Arnold Kling  

Monetary Theory, Once Again

Unchecked and Unbalanced My First Day of Class...

Josh Hendrickson and a reader raise interesting questions.

First, the reader, who asks how my views relate to post-Keynesian economics.

I have been reading your recalculation theory of the crisis with great interest. If I understand it correctly, your theory essentially says that money is endogenous (and hence irrelevant) and taken to its logical conclusion will perhaps entail a belief in the fiscal theory of the price level. Such endogenous-money-real-disequilibria theories have also been proposed by Minsky-style Post-Keynesians, for example Steve Keen (of Debunking Economics and Debt Deflation Watch). The differences as I understand are

1) Both theories agree that risk premia are currently being re-evaluated but while you see this as a structural shift from some industries to others, the post-Keynesians see this as simply the after-effects of an irrationally exuberant debt overload.

2) In your system money arises from the real economy itself (and is hence irrelevant) and recessions are simply structural re-allocations, some of which may involve re-allocations from private debt to risk-less securities. In the post-Keynesian view, money results from debt (and is relevant, but only because debt is relevant) and recessions are the result of over-leveraging and financial instability.

3) You believe recalculations are not to be meddled with because the meddling (monetary or fiscal) can only make it worse without really changing the natural course of events. They believe running fiscal deficits can help 'tide over' recessions and tighter financial regulation can even prevent them.

I would be happy to call my views Austro-Keynesian. I do like the Keynesian (or post-Keynesian) notion that attitudes toward risk follow a cycle: risk averse, then more risk tolerant, then very risk tolerant, then a crash leading to risk aversion. The post-Keynesian financial instability story has merit, in my view. I also am happy to think of markets as having many adjustment problems and other flaws that cause deviations from instantaneous market-clearing. That can be a post-Keynesian view, but it also can be an Austrian view.

Where I part company with the post-Keynesians is when they want to speak of broad aggregates, like consumption and investment. I wonder if trying to think in aggregative terms is not a methodological error.

Instead, I like to think of the economy as if it had an imperfect central planner attempting to solve an incredibly difficult resource allocation problem. In reality, there is no central planner--market prices try to play that role, but they do so much less well than in the neoclassical model. If yesterday's resource allocation was relatively efficient, our central planner (the price system) can do a reasonable job with today's allocation just by making small tweaks to what it did yesterday. However, if yesterday's allocation was way off, or today's conditions are very different, small tweaks of yesterday's plan will not get us to full employment. There may be no way to get to full employment quickly.

Keynesians and post-Keynesians see a recession as an imbalance between aggregate saving and aggregate investment, and government can step in to spend the savings that the private sector will not spend. I doubt that this aggregative approach is the right way to think about the Recalculation problem. I don't think in terms of homogeneous labor and homogeneous capital waiting for someone to come along to deploy arbitrarily. I think in terms of heterogeneous labor and heterogeneous capital, and a lot of the planner's problem is to convert labor and capital to better uses. If the government wants to help with the planning problem, it cannot just blindly run a deficit. It needs to outguess the market about where resources should wind up and how to get them there.

Josh Hendrickson writes

I would argue that it is possible to minimize the role of money and monetary policy without eliminating it from analysis. This, however, puts Arnold's theory in a bit of a bind. If money is included, I fail to see how his theory can be distinguished from the natural rate hypothesis. Couldn't the process of recalculation simply cause the natural rate of unemployment to rise for a period thereby limiting the ability of monetary policy to alleviate unemployment without causing inflation

He may be right that I am taking my anti-monetarism to extremes. I note that in my last post, I dropped a reference to the Nixon re-election monetary expansion, which suggests that I must think that something happened then other than an offsetting decrease in velocity.

However, I am trying to draw a contrast that gets people out of their old habits.

