Applying Occam’s Razor to the Cowen/Black model

A new paper by Tyler Cowen presents an interesting new take on the economic crisis.  Building on the insights of Fischer Black, Tyler argues that:

In essence, the story of the current financial crisis can be told in three broad chapters:  (1) the growth of wealth, (2) the decision to opt for risky investments, and (3) the underestimation of a new source of systemic risk.

Before discussing the specifics of Tyler’s proposal, I’d like to briefly summarize what we have learned in the past 250 years of macroeconomic research:

1.  Nominal GDP fell roughly 50% during the Great Contraction, and this decline seemed to have also depressed real GDP (perhaps due to sticky wages and prices.)

2.  More modest recessions are generally associated with smaller declines in the rate of NGDP growth.

3.  Inflation is a monetary phenomenon, and under an unconstrained fiat money regime the government has nearly unlimited ability to influence nominal aggregates.

And that’s about it.  Of course there are numerous other theories, but we are pretty sure about these three items, or at least as sure about them as we are about anything else in macro.

When I read Tyler’s paper it struck me that the analysis was strangely detached from the central core of macroeconomic theory.  Thus there is no discussion of the distinction between nominal and real shocks, and no discussion of how the Fed might have caused or prevented the recent sharp fall in NGDP.  It seemed like a macro paper written by an unusually perceptive microeconomist—with all the advantages and disadvantages that entails.  For instance, consider how he elaborates on the second causal factor:

The second basic trend was the increased willingness of both individuals and financial institutions to make risky investments, including the purchase of overvalued equities, risky derivatives positions, loans to such highly leveraged companies as AIG, and real estate loans (especially subprime
loans). Many of these risks were not based in the financial sector but, rather, involved unduly optimistic revenue models, as we have seen in the automotive industry, state and local governments, and such “Web 2.0″ companies as Facebook. Some of the risky investments included speculation in
volatile commodity prices, which spread the boom-bust cycle to such commodity exporters as the oil-exporting countries.

I like the fact that Tyler sees the problem as being much more pervasive than the financial sector.  The comment about state and local governments rebuts the view that the economic crisis is a failure of capitalism—governments forecast no better than the private sector.

Where Tyler and I differ is whether it makes sense to think of the problem as being a single mistake, two mistakes, or many different mistakes all made for roughly the same reason.  I have argued that two mistakes were made; investors overestimated the value of sub-prime mortgage bonds, and the Fed (and ECB and BOJ) let NGDP growth expectations fall sharply last fall.  Tyler’s paper is in some ways even more simple—he leaves the Fed out.  But it is also more complicated, as he assumes many different sectors in many different countries underestimated risk.

I was particularly struck by the comment about “unduly optimistic revenue models.”  Why would so many industries in so many countries have been overly optimistic about revenue?  Perhaps the answer is that aggregate revenue fell far short of expectations.  Suppose automakers’ plans assumed aggregate revenue (for all industries) would rise about 5% per year for the foreseeable future.  Would that forecast have been unreasonable based on past history?  I don’t see why.

Tyler would presumably argue that many sectors, including autos, underestimated the risk of a fall in aggregate “revenue” (aka nominal GDP.)  This raises a deep philosophical question (are there any shallow philosophical question?) about the nature of risk and uncertainty.  I seem to recall that Nouriel Roubini argued against the “Black Swan” view that our economy drew an unusually unlucky chip from the urn of history.  He argues that the crisis was eminently forecastable.  Perhaps the banks should have foreseen the bursting of the housing bubble.  But I can’t see how anyone could blame for banks for making industrial loans that only went bad because NGDP starting falling at the fastest rate in more than 50 years.  After 25 years of the “Great Moderation,” why would investors have expected a deflationary monetary policy from Ben Bernanke—an economist who wrote papers criticizing the BOJ’s deflationary policies?

