MMEs I criticize, you don’t consider it the defining factor of why we can’t get out of this recession; nor do you believe that encouraging people to spend on things they don’t currently consider worth buying to be a good idea.
In contrast, MMEs, when talking at the most lay level, take it as a given that increasing the amount of money being lent or spent (preferably to its c. 2005 levels) is a good thing, something that will get us out of the economy — that changing bank incentives until they regard it as optimal to lend 4% is “better than nothing”. I have yet to see any layman-oriented writing from MMEs that recognizes the changes in any metric are only good to the extent that they reflect a better ability to collectively satisfy our wants.
Unlike you, I don’t think this is a case of “poor communication”. Rather, I think it reflects poor understanding: they have made the metrics an end in themselves, turning the concept of an “economy” into something more like a volcano god. Gene Callahan, someone I normally disagree with strongly, surprised me recently in making essentially the same point.
Long story short, any economy that is “hurt” by “too much saving”, or that requires people to keep spending on the same things and taking out the same level of loans … is not an “economy” I care about.
***
With that in mind, I want to discuss my area of agreement with you, and where I see the value of USD as ultimately coming from.
I find it helpful to think of the Fed as like Nintendo issuing Pokemon cards (or Wizards of the Coast issuing Magic: The Gathering cards, etc.). Ethically, there is nothing wrong with Nintendo having a monopoly on the right to issue Pokemon cards, and they must strike a balance between flooding the market with worthless cards, vs. making them so scarce no one can actually play. In that sense, there is an optimal supply of Pokemon cards.
The difference is in why money is valued. Unlike Pokemon cards, the ultimate valuation doesn’t come from the desire of people to play card games with USD. Rather (I claim), the value of USD comes from this:
1) Past debts have been denominated in USDs; people with such debts have an immediate use to put the USDs toward.
2) USDs can be used to pay taxes.
3) A large enough number of people value USDs for purposes 1) and 2) above so that most Americans are removed a few steps from such a person that they know their dollars can always be traded for real goods.
Now, if the Fed were simply conducting operations to keep USDs in optimal quantity to meet the needs as described above, I wouldn’t have as much objection to its operation.
But that’s not what’s going on. Rather than printing USDs to make sure that they continue to be used for commerce, they find themselves in the position of favoring certain activities over others. For example, when an institution is suffering liquidity problems that ultimate derive from a _real_ shortcoming, and the Fed lends them money on favorable terms, it is propping up an inefficiency, not eliminated.
Or when people currently regard certain production processes or output as crap and not worth buying, and the economy needs to fundamentally retool, the Fed (with support from its court economists) has shown a willingness to make whatever loans are necessary to keep existing businesses involved in these inefficient processes afloat, rather than letting market forces run their course.
Contrary to what you claim, the financial markets don’t simply route around such inefficient purchases — they can’t, as they rely on evaluations from the real economy. Rather, they change the apparent optimality of different processes. If a production process is a really stupid use of resources, then a committment by the Fed to make all the loans necessary to keep it in business will make it appear to be a good use of resources — and other businesses will retool around this inefficient enterprise, workers will optimize stupider and stupider skills to better interface with this process, etc.
I think that by focusing on whether free money is injected into the right places for an optimal monetary policy, you and MMEs are ignoring the effects on real production that this will encourage. I think that these policies keep us in an economic dark age.
Hope this is responsive to the issues you’re raising.
October 27, 2010 at 1:52 pm
jsalvati
art1)
To clarify, I do still think that poor recent monetary policy is a very large contributor to current problems, 60%+. If we had had decent monetary policy, we would have avoided 60%+ of the problems we’ve experienced in the recent past. I am less confident that good monetary policy now could undo as much damage; I find hysteresis plausible. I am not sure what I said that implied otherwise.
The second part of the first section is reasonable summary of our disagreement on MMEs (except the part about “too much saving”).
Part2)
I agree with you on 1-3, I will only add that I think the government requires that people/businesses accept US dollars for payment (“legal tender”) which clearly adds to the value of the dollar (perhaps undesirably).
Lets clarify our point of departure.
FedGoal1) Do you agree that a Fed that just focuses on maintaining monetary equilibrium is a desirable thing? a not terrible thing?
Lets separate out possible Fed actions, and you tell me whether you think they are if you think they are 1) good 2) not terrible 3) terrible.
Action1) Normal monetary policy, defined as buying and selling government debt creating/destroying money?
Action2) Quantitative Easing, defined as buying and selling liquid corporate bonds at their market price? Assuming this was an established procedure (to avoid liquidity concerns).
These next ones I know you object to, and I don’t really have a strong opinion on them, but I do not understand why you think they are as bad as you do.
Action3) TARP style loans to financial institutions – I can understand why you might think that this is a bad idea, moral hazard ect. However, the government made money on these loans (this is my understanding anyway, correct me if I’m wrong), so I have trouble seeing why you think these were a really super terrible idea, even if they are not a great idea. The fact that these loans made money suggests that saving these specific institutions was not a terrible thing, even if the way the government went about is likely to encourage future problems. If a different large but unaffected bank (or perhaps a different government) had made similar loans to these troubled banks because they saw a profit opportunity, would you still say this is contributing to our current problems?
Action4) MBS purchases – The way that MBS purchases can distort the economy is by encouraging excess production of MBS (and whatever bad things this may entail). In order to do this, MBS purchases need to change people’s expectations about cash flows on new MBS. I have a hard time seeing how the government buying some MBS distorts people’s expectations of future cash flows a lot for new MBS, even if they imply they will overpay by some (probably not huge) amount in the case of a crisis. This is especially true given the fact that the Fed is no longer actively purchasing MBS (I think they still hold some though).
Again, I can understand why you think these are a bad idea. What I don’t understand why you think that these activities are not only terrible, but so obviously terrible that not seeing them as terrible policies is a sign of stupidity.
October 27, 2010 at 1:57 pm
Silas Barta
Thanks for carrying this over, and sorry for not doing the cross post myself. I will reply once I’ve had a chance to think this over (there are still things from a few posts earlier in the thread I still need to address).
January 2, 2012 at 8:26 am
Debating Silas Barta « Good Morning, Economics
[…] 7, 2010 in macroeconomics | by jsalvati Silas Barta and I have a long ongoing debate (part 1, 2,3 part 2 actually comes before part 1) about monetary economics, 2008 recession policy and the views […]
December 8, 2013 at 12:29 am
washingtonfreeman
Epic!