One habit I want to break is the habit of speaking about the natural rate of unemployment. At some level, yes, one can equate the statement "the economy is going through a protracted, complex recalculation" and the statement "the natural rate of unemployment has temporarily risen." Both statements imply that that the supply side has changed. However, in the Recalculation story, the sudden shift in the pattern of demand due primarily to the housing market crash is the supply shock. In the Keynesian tradition (and even moreso in the monetarist tradition), the fall in nominal GDP is a demand shock, and the supply shock is something you may have to tack on later if you need to explain why raising aggregate demand doesn't get you to full employment. In fact, I think there are a lot of Keynesians out there (I have in mind the folks who are arguing that we have not yet had enough stimulus) who don't seem to consider that we've had an increase in the natural rate of unemployment.

Another habit I want to try to break is the habit of thinking that nominal income is proportional to money. At any given moment, one can take the ratio of PY/M and say "there's your proportion for you," but you can do that if you define M as mackerel as easily as if you define M as the monetary base.

I think of real output as determined by how well the central planner (the market) is doing at allocating resources. I think of the average rate of inflation as determined by people's habits under what I call the monetary "regime," meaning the long pattern of government issuance of interest-bearing and non-interest-bearing debt. Most economists want to draw a bright line between those two types of debt, with only the non-interest-bearing debt affecting the price level. I wonder if that distinction is as meaningful in practice as it is in theory. Ironically, the economists that Hendrickson cites as having articulated a "fiscal" theory of inflation are economists of the rational expectations tradition, which I very much reject. I don't think that people base their day-to-day economic decisions on their assessment of intertemporal government budget constraints.

Instead, I would prefer to think in terms of interest-bearing government debt and non-interest-bearing government debt as sufficiently close substitutes that the total matters more than the composition. Swapping like for like is not a significant event.

The 1981-1984 period is embarrassing for my view. Reagan runs a big deficit, and Volcker does not finance much of it with money. We get a recession and disinflation--exactly what would be predicted by those who take the view that money matters. I would back off and say that when inflation gets to be as high as it was in the late 1970's, then interest-bearing debt and non-interest-bearing debt are not close substitutes. The higher the inflation rate, the more the economy will conform to monetarist predictions. Even so, it took over a decade to wring inflationary expectations out of the system.

The Recalculation story suggests that unemployment will remain high.* The fiscal theory of inflation says that inflation could be re-ignited by more issuance of government debt.** Those of you who are too young to remember the "misery index" might prepare to become acquainted with that concept.

*If you have an instinctive belief in a trade-off between inflation and unemployment, then remember that in your world I am positing a large rise in the natural rate of unemployment.

**If you have an instinctive belief in the proportionality of PY to MV, then in your world I am suggesting that an increase in government debt will greatly lower the demand for money, increasing V.

Comments and Sharing

COMMENTS (16 to date)
winterspeak writes:

ARNOLD: Glad to see that you finally mentioned "post keynesian" on your blog. I don't think you characterize that particular school of thought well. Keynesians think the Govt can step in and make up for lack of private demand. Post-Keynesians understand that Govt deficits fund private savings (which they must, by accounting) and are quite happy to have this happen through tax cuts, which also avoids all that terrible Industrial Policy that Keynesians are so fond of (and you, rightly, recoil from).

This is unsurprising when you remember Keynes came of age at a time when taxes were low, the Federal Govt was small, and the US ran on a gold standard, while the Post-Keynesians are talking about a time when taxes are high, the Federal Govt is large, and the US is on a fiat currency (although Hummel would disagree and beleives the US default is "likely" when, in fact, it is impossible).

Post-Keynesians also understand how the monetary system works, and therefore know that the "money multiplier", which is critical to monetarists like Sumner, does not exist. Reality is currently showing how much it does not exist either, but if you can't ignore reality in the Ivory Tower, where can you escape it's clammy grasp?