As I read through the paper I kept nodding my head in approval as Tyler pointed out how most explanations of the crisis were not able to account for the size and scope of the problem:

The obvious question is, How were so many unsound decisions in so many countries made? A number of specific answers can be given, ranging among hypotheses about home prices, the weak transparency of mortgage securities, corporate malgovernance, excess subsidies to housing, and excessively
loose monetary policy. Although these answers may have merit in explaining particular aspects of the crisis””given that bubbles have burst in just about every asset market and in many countries””they do not seem sufficiently fundamental.
Once we liberate ourselves from applying the law of large numbers to entrepreneurial error, as Black urged us, another answer suggests itself.  Investors systematically overestimated how much they could trust the judgment of other investors.

Replace the final “other investors” with “central banks” and this would be exactly my view.  Now consider the following attempt to embed his hypothesis within a rational expectations framework:

One view of rational expectations is that investors’ errors will cancel one another out in each market period. Another view of rational expectations is that investors’ errors will cancel one another out over longer stretches of time but that the aggregate
weight of the forecasts in any particular period can be quite biased owing to common entrepreneurial misunderstandings of observed recent history. In the latter case, entrepreneurial errors magnify one another rather than cancel one another out. That is
one simple way to account for a widespread financial crisis without doing violence to the rational expectations assumption or denying the mathematical elegance of the law of large numbers.

If investor expectations were rational, then I don’t see how one can argue that over-optimistic expectations were a cause of the crisis.  Perhaps I misunderstood his “without doing violence” remark.  I have no doubt that most children don’t expect a serial killer to appear in their schools, and when the rare Columbine-type event occurs, the cause is not usually viewed as being student overconfidence.  Tyler seems to envision a world with victims but no villains.  That’s often the tack I take in arguments, and I suppose it is the right way to look at almost any major economic policy screw-up.  To understand all is to forgive all. The Fed was probably trying to do the best they could.

But where does that leave us?  If the Fed isn’t the villain I claim they are, then I have no problem with the dispassionate and fatalistic tone of his analysis.  But if there are no villains, I at least see more policy implications that Tyler does.  It’s not feasible to send kids to school with bulletproof vests, and it’s also not reasonable to expect the auto industry to adapt well to a 40% drop in revenue (factories are long term investments.)  Instead, the Fed needs to prevent a situation where almost all industries are seeing major losses of “revenue.”

I would never dispute the fact that real estate was overbuilt in sub-prime markets, and that a painful adjustment would have been necessary even if NGDP had continued sailing along at a 5% growth rate.  But that is all I will concede.  I don’t see much evidence for the following sweeping claim:

At the same time, the U.S. economy needs to undergo significant sectoral shifts. Resources need to be moved out of finance, out of construction, out of luxury goods, out of big-box retail, out of domestic auto production, and out of many economic activities sustained by bubble-driven borrowing.  Arguably, large adjustments are also needed in the
energy and health care sectors.

People love to ridicule Andrew Mellon’s famous line “liquidate labor, liquidate stocks, liquidate farmers, liquidate real estate.”  But where did it come from?  Isn’t there a deep human impulse to assume that an economic crash exposes the excesses of the preceding era?  Tyler’s thesis is certainly more sophisticated than Mellon’s–he assumes people underestimated the risk of certain investments.  He’s not arguing there was too much employment, or too much consumption.  And in the case of residential real estate he is partially right about the need for reallocation.

But even in real estate one must be careful about drawing hasty conclusions.  In the past several years housing prices have fallen in virtually every market in America and Canada.  But housing was overbuilt in only a minority of those markets.  In mid-2008, long after the sub-prime fiasco had been exposed, the housing markets of much of the US heartland and almost all of Canada were still relatively healthy.  Only when aggregate “revenue” began to fall sharply in the US, and Canada, and indeed most of the world, did the crisis spread out of the sub-prime markets and into more stable markets, as well as a thousand other industries.

It would be easy to look at the list of overbuilt sectors provided by Tyler and ask sarcastically whether we are to become a nation of hamburger flippers.  But I’m a Chicago School economist, and if the market tells me that’s what we need, so be it.  But is that what the market is telling us?  How do we know a sector is overbuilt?  Is it because sales are expected to remain far below capacity for many years to come?  If so, why not try to increase sales by boosting NGDP, to provide more revenue for all industries?  Perhaps the reallocation problem is as severe as Tyler asserts—in that case a policy of 5% NGDP growth might simply generate stagflation.  But we won’t know until we try, and 5% NGDP growth will not lead to rapid wage growth, and hence even in the worst case the core inflation rate will remain low.

