February 2021

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
Blog powered by Typepad

« This Week's "The Calling" | Main | The False Dichotomy of Rothbardian Anarchism and Hayekian Classical Liberalism »


Feed You can follow this conversation by subscribing to the comment feed for this post.

I could not agree more.

Fritz Machlup cystalized ABCT when he said that monetary factors cause cycles, while real factors constitute it. The monetary disturbance is viewed as a macro cause of micro factors. Even the monetary disturbance reduces to micro categories of demand and supply.

Pete mentions Alchian. When I went to UCLA graduate school, it was under Alchian's strong influence. Macro issues were analyzed in micro terms. Leijonhufvud's macro fit neatly into the Alchian paradigm. Most professors were open to Austrian variants of the paradigm.

Current "fashion" in macro is to have models in which there is a unique interest rate, "the" interest rate. No coherent analysis of the financial crisis, nor of the way out, can be made using such models. There are a multiplicity of rates, whose interaction generated the recent cycle.

Current policy is being constucted using such models. Policy is almost literally coming off classroom blacboards with no attention to the complex reality of actual financial markets. There is no end of harm that can come out of such a procedure. If you want a definition of extreme apriorism, look at fashionable macroeconomic thinking.

that was the smartest thing on macroeconomics that i have read since the lucas critique.


I am also unhappy with the state of macro. I share your distaste what what Buiter calls economics of Dr. Pangloss. But you build up a strawman. I do not see the relation between minimum wage proponents and mainstream macroeconomics. Thoma is not the representative macroeconomist. Like most other economists, macro-guys do not favor minimum wages (even though a surprisingly large fraction of economists in general supports this fallacy). And even though growth of knowledge is not linear, it may be equally wrong to assume that it is not growing at all. Macro is simply difficult and policy advice always in high demand. I speculate that Hayek would be more comfortable with modern macroeconomists than he was with his contemporaries. And I am not convinced that you find in Smith, Ricardo, Mises, etc. any insight into expectations formation and its significance for monetary policy. Not everything can be found in Adam Smith.


also in ABCT there is only "the" rate of (natural) interest! That was part of Sraffa's critique. Modern theory at least acknowledges that outside (or rather before) stationarity, there are plenty of (own) rates of interest. Or do you mean money vs. natural rate of interest? The natural rate, however, is today introduced as interal rate of return of the representative firm (ugh). I have to admit however that workhorse models usually ignore investment. Bad enough.


Sraffa was wrong.

In Prices and Production, Hayek juxtaposed the bank rate against the natural rate. Later elaborations imply as many expected rates of return as there are firms producing in different stages of production.

I agree here: Hayek's Ricardo effect, inclusive is elaborations on the upward slopping saving curve, displays divergent equilibrium rates of return (Kaldor attacked this point). I mention the Ricardo effect because it is much better theory than P&P. I disagree here: Sraffa is not wrong: outside stationary or steady states, own rates of interest differ (I do not mean that Sraffa is correct in general). Hayek acknowledged this and it is core of neoclassical capital theory a la Bliss. That modern macro does not consider Wicksellian interest spreads is not true, see Woodford, even though I admit that such spreads are restated as equilibrium phenomena, thereby missing Wicksell's and the Austrian disequilibrium-coordination nexus. What is also true, however, is that in modern theory spreads are explained by a much better interest theory and with better tools. Going back to the Austrians may not be useful(except for Hayek, who however accepted Fisher's theory). I wish Austrians would rely on Hayek's Pure Theory of Capital (very Walrasian), rather than on his Prices and Production (Böhm-Bawerkian).

To amv: you raise too many issues to respond in detail in this format.

The macroeconomic way of thinking really is an alternative and not a supplement to the micro way of thinking. It may appear as if the boundary is because of the different issues studied. In reality, however, the divide is more fundamental than that. This can be seen in, for example, Paul Krugman's continual attempts to show how macroeconomic imbalance (eg, recessions) invalidates the micro ways of thinking. So prices, wages, interest rates don't adjust their respective markets because of liquidity traps, cumulative processes of price decline, etc.