Right now the market wants to save more, but the monopoly supplier of savings is not producing them in sufficient quantity to meet demand. Post-Keynesians would say that the monopoly supplier should find this demand by simply taking away less private sector income (lower taxes). Keynesians would say the Govt should pick industrial winners and losers (higher spending). Austrians would say that collapsing private spending and deflation is purging the rot out of the system, and it's good to get rid of these bad debts (and we should get rid of the central bank while we are at it). Recalculationists would say that it's all very complicated, and you cannot run a deficit/fund private savings through lower taxes in a useful way (???)

Ritwik writes:

So, you're saying that deficit spending *could* boost an economy in a recession/recalculation as long as the government chooses the correct sectors? In that case, the debate reduces to the socialist calculation debate. In fact, since the government only needs to intervene a few times rather than always, it reduces to a watered down version of the socialist calculation debate in which the onerous burden of proof is on the other side - a 'free-market usually, government in recessions' view seems remarkably attractive.

Given that many Keynesians (of all hues, ever since Samuelson admitted he was wrong in the capital controversy) will happily admit heterogenous capital and labour, I think your main departure with post-Keynesians is still on the issue of money (my point 2). You believe money is neutral, but are not led towards the Lucas path because yours is fundamentally a disequilibrium analysis. They believe money is non-neutral but administered monetary policy is ineffective because money is the result of private debt.

It is interesting that you mention that the economy looks monetarist when interest rates are suitably high and non-monetarist otherwise. This is entirely the same as saying that the LM curve is more horizontal at low interest rates and more vertical when interest rates are high - something that all Neo/New Keynesians will identify with. In your view, interest rates are *usually* low enough to keep the LM curve horizontal. In their view, interest rates are *usually* high enough to keep the LM curve vertical.

It is also interesting to combine your two starred comments. Let's say I believe in both the Philips curve trade-off and the MV=PY theory. Then, your argument about the unresponsiveness of PY(or equivalently, a shift in the NAIRU)implies either that M should go down during deficit spending or that inflationary expectations adjust quickly enough during a fiscal stimulus. The former seems like a monetary policy choice - the fed could print more money that ultimately finances the deficit. The latter does not sit well with your idea of sticky expectations - why are they sticky only in a monetary expansion and not in a fiscal one? On the issues mentioned in the last two paragraphs, I think your main contentions are best challenged by the work of Willem Buiter. (

In all of this, I see elements of a new synthesis. One could proceed from the neo-Austrian/post-Keynesian view and give exogenous monetary policy a second chance. Or one could begin with monetary disequilibrium and try to move away from an 'only money' perspective which tries to argue that the velocity of money just collapsed one fine day (and is going to remain there until the next such shock comes along). Alternatively, one could begin from the mainstream Neo-Keynesian/New-Keynesian view, drop rational expectations, adopt adaptive sticky expectations, add huge dollops of disequilibrium (both monetary and real) and include asset price dynamics. This I believe will be the most fruitful research program, unless it is hijacked again.

For theory, we could turn to Phelps, Leijonhufvud, Fischer Black and Minsky. For this crisis, Buiter seems most reasonable.

bakho writes:

It is interesting that you bring up the 1981-1984 period.

Volcker and Reagan spending are selective parts of the story but one that is missing major pieces.

As a result of the oil shocks of the 1970s, businesses transformed their energy utilization. Efficiency greatly increased as did ability to switch to the least expensive fuel. This investment in energy efficiency (you can think of it as increases in energy productivity) took investment away from other potential investments in productivity. During this time oil consumption dropped by 22 percent. CAFE standards kicked in. The price of oil, (which affects prices of many other products) deflated. By 1981-2, much of the energy transformation had already been put in place, so investment was freed for other areas.

If you consider the entirety of events, many of the events from 1981-4 align with your economic dislocation story.

Not everything has to boil down to a single cause and effect. Are there economic situations where monetary policy might dominate and other economic situations where dislocations become predominant?

fundamentalist writes:

Arnold: "He may be right that I am taking my anti-monetarism to extremes....However, I am trying to draw a contrast that gets people out of their old habits."