I would also argue that a severe stagflation scenario is highly unlikely.  The worst re-allocation problem that I can ever recall was the housing slump of mid-2006 to mid-2008.  If I am not mistaken we saw eight straight quarters where real GDP growth was reduced as resources moved out of housing and into other sectors.  By mid-2008 the transition was mostly complete.  And during this period the economy also had to absorb a severe energy price shock.  And what did that massive re-allocation problem lead to?  A small increase in unemployment, from slightly below the natural rate (4.6%) to slightly above (5.6%.)  That’s all.

Tyler’s paper is very short and worth reading in its entirety.  My excerpts don’t adequately capture his model, and Tyler is especially good at exposing the flaws in previous explanations of the crisis.  But in one key respect our views could not be more different.  Tyler simply ignores the mainstream approach to business cycles.  I don’t know if this was his intention, but I get the feeling that he views this crisis as having exposed the inadequacies of simplistic aggregate models such as AS/AD.  In contrast, I think we ended up here precisely because we ignored the AS/AD model; and especially the message from the Great Contraction—never, ever, let NGDP fall.



52 Responses to “Applying Occam’s Razor to the Cowen/Black model”

  1. Gravatar of Nick Rowe Nick Rowe
    9. June 2009 at 01:38

    Good post Scott.

    But those of us who argue for a monetary explanation also have our own “law of large numbers” problem. Perhaps it’s a law of middling numbers.

    Why did so many central banks around the world make the same mistake at the same time?

  2. Gravatar of Bill Woolsey Bill Woolsey
    9. June 2009 at 03:44


    If the problem is readjustment, where are the sectors of the economy signalling that they are starved for resources–rising prices, high profits and growing output and employment? While we can imagine that these sectors will face bottlnecks. It seems plausible that shrinking sectors might be able to shrink faster than growing sectors can grow. But there should be growing demand somewhere–why else is this a resource reallocation?

    If there was too much risk before, then less risky investmnet projects or consumption should be expanding. Are they? (I would guess that private equity firms investing in firms with high growth potential shrinks. Pedestrian investment in large firms should grow. Well, that isn’t what I see.

    If people are so worried about risk that they refuse to invest in anything, then the other option is consumption now. I think consumption is dropping.

    In the “long run” less investmnet and even safer investment is going to result in lower productive capacity. And, I suppose, to the degree this is forseen, this means lower real incomes now. (Current capital income is based upon expectations of future output generated by expeced income. And so is labor income, for that matter.) Still, there shouldn’t be any problem with surpluses of everything. It is rather shortages that should be the problem.

    It aways comes down to people choosing to enjoy more leisure (or home production as Kling says) because the real compensation for labor is less. Again, there is a decrease in the amount of resources available for production. Production falls because of less capacity. And, we come up with explaations as to why this happens.

    Frankly, I think it all gets back to market clearing. If you assume market clearing, nominal income is never a problem. Prices (and wages) adjust. And so, the fluctatiosn in real output are fluctuatoins in productive capacity. And we grasp at any available straw.

    If, instead, you go with the obvious–prices and wages are sticky. Nominal expenditures fall. Sales fall, and production falls..then the reality is so much more obvious.

  3. Gravatar of ssumner ssumner
    9. June 2009 at 03:48

    Nick, Good question. But I would be very surprised if it was any other way. My argument is that the 100s of economists who work at the Fed are likely to see things in pretty much the same way as the 1000s of economists who work in US academia. In an earlier post I argued that the Fed generally reflects the consensus of economists. If I am right, then you’d expect the Fed and ECB to look at things in pretty much the same way. They use the same model, and they look at roughly the same set of facts. My argument is that their model is flawed in several ways:

    1. The low interest rate equals easy money illusion
    2. Not forward looking enough
    3. Too much focus on long and variable lags
    4. For the ECB, too much focus on monetary aggregates

    Lots of other countries are latched on to the $ or euro through fixed or quasi-fixed exchange rates. So mistakes by those two central banks are enough to drive world AD much lower. Once that happens, smaller countries like Taiwan will lose much of their export markets, regardless of what they do to NGDP. If they try to inflate, they’ll get distortions in their economy. (I am basically working with a closed economy model in this blog.)