This was Samuelson's basic point in Economics as well. Unless the macroeconomic system is in balance, the microeconomic laws of supply and demand do not hold.

And the puzzle that enlightened policy was to solve was how can we have poverty admist plenty?


Mario, Peter,

I really hope that's a straw man, because otherwise ... wow.

If your macroeconomics appears to invalidate your microeconomics (under any circumstances), then one, the other, or both are wrong.


Careful --- only invalidates when the macroeconomic system is out of balance. If you get the macroeconomic system in balance, then microeconomic analysis can proceed.

Read Mario's comment a little closer, and read the first edition of Samuelson's Economics to get a very clear picture of the basic idea. Then go and read Krugman or Brad deLong and it will make perfect sense.

To AMV -- follow the debate on minimum wage --- it is Krugman that Thoma is picking up on. And over at Marginal Revolution Tyler and Alex have also contributed.

Recently I started reading something about monetary policy in Japan and studied some neokeynesian models of the liquidity trap, etc. I'm appalled by these sorcerer's apprentices.

I hope I'm misreading them in saying that their approach to monetary policy analysis is: (1) work out a model with a couple of frictions chosen to obtain some good fitting of time series data, (2) compute the model stationary equilibrium, (3) linearize around this equilibrium, (4) fit the model on time series data, (5) assume parameters are stationary, (6) analyze policy issues on the model, (7) apply these policies to the real world, (8) when all goes wrong, blame the market.

I really hope I'm wrong. Maybe academic analyses are made in this way, but guys at the Fed are not so crazy (that would be a surprise). Maybe some of my points can be easily overcome or are inconsequential. I didn't work out the details of their models nor I have a fairly broad knowledge of the related literature (aggregators, imperfect competition, habit persistence...), nor I know the details of data fitting procedures.

However, as an engineer, it appears to me that the approach is just like: design a rocket assuming there is no atmosphere, launch the rocket, blame the air.

The greatest problem with actual macro policy is not that it is not microfounded: all recent models are based on some sort of apriori (in the bad sense: uncritical) microeconomic foundation.

The problem is that the NUKE camp acts as a crazy engineer and causes disasters with their oversimplified view of policy analysis, and the RBC (broad sense: all "free market walrasian guys") camp don't have any reason to supppose that the other camp chooses disastrous means.

As long as the latter camp will keep on insisting that money is neutral and the fiscal multiplier is zero dot something, it will not be able to understand the real subtle costs of the policies of the sorcerer's apprentices. The problem with policies is not their irrelevance, and lack of neutrality is not a market failure to be corrected, but a backdoor to interventionist policies which may be disastrous.


well Krugman ... is he representative for the macro tribe? I don't know. What I do know is that already classical economist like Ricardo und the Bullionist understood that changes in nominal wages alter real wages and therefore labor demand iff aggregate demand (or effective money supply) remains constant. If aggregate demand falls, falling nominal wages leave the real wage unaltered - since all nominal values decrease to assure long-run neutrality. Don't get me wrong: I do not favor minimum wages. For macro-theory also tells us that nominal rigidities leads to falling real output whenever expenditures falls short. This is also Austrian, isn't it? Otherwise: why did Hayek favored constant Mv after realizing that deflationary pressures do not break nominal rigidities? I argue that (1) we need macro-theory as a complement to micro-theory (2) macro-theory is very demanding and thus less developed than micro-theory, and that (3) ignoring macro leads to systematically false predictions, e.g., that falling nominal values like wages lead to more employment in face of negative demand shocks. Further, we know today that it is impossible to derive the behaviour of the economic system from pure micro, that is, by the fact that individuals act, that they choose rationally, etc. Austrians HOPE that it is the profit and loss system that keeps the economy on track. I BELIEVE this too. But since today, we do not have a good theory for that. All Kirznerian alterness cannot ensure that aggregate excess demand functions are well-behaved, that is downward slopping. Implicit in all Austrian theory is that there is gross substitutionality .... but how do they know?