Exactly! When people are stuck in a paradigm, sometimes you have to go to extremes to crack the shell and get them to see something new. It's good pedagogy.

All Arnold needs to become fully Austrian is to see the damaging effect of inflationary monetary policy in causing the misallocations during the boom. They don't happen by accident. Prices guide the market. Monetary pumping distorts prices and causes disequilibrium. However, that doesn't mean that monetary policy can fix the problem as monetarists claim. Once monetary policy causes misallocations of labor and capital, it becomes a real economy problem. The market must recalculate and re-allocate labor and capital. Yes, in the short-run monetary pumping can boost employment temporarily and to a small degree, but the danger is that it will set in motion the next misallocation of resources, and that is usually what happens.

bakho: "Are there economic situations where monetary policy might dominate and other economic situations where dislocations become predominant?"

That is Austrian econ in a nutshell. Very small nutshell. Monetary policy dominates during the boom, the real economy causes and dominates the bust. I like to say that monetary pumping inflates bubbles, the invisible hand pops them.

winterspeak: "Austrians would say that collapsing private spending and deflation is purging the rot out of the system, and it's good to get rid of these bad debts..."

Austrians wouldn't say that there is anything good in collapsing spending and deflation. Austrians say the bad things, the misallocation of labor and capital, take place during the boom, the part that every other school of economics claims is totally virtuous and should be maintained. The bust that follows is necessary, but not good. It corrects the mistakes made during the boom but is very painful. Trying to stop that re-allocation of resources only prolongs the pain. Austrians simply don't want to prolong the pain.

winterspeak writes:

FUNDAMENTALIST: I cannot think of a school of thought more anti-Post-Keynesian than Austrianism. Austrians want to go back to a gold standard regime. Post-Keynesians embrace fiat currency.

In your own response you say that "The bust that follows is necessary... It corrects the mistakes made during the boom" PKs would say that bad decisions are in the past, and destroying real output (through unemployment) does not make them better. The Govt should do it's job as monopoly issuer of fiat currency, fund the private sector's demand to save (which will cause no inflation) and do so in a way that lets labor and capital be allocated correctly.

A bust in housing that has car sales drop 2-3x below replacement level is *not* allocating anything correctly. A 10% unemployment rate, like wise, is not allocating things particularly well either.

Please tell me how a payroll tax holiday would create misallocation of resources. Thanks!

pcemberton writes:

Arnold, the model you are looking for has already been written down. It is "Bottlenecks and the Phillips curve: A Disaggregated Keynesian Approach" published by George Evans in the Economic Journal, 1985 (!!)
Now building a more sophisticated financial sector into Evans model, maybe integrating with stock-flow consistent models like Tobin and Brainard and the more recent Godley/Lavoie model could be the means to integrate Minsky and your Recalculation.

fundamentalist writes:

Winterspeak, not all Austrians want a return to the gold standard. Hayek didn't think it would do much good.

Winterspeak: "PKs would say that bad decisions are in the past, and destroying real output (through unemployment) does not make them better."

The wealth was destroyed when it was misallocated in the boom. The Depression is nothing but the manifestation of that wealth destruction. Unemployment is an inevitable consequence of that wealth destruction.

Winterspeak: "The Govt should do it's job as monopoly issuer of fiat currency, fund the private sector's demand to save (which will cause no inflation) and do so in a way that lets labor and capital be allocated correctly."

There is a huge gap between what PK's want and what they can actually do. They need a little dose of humility. Show me one depression in which that nonsense worked and I'll give it some more thought.

winterspeak: "A bust in housing that has car sales drop 2-3x below replacement level is *not* allocating anything correctly."

All you're saying is that you're smarter than the market, which is the hubris all socialists are guilty of. Besides, no one ever claimed that the market would re-allocate instantaneously. That happens only in the general equilibrium math models. In the real world, re-allocation takes time; the greater the misallocation during the boom, the longer re-allocation takes.

winterspeak: "Please tell me how a payroll tax holiday would create misallocation of resources."