    So I would give two answers:

    1. The Fed and ECB (and perhaps BOJ) are all you need.

    2. Even if you needed 20 central banks, you’d expect them all to reflect the intellectual consensus as to what constitutes sound control banking.

    In 1930 all central banks should have devalued against gold. I seem to recall that none did. Highly correlated mistakes are very possible when everyone has the same model. And in a sense this is the same point I am making about errors in different industrial sectors. They all concluded that roughly 5% NGDP growth was to be expected, and they were all rational in their expectations, and they were all wrong.

  4. Gravatar of ssumner ssumner
    9. June 2009 at 04:04

    Bill, That’s a very good observation—I don’t know why I didn’t think of making that point.

    I seem to recall in the comment section of the Austrian post that some people pointed to a very critical appraisal of Austrian economics by Tyler Cowen. Do you recall that Cowen piece? Is he skeptical of Austrian economics? I ask this because his view of the re-allocation problem seems similar to points made by some of my Austrian commenters.

  5. Gravatar of Alex Golubev Alex Golubev
    9. June 2009 at 05:37

    Scott, i don’t know of a more effective way of sharing info with you. it’s not directly related to this post, but i think you’ll find this (short) analysis interesting:

  6. Gravatar of Thruth Thruth
    9. June 2009 at 07:56

    Scott: Nice post. I was curious as to what your reaction to Tyler’s piece would be. I also wonder if Tyler’s point 3 is an implicit assertion that systemic risk can undermine the Fed’s control of monetary policy?

    Bill: “If the problem is readjustment, where are the sectors of the economy signalling that they are starved for resources-rising prices, high profits and growing output and employment? While we can imagine that these sectors will face bottlnecks. It seems plausible that shrinking sectors might be able to shrink faster than growing sectors can grow. But there should be growing demand somewhere”

    wasn’t that China and commodities (+commodity intensive industries) until last September?

  7. Gravatar of azmyth azmyth
    9. June 2009 at 09:13

    George Mason has a very strong Austrian program, and so Professor Cowen is well versed in Austrian arguments, but is still skeptical. Here are a couple of his Austrianish posts:

  8. Gravatar of saifedean saifedean
    9. June 2009 at 10:20


    You say that the lesson of the Great Contraction is Never let NGDP fall.

    If I could summarize the Austrian lesson in one sentence, it would be: never raise NGDP by printing money.

    I hope you can now begin to see the meaning of the Austrian view on this: raising NGDP can easily be done by anyone with a printing press and green ink, but it doesn’t increase real GDP. This monetary expansion would then create the illusion of increased wealth. This leads people to make miscalculations like the ones you ant Tyler mention. Once the monetary expansion stops, these calculations that were built on the assumption of increased NGDP will backfire. The result is massive malinvestments like today’s housing sector.

    Whatever production happens because of an increase in the money supply is production that happens because of the distortion of the price mechanism. This means that the economy, without price distortions, suggested to individuals that certain enterprises were not worth undertaking. With the price distortions, these enterprises appeared to be worth undertaking. This will generate jobs and output, leading to the boom that the Keynsians tout, and make it look as if things are good. But, the reality is that there was no reason for these investments to be undertaken, since that was what the true price signals (shaped by supply and demand, scarcity and preferences) conveyed.

    Once the monetary expansion ends, the real relative prices will be re-established, after a period of adjustment in which inflation is dissipated from the boom sectors to the rest of the sectors, and the real relationships that initially conveyed that these undertakings were not worthwhile will re-emerge. The result is massive losses that are LARGER than whatever gains happened during the temporary boom.

    You in Chicago view the drop in NGDP as the root of all evil. In Vienna, they view the artificial increase of NGDP to be the root of all evil–including the evil of the drop in NGDP. That’s why the Austrians would oppose the increase in the money supply always, and that’s why they sound like a broken record to you and the rest of the mainstream.