I was careful. I said "under any circumstances." If your microeconomics is invalidated when the macroeconomy is "out of balance," then one, the other, or both need fixing.

What a terrific summary. Thinking politically, as I often do, I wish right thinking politicians would make this a foundational statement of their economic policies.

Tyler is scaring me over there at MR these days--his replies to Krugman do not seem to get any the really basic flaws in Krugman's analysis-- maybe this was on purpose, reply to a Keynesian on his own terms-- but the discussion of the minimum wage should not be framed by Keynes, it is ridiculous. Consider Tyler's response to this paragraph:

"1. Why did I go from minimum wages to overall wages? Clearly, a cut in minimum wages –which only apply to some workers — can raise the employment of those workers at the expense of other workers. But the advocates of a cut are claiming that they can raise overall employment. The only way that can happen is if a reduction in average wages raises employment."

Tyler responded by talking about wage stickiness and AD and AS... My response is much more basic: Did an economist really say this?? The minimum wage is a price control. As with the price control on any input to firms, relaxing the control should allow the firms to become more productive - because they can choose a more efficient combination of inputs.

Firms freed of the minimum wage can now adjust their labor to capital ratio according to what is actually most efficient rather than basing it on arbitrary prices set centrally. This greater efficiency made possible by hiring more low-wage workers in firms where this makes sense should most certainly increase overall employment because it should produce growth which will produce more hiring.

This is how macroeconomics is based on microeconomics (and there are some great lessons from the Soviet experience here), but Krugman and Tyler are instead bouncing back and forth ideas about aggregate pressures and stickiness and focusing solely on how to make the numbers grow, forgetting all about efficiency...Is this what we call economics today?

Follow up quote to my comment by J. Viner (also Tyler on MR):

"And what Keynes had to say then is as valid as ever: under depression-type conditions, with short-term interest rates near zero, there’s no reason to think that lower wages for all workers — as opposed to lower wages for a particular group of workers — would lead to higher employment.

Suppose that wages across the US economy had been, say, 20 percent lower than they actually were. You might be tempted to say that this would make hiring workers more attractive. But to a first approximation, prices would also have been 20 percent lower — so the real wage would not have been reduced. So how would lower wages lead to higher demand for labor?

Well, the real money supply would have been larger — but the normal channel through which this might increase demand, lower interest rates, was blocked by the zero lower bound. Yes, there would have been a slight Pigou effect: real private sector wealth would have been higher, because cash under the mattress (or wherever) was worth more. But on the other hand, real debt burdens would also have been higher, probably exerting a contractionary effect. Overall, there’s no good reason to think that lower wages would have helped raise employment."

Bad for the brain?


Maybe it's an immanent criticism (IC), as opposed to transcendent criticism (TC). ICs assumes the hypotheses of the rival theory in order to prove that rival theses do not follow. TCs assumes other hypotheses.

An IC is possible only if the rival thesis is really contradictory in the rival framework, which is not always the case.

It's a good thing to focus on ICs rather than TCs: ICs assume less, are more comprehensible, are more convincing (if successful), and force the critic to enter the details of the rival theory, making dialogue easier, and work even if the object of criticism has no knowledge of the other (transcendent) theoretical framework.

A possible objection is: is it worth the effort to argue with Krugman? :-)

PS For instance, a paper called "A rational expectations model of ABCT" would be an IC. I don't know if it can be conceived, but to say that RE don't hold is a TC, and may not convince those who believe in them. I fear that such a paper would fall under friendly fire.


Note that in my post I deny that this crisis is a crisis caused by collapse of aggregate demand, but instead a crisis caused by regime uncertainty.

I read all the points you raise and cite others as raising as speculative answers to a situation, I want us to challenge the way we think about those answers and the way we frame the situation.