I have no idea. Who claimed it would?

winterspeak writes:

FUNDAMENTALIST: Forgive me if I said all Austrians wanted to go back to a gold standard, and some did not. The overlap between Austrians and gold-standard advocates is big, but it may not be 1:1.

Unemployment (much beyond the standard frictional level) is *not* an inevitable consequence of wealth destruction. Bad decisions were made in the past, why can't everyone go to work implementing better decisions today? PKs have some far out views on utilizing labor that I don't agree with, but past malinvestment need not require current unemployment. There are shades of sunk cost fallacy in this assertion.

I laughed when you said PKs have power. PKs cannot do anything! They are even more marginalized than the Austrians ; ) Scott Sumner is about 100x more famous than the most famous PK. Intellectually, they remain so far on the periphery that they might have fallen off the surface of the planet all together. There are lots of examples where big deficits pulled an economy out of a recession, and one excellent example of deficit reduction plunging an economy back into a depression. I am sure you are familiar with these.

Fundamentalist, you seem to deny that nominal currency is also part of a market economy. Greater private sector savings is part of re-allocation, and it CANNOT HAPPEN unless the Government does its job and funds that.

There is a market failure here, and the market failure is that the monopoly manufacturer of net private sector savings is setting output below the optimal level to maximize real private sector wealth in the short and long terms.

It is delusional to claim there is no monopoly manufacturer of net private sector savings. It is blind to not see that this is what is limiting employment today. It is anti-market to keep it from doing its job, especially as it can do it without interfering with any re-allocation or re-calculation. And it is ignorance of double entry book keeping that keeps alive fairy tales like "banks lend out deposits", "the US Govt will default on its debt", "Government deficits are not necessary for the private sector to save more" etc.

fundamentalist writes:

winterspeak: "why can't everyone go to work implementing better decisions today?"

In a primitive economy that would happen. But as Hayek points out, workers require capital. A great deal of capital is destroyed in artificial booms. It takes a while to build the savings to supply the loans to build new capital for workers in different industries.

winterspeak: "Greater private sector savings is part of re-allocation, and it CANNOT HAPPEN unless the Government does its job and funds that."

Au contraire! Real savings is nothing but reduced consumption. No new paper money is necessary. In fact, it's harmful. As prices fall, the money that people have in the bank already becomes worth more and reduces the need for more money. The key to recovery is falling prices.

winterspeak: "It is delusional to claim there is no monopoly manufacturer of net private sector savings."

I guess I'm delusional then. Although I'm not quite sure what you're saying. Is the state the "monopoly manufacturer of savings" in its role as printer of paper money? If so, how did people save under a gold standard? Are you saying that saving is impossible unless the state cranks up the printing presses and floods the world with money? Is saving totally impossible in a barter economy? In other words, if I live in a barter economy, is it impossible to save potatoes by hoarding them in a root cellar?

winterspeak writes:

FUNDAMENTALIST: Please explain to me how having old capital destroyed means that I cannot work on something productive if given the chance. My house burns down -- I'm still able to work.

Savings DO NOT produce loans. It is the opposite, bank LOANS create the DEPOSITS. Think through the accounting and you will see why this is the case.

It is a fact of accounting that it is impossible for one sector to increase its net savings of financial assets unless another sector dissaves by that exact same amount. This is double entry bookkeeping, and noticing how a bank loan creates both a liability (deposit) and a receivable (loan), triggered by the loan, will help you see the difference between currency issuance and currency usage.

If prices fall, income falls too. The private sector cannot increase net savings. Your "liquidationist" approach is vandalism at a sector level.

If I grow a potato and put it in my cellar, there is no liability connected with that asset. But if I take a $ of earning and put it in my bank, there is a $ I am taking out of someone elses income and there is a liability associated with the asset.