    The important thing to remember, as Mises pointed out almost a century ago, is that there is NEVER any reason for increasing the money supply. Any money supply is always and forever enough for any amount of economic activity. What matters is money’s purchasing power, not its denomination. Turkey knocked six zeros off of its currency recently, and the people did not become a millionth as wealthy.

    This is the Austrian story in a nutshell: there is never a need to increase the money supply. Increasing the money supply is always redistribution, it is always inflation, it increases NGDP unsustainably, it distorts the price mechanism, it causes malinvestment, and it inevitably backfires.

    History has been unequivocal in its vindication of the Austrian story.

  9. Gravatar of Lord Lord
    9. June 2009 at 14:26

    Isn’t long and variable lags the time it takes for central banks to get it right? I can understand focusing on rapid and sharp changes in expectations in the real economy as causal whereas central banks react belatedly and slowly to them appearing more as effects. They may fear rash actions either might not work or work too well and have difficulty calibrating their response, as well as fearful of a loss of confidence and even of shifting the blame to themselves as a result. With a manic market, they wish to appear deliberative and sober. It really seems to me an expectations game with both bluff and action, trial and ordeal, in which both test each other.

  10. Gravatar of Winton Bates Winton Bates
    9. June 2009 at 17:14

    It is interesting that you and Tyler both acknowledge that investors made systematic errors. Does it matter whether their mistakes arose from overvaluation of sub-prime mortgages or from a more general tendency to systematically overestimate how much they could trust the judgement of other investors?
    It seems to me that whatever the mistakes made by investors your point still stands that the recession would not have occurred if central banks had acted to prevent a sharp fall in NGDP.

  11. Gravatar of Jon Jon
    9. June 2009 at 18:16


    My own general opinion of Cowen’s work is that he got the critique of the ABCT wrong, but Barnett’s defense of ABCT is flaccid in certain regards.

  12. Gravatar of TGGP TGGP
    9. June 2009 at 19:06

    Andrew Mellon quite likely never said that, but rather Herbert Hoover just made that claim to make himself look better in comparison.

  13. Gravatar of ssumner ssumner
    10. June 2009 at 05:51

    Alex, TIPS show low real rates and rising inflation expectations.

    Thruth, You may be right about Tyler’s view of risk and monetary policy. I don’t understand risk well, and try to stay away from it (just like IS-LM.) My gut feeling is that like other famous Keynesian concepts (interest rates, investment, etc.) it is a symptom, misdiagnosed as a cause of business cycles.

    azmyth. Thanks. BTW, Treasuries have risen sharply, and the Asians central banks haven’t “come to their senses”, so it’s not clear Tyler was right. Or perhaps I should say it is not clear the Austrian prediction that Tyler made was right.
    If I make lots of predictions, some will come true.

    saifedean, What about Japan? Their NGDP has been stable since 1994. Is that good? Their money supply has increased considerably.

    Hayek originally agreed with those who opposed monetary stimulus during the Great Depression. That policy error led directly to WWII. Later Hayek realized that he had been wrong, and argued that monetary policy should have been stimulative, to offset the fall in velocity. Hayek is my favorite Austrian economist.

    Lord, I don’t think markets are “manic,” indeed I think they are the most sober and thoughtful forecasters we have. When markets fell last summer and fall they were warning the Fed that money was too tight. The Fed ignored that warning, and now we are paying the price with massive fiscal stimulus which will lead to much higher taxes.

    The Fed should follow markets every single day. Every little up and down is important. Nothing should be ignored. They should use markets as their compass.

    Winton, Tyler doesn’t really talk about policy implications. I would say that our differences only matter if they result in Tyler reaching different policy implications. And I don’t know if he did.

    TGGP, It doesn’t surprise me that Hoover made up the line. I like Mellon much better than Hoover.

  14. Gravatar of Bill Woolsey Bill Woolsey
    11. June 2009 at 15:04

    We can treat the results of a single firm’s investment project like some kind of random process. But it is a fallacy of composition to add them up all up and then say that all firms may have bad luck at once.

    People earn income and buy the things on offer. If a firm develops a new “better” product, but in reality, people don’t prefer it more, then its value may well be less than its opportunity cost. The value of the other goods sacrficed. But it is impossible for every firm to be in this situation. The problem for the firm making the entrepreneural error is that it is pulling resources away from other more valuable products.