If I live in a Keynesian world, then I guess Keynesian solutions would be the way to go. But I deny that we live in a Keynesian world of wilds swings of optimism and pessimism which means markets are inherently unstable. Instead, I remain more on the side of Say in his letters with Malthus.

Bottom line: the problem here is not the markets inability to adjust to changing circumstances, but government policies which raise the cost of adjustment. Government policies have turned a market correction into an economy wide crisis. And you cannot get out of that crisis by government continuing to do the same things which led to the crisis.

So yes, in my opinion, still bad for the brain, and certainly bad for the economy. I don't deny this makes me out of step, I do deny, however, that my position is absent grounding in theory and/or evidence. But even note how in your statements you invoke aggregate variables in the discussion, I am denying that these aggregates help us understand much of anything in an economy. It is not really microfoundations of macroeconomics, but microeconomics of issues such as fiscal policy, monetary policy, regulation, trade, etc. Economics is, as I said, about property rights, relative prices, and profit/loss.


I'm in a horrible position. I share 99% your sentiments. Government (excl. Fed) made things a lot worse sure. Regime uncertainty? Granted. Raising wages? Bad idea. Only now, after the Fed has let NDGP slump by insufficient reaction and bad prior regime (no level targeting to insure against such demand shocks) I find myself in a position to promote policies that you call Keynesian but which I detect in almost all classical writings, in the writings of Friedman and early Chicago economists like Knight and Viner, even in Austrian literature on "secondary deflation" (e.g. Haberler), etc. Just think of Sumner, a Chicago Boy, now taking sides with monetary dowes. The only other episode in history, where I would equally favor more demand is the Great Depression.

Yes, I do speak in terms of aggregates. I am aware of the pitfalls. But to be consistent you have to bury ABCT, which is as aggregate as anything can get (despite the lip service to methodological individualism).


I am on record as in doubt of Sumner's position. My main reasons for objection relate not just to the problems I have with macroeconomics, but also to the public choice implications. I am concerned with the "end game" politics as well as the incentives that policy makers face in pursuing optimal policy even assuming that they have the knowledge and ability to pursue monetary policy.



don't you think that "public choice implications" are worse now, after the recession takes full course because of bad monetary policy. Just imagine the Fed would have followed a NGDP level target and thus nominal spending expecations were anchored. No need for fiscal stimuli, no need for regulation of financial markets, no European turn of American policies, etc... Leaving money supply contract caused the Great Depression and thus allowed for the worse kind of "public choice implication". I am not only talking about Roosevelt and the New Deal. Just think of Hitler and WWII with 55 million bodycount.

I'm unconvinced by all this NGDP story, and by unfounded comparisons with the Great Depression.

It's not the first time I post on this, but it's the first time I make up my mind. I've been inspired by AMV's comments, but I had thought similar things, unsystematically, when reading Sumner, Woolsey and Horwitz (whose position is intermediate, I think, but I don't have it clear).

I tried to explain all my doubts with the monetary disequilibrium idea of keeping MV from falling. Despite the length, the main idea is that if the banking sector is overstretched, the new equilibrium will imply a lower NGDP, and any attempt to impede this painful process will just keep malinvestment where it is, prolonging the recession.

These are my thoughts, maybe occasionally too clearcut not to be annoying (that's a convoluted way to ask sorry). Sorry for the length, I hope I touched something important.

(Irrelevant, unrespectul but funny paragraph, I leave it because the rest is boring) If a drunk person has headache, I don't say "you must have fallen and smashed you head on the ground", I say "you have a hangover". That's not to claim that drunk people don't usually fall to the ground or that to hit something hard with the head is not painful, it is simply of little relevance.

Now, the US economy has been having a hangover. A huge one. Everything is or has recently been excessive, and not for six months, but for one or two decades: consumption, leverage, debt, trade deficit, maturity mismatch, bad or complex or opaque lending...