Once we came off the gold standard, money became unlike potatoes and like numbers in a spreadsheet. It is this vestigial belief, that money is still like potatoes, that creates the fairy tales like "US will default on its debt".

Excellent discussion of this here:

winterspeak writes:

FUNDAMENTALIST: If you prefer, we can continue to run with your potato analogy.

Suppose you have a mortgage you need to pay, and the mortgage is denominated with potatoes. The mortgage is too high, as are the mortgages of all of your neighbors, because there was a housing bubble, which has now popped.

Unfortunately, your land cannot grow potatoes. Neither can any of the land of any of your neighbors. Everybody wants more potatoes, but no one can grow any.

There is one source of potatoes, the Govt potato factory, and they refuse to issue any. "It won't help," they cry "go recalculate amongst yourselves."

As you wait for prices to fall, which means you are more underwater on your mortgage and your salary will be cut as your company's revenue falls, how do you, and your neighbors, get more potatoes collectively?

Looking forward to your solution!

fundamentalist writes:

Obviously there is no solution. You have set up your analogy so that no solution is possible. I suggest the analogy is faulty. You have arranged the analogy so that potatoes take on the role of money. I was talking about a barter economy in which there is no money and people exchange what they produce. If you produce potatoes, you trade potatoes. If you produce cows, you trade cows for whatever else you need.

fundamentalist writes:

Also, if you save, you save what you produce. If you produce potatoes, and you want to save, you save potatoes. But if you produce shoes, you save shoes. Saving is nothing but reducing your consumption today in order to have something to consume in the future.

winterspeak writes:

FUNDAMENTALIST: There is a solution. The Govt potato factory starts making some potatoes. Or, if the Govt is in the business of taking away peoples potatoes, they could stop doing that for a while.

I think you are very close to understanding my point though: in a gold standard world, money is like a potato in that it is just another element to be bartered, where it's value lies in convenience, etc.

But we are no longer in a gold standard world, we are in a fiat currency world, where money is a unit of account created by the state. Therefore, the state should take responsibility and manage its production responsibly: cut back when it makes too much (inflation), but ramp up when it makes too little (deflation). Right now, the private sector, at a sector level, wants more net savings, but the US Govt is not doing that sufficiently, and there are lots of people opposed to it doing so ("go recalculate amongst yourselves!") The way the state transfers that unit of account to the non-state sector is by spending it, and then not taxing it back. This is called running a deficit, but is properly understood as funding private sector savings desire.

fundamentalist writes:

winterspeak: "Therefore, the state should take responsibility and manage its production responsibly: cut back when it makes too much (inflation), but ramp up when it makes too little (deflation)."

That's easier said than done. Milton Friedman was right that the lead times are too short and the lag times to long. As I have posted many times, the lag between Fed policy implementation and its effect on prices can be 4.5 years. There is simply no way the Feds can do what you suggest. They have been trying for 100 years and failing miserably every single time.

But my point about the barter economy is that saving is possible without money because saving is nothing but reduced consumption. That has been the definition of savings in all economics from before Adam Smith. All money does is make saving more convenient. It has nothing whatsoever to do with the amount of savings. People can save even when the money stock is falling. In fact, they usually save more then.

winterspeak writes:

Fundamentalist: It is not hard. If you declare a payroll tax holiday, you will start injecting $2B into the economy at a rate of $38M/week. This would put more money that the entire Obama stimulus into the economy in one year than his does over its 5 year span.

All of mine would go to households, who need money, not banks, who do not. It would instantly flow into debt repayment (helping banks), savings, and spending.

Friedman was against the Obama style fiscal stimulus where Govt picks winners and losers and dishes out favors. In our high-tax society, the Govt can enable the private sector to recapitalize itself by not taking away its money for a few months.

And when employment goes up/inflation increases, declare the holiday over.

It is not that hard.

In a barter economy there is no money to save. And since we aren't in one, I don't see why it's pertinent.

Comments for this entry have been closed
Return to top