    Now, firms do introduce new production methods. And, perhaps they are not as effective as planned. And if that is the sort of “investment” we have in mind, then I suppose it is possible that production could drop. Lots of firms produce less output because of production processes that are in fact less productive (though they were tought to be more productive.)

    And, of course, this would result in shortages of goods and services, rather than a general glut.

    Perhaps I misunderstood Black/Cowen’s argument, but it seems to me that there the this notion that all the firms produce the wrong thing at once is impossible. And, further, what counts from the point of view of profits, is how good the products are relative to one another.

    Is it really just an artifact of market clearing reeasoning. The price level always adjusts to that real aggregate demand equals productive capacity. Aggregate fluctuations must be do to fluctuations in productive capacity. Changes in employment must be due to valuing home production and leisure over the products of firms. Unemployment of labor is just taking time to gradually search out jobs

  15. Gravatar of ssumner ssumner
    12. June 2009 at 05:09

    Bill, I also have problems with the Cowen model, for much the same reason as you do, but I’ll first try to defend the approach for the sake of argument:

    1. The public as a whole can misforecast a given macro variable, such as the future inflation rate. Everyone agrees on that point. So in the early 1960s most expected low inflation in the late 1960s and 1970s, but most were wrong.

    2. People’s forecast errors in individual industries tend to be uncorrelated, when applied to industry specific issues.

    3. Cowen is saying that people made an error analogous to point 1, they almost all underestimated the way in which systemic risk would increase over time. Consider “systemic risk” like “aggregate inflation.”

    4. Once they understood that they have made the error in point three, they found themselves with too much investment in certain industries. Then you use an Austrian-type explanation to go from mis-allocation to temporary unemployment. I think that is Cowen’s argument.

    I think this is possible in theory, but as a practical matter I think it only applies to housing, and only to a very specific part of housing–the sub-prime fiasco. I don’t even think it explains the more recent problems in the housing market, which I think are due more to falling NGDP.

  16. Gravatar of saifedean saifedean
    12. June 2009 at 08:11

    “saifedean, What about Japan? Their NGDP has been stable since 1994. Is that good? Their money supply has increased considerably.”

    — I think you’re misunderstanding my point, since I see no way in which this comes to bear negatively on my argument. My point isn’t that NGDP should stay constant; it is that the money supply should never be increased, ever. Productive activity increases wealth, with a constant money supply, causing prices to fall. Increasing the money supply just causes miscalculations and inflation.

    Japan has been inflating and increasing the money supply considerably, which I think is bad. Their economy has not been growing much (even contracting), for whatever reasons. The money printing does not help in this, it raises meaningless metrics like NGDP, but does not increase actual wealth and production. On the contrary, it makes planning very difficult and so it HURTS real economic growth.

    In fact, Japan is a perfect counter-example to your viewpoint: They’ve done everything to raise NGDP and their economy just would not grow in real terms.

  17. Gravatar of Bill Woolsey Bill Woolsey
    13. June 2009 at 02:28


    If Cowen’s point is that the price level may be inconsistent with monetary equilibrium, and that the question is why isn’t the average forecast always correct, then… yes.

    But then, why isn’t the discussion of price setting strategies rather than investment?

    And, when we are speaking about investment, then cost as opportunity cost implies that there is a necessary negative correlation. Overproduction of one thing is under production of something else. Pulling resources into a failed project, pulls resources away from something else that is now underproduced.

    I don’t see how it can be anything else other than the naive monetary disequilibrium hidden in the background. If the firms don’t entice the people with a good deal, they won’t spend and so sales are low. And why can’t this happen across the board?

    Well, what do you mean, “not spend?” It means that you are using income on something–accumulating money balances. And then, the question is, why doesn’t the nominal or real quantity of money increase, or the yeild on money decrease, to clear the market for money? And, at the same time, maintain a level of nominal expenditures so that changes in spending in the economy reflect the reality of opportunity cost? Roughly.. shortages and surpluses match?