This hangover cannot be painless. It must last for several years, because nothing can turn a 100% household debt / GDP ratio or a 80% consumption / GDP ratio, or millions of now submarginal construction workers and capital goods into something that makes economic sense without a huge headache. Probably, the mere increase of the natural (not actual) rate of unemployment due to such massive structural shifts (probably, banking, finance, construction and import industries, not to forget automotive, suffer large overcapacity) may account for some percentage point of unemployment for several years.

(Ok, it may be more human not to say to a laid-off worker "you don't make sense from an economic point of view", but it's clearer than "structural unemployment" or "submarginal specific human capital in need of complementary investments for reallocation".)

Even in the unlikely case that reflating the economy does not prolong the binge that has been cutting the roots of the US economy for at least a decade, it has little to do with the actual pains in the capital, financial and job markets.

The present stance of Fed policy is massively expansionary: the fact that it fails to make monetary aggregates explode is the result of real structural problems in the financial and banking sector, that can be cured either by bankrupties or by plundering tax-payers, but surely not by cutting rates that have already hit the floor.

If the economy overshoots its sustainable debt levels, only repaying that debt can put the economy again on the right track. The same applies to all overstretched positions: financial leverage, maturity mismatch, reliance on foreign credit, the relative weight of consumption on total production, and monetary multipliers. If the "equilibrium" level of the deposit/reserve ratio is, let say, 10, but its actual level is 20, there is nothing but bank credit deflation that can bring the economy back to its sustainable path without reflating bubbles. The only alternative to keep M from falling (and the same applies for all other M's or for all the MV's) is to double the monetary base, and this hasn't helped for the previously discussed reason.

Now, banks are crumbling, and they would have failed en masse without taxpayers involuntary help. Non-bank financial institutions are largely in the same conditions. The whole sector has accumulated sufficient debts and risk factors to create a black hole in the heart of Manhattan. How can they be forced to keep MV from falling, or even to rise at an early rate of 2-3-4%?

The money supply doesn't exist: it is an aggregate, it has no more meaning than consumption, investment and GDP. To stabilize any stock or flow definition of money doesn't imply anything regarding the stability of the economic system. And the very reason of intertemporal discoordination, the expansion of bank and financial credit, may now cause an undershoot (a short term, at least under flexible prices, reduction of production below the already reduced after-binge path) of real production and employment.

Can it be avoided (the undershoot, not the absolute fall in production, which is necessary, and I have no idea of the relative importance of the two factors)? It would be nice: liquidate all the rotness but nothing more than the rotness. That's a terrible task for entrepreneurs, especially in a rapidly changing and deteriorating environment.

I claim that I know of no plan that can realistically avoid the deflationary undershooting without keeping malinvestment from being liquidated. The banking and financial system shall find its equilibrium values for multipliers, mismatches and leverage: in the meanwhile, credit destruction will cause a reverse Wicksell effect which may cause more liquidations than strictly needed.

If to go from A (malinvestment) to B (a new equilibrium) credit must be destroyed, probably there will be an intermediate point which is worse than A and B (C: overcontraction due to credit destruction). To avoid reaching C means to avoid a reduction in the overexpansion of banking and financial markets, i.e., to impede the reaching of B.

Theoretically, deflation is costly, and credit destruction worsens, let's also say unnecessarily, the pain of readjustment. But how can it be avoided? With "how" I mean a fairly realistic set of instructions to be handed to Bernanke.

If the Great Depression can't be explained only by ABCT, it's because externally imposed restrictions of competition in goods and labour markets, protectionism and other anti-market policies are to be taken into account. Moreover, I have troubles imagining why there should be anything more needed to explain its depth and length: isn't Cole/Ohanian enough? Great Depression policies forced a point D on the market that was worse than B and C, and kept the economy from reaching B by suspending or hampering the microeconomic readjustment process.

The Fed has been trying to keep the economy from realizing its own past mistakes. There is nothing that can be done to undo past mistakes execpt paying for their consequences. The real and financial structure of the US economy is experiencing a painful readjugstment, despite attempts by part of policymakers to stop this process and revert back to the previous Ponzi scheme.

I will be told that not all money creation is credit. Yes, this is as true as saying that not all money creation is intertemporally neutral. So what? The present stance of monetary policy is to try to hide malinvestment. It's failing, because malinvestment has outgrown the power of monetary policy. That doesn't mean that these sterile and harmful attempts to revive the US economy through monetary stimuli will not cause another japanese malaise. The only thing the Fed can do now is to turn a deflationary recession into an inflationary recession. A better alternative would be to go to sleep and wait for the economic system to recover from the hangover. Fortunately, with no Hoover and no Roosevelt, there will be no Great Depression.

For the very reasons why general equilibrium is unrealistic and bank credit inflation causes malinvestment, bank credit deflation implies a reduction of output below the general equilibrium path.

I'll inject some historical perspective.

The Great Depression was the first time that the federal government engaged in an activist policy to counteract an economic downturn. Hoover made that point in his memoirs, and he was correct.

So was it the Great Depression because or despite government activism? I submit that question can only be intelligently answered by looking at how markets recovered from prior economic downturns.

My question cannot be answered simply by invoking a theory.

The long quote, I think from Delong, about lower wages, prices, real balances, and the zero bound is all correct until he gets to the change in real debts. While it is true that the higher real debts makes the debtors poorer and reduces their demand, it also makes the creditors richer and raises their demand. If that transfer raises the demand for money futher because creditors demand more money than debtors, then, yes, even lower nominal values are necessary to clear markets. Personally, I found that argument evidence of bad faith. (From Delong, I think.) Everything that could prevent recovery is thrown up without mentioning the other things.

For example, expectations of falling prices will depend to depress spending. But then, recognition that prices have overshot will expand spending. A RE notion that it will work perfecly is implausible to me. But to just say, deflation is contractionary (because it will be projected into the future and raise money demand) and nothing more is just building a lawyers case.

If you assume that money demand is proportional to nominal income, then if real output is 10% below capacity because of a shortage of money, then a 10% cut in prices and wages will do the same job. Adjusting nominal money demand to the nominal quantity of money.

If, instead, the demand for money depends on real income because it is a normal good, and the demand for money is a demand for real money balances, but the two don't fit together, it is possible that a 10% decrease in real income can reduce money demand to meet the existing quantity (becaues people are poorer) but a 50% decrease in the price level would be needed (because people really want to hold money rather than capital goods, securities, or whatever.)

Macro is the micro of the medium of account. And the only way the medium of account can be effective in a market economy is by tying it somehow to the medium of echange. It is all about the quantity of money and the demand to hold it.

If you refuse to consider shortages or surpluses in a single market, assuming the price has already adjusted, then I suppose monetary disequilibrium will make no sense either.

I don't think market clearing always is the sensible way to think about micro. Some markets clear better than others. Even when there is macro equilibrium, I see a world where shortages or surpluses in labor market persist for years. (There just aren't enough nurses.)

I think stock markets clear pretty continuously.

Finally, the case for nominal expendiutre targeting isn't that it results in bliss. It is rather that the various sorts of microdisequilibrium that are occuring all the time, including perhaps intertemporal disequilibirum, will be better adjusted by market forces in the context of an envirnment of stable nominal expenditure growth.

Too often crticisms of this view confuse it with some kind of real targeting. Like expanding nominal expenditure enough to get real output or unemployment to some particular target. No. That isn't it.


the quote is taken from Viner.

Jacob Viner?

Wow! Thanks.

Nick Rowe asks What is it with Microeconomists?

The problem that never seems to be discussed by Keynes is that even if government stimulus were a 'correct' response in theory (and I don't generally agree with this), this provides no evidence that government is a correct response in practice. Why? Because the government is very inefficient, and thus has very low and frequently negative returns on its expenditures. This is the primary reason the Soviet Union failed.

The comments to this entry are closed.

Our Books