    And, of course, if it is that people don’t want to use their resources, chiefly labor, to earn income because their is nothing to buy with that income that they want.. then yes, total production can and should fall. And so, those sorts of recessions should involve lots of long vacations. And, our antedotal (and polling) experience of recession shouldn’t be.. I just can’t find a job. It should be, I am dropping out of the labor force for a while, because I just can’t find anything worth buying.

    We can imagine that people work and accumulate money, hoping to buy other stuff in the future. But it is still a problem of monetary disequilibrium. The quantity and yeild on money aren’t adjusting in a way that clears money.

    Similarly, if people can’t find a decent investment opportunity, and so, consume their income, then that will eventually result in less production (or at least less growth.) So, we should see recessions that are like the Austrian theory… consumer industries are booming.

    Finally, I think it is entirely possible for there to be reductions in the productive capacity of the economy because of shift in the allocation of resources. It is because of specific human and physical capital and the time needed to redeploy resources. But there should be shortages to match the surpluses.

  18. Gravatar of Bill Woolsey Bill Woolsey
    13. June 2009 at 02:48


    An increase in the production of goods and services, in aggregate, raises income (real income,) not wealth. The increase in income raises the demands for most everthing. Increases in production (due to more resources–population and newly built capital goods–or due to improved productivity–due to better technology) results in increases in the supplies of everything. When supply and demand both increase, prices remain steady. Of course, basic supply and demand is about relative prices and real income and the like, and, of course, relative prices can’t all fall in relation to one another.

    Anyway, the only reasson why growth causes lower money prices is if the demand for money rises with income. So, as real income rises, poeple want to hold more money. Given the nominal quantity of money, the price level must fall until the real quantity of money rises to meet the growing demand for real money balances. If the demand for money did not rise with real income (if money isn’t a normal good,) then any decrease in the price level because of increased supply would result in an excess supply of money, greater spending, and the price level returning to where it started. If money is inferior, then growth would result in a higher price level.

    Always, the question is how is a shortage or surplus of money cleared up by the market economic system. That is, an imbalance between the quantity of money and the demand to hold that quantity of money. The market process for a given nominal quantity of money is for all prices, including nominal incomes like wages, to adjust. But all of these prices already have “jobs” in the market economy system. The money prices adjust to refect changes in the relative prices of things–the values that buyers put on them and the opportunity costs of the other things that could be produced with them. Most goods have specialists that try to gauge the proper prices and production of these goods, and they are rewarded based on their success with profit and loss.

    The market process by which a given quantity of money is cleared up requires all of those specializing in the prices nad production of a few products to take on a second job–determining the demand to hold cash balances through out hte economy. With contracts made in money terms, this also involves judgements about the future path of the demand to hold cash balances. (With durable capital goods, they do, of course, make judgements about their particular markets in the future too.)

    I find it no surprise that making everyone a part time money entrepreneur works worse than having specialist entrepreneurs running various markets.

    And that is why it is better to have a monetary institution where the nominal quanity and/or yield on money adjusts so that it accomodates changes in the demand to hold money. Those who make these adjustments can specialize, and leave others to concentrate on the proper relative prices and production in particular sectors of the economy.

    The notion that nominal GDP is an irrelevant metric is naive. If nominal income grows with the real productive capitacity of the economy, then the shortages and surpluses in the economy match. This refelcts the reality of production–where overproduction of particular products means that resources are being pulled away from the production of other things. With nominal income equal to the real porductive capacity of the economy, and the price level stable, then higher prices of one thing generally do signal a need to produce more. And lower prices, and desirability to produce less. Changes in money prices reflect changes in relative prices.

    Anyway, I think you are in error. Increases in the quantity of money that meet an increase in the demand to hold money are equilibrating. They don’t distort anything.

  19. Gravatar of ssumner ssumner
    13. June 2009 at 06:08

    Saifedean, I don’t understand your point about Japan. You seem to say Japan raised NGDP but it failed to raise RGDP. Just the opposite. NGDP has been roughly stable for 15 years. The real GDP rose by roughly 15% and the price level (GDP deflator) fell by about 15%. I seem to recall Austrians arguing that one should keep the money supply constant and that NGDP would be pretty stable, RGDP would grow and prices would decline due to improved productivity. What’s interesting about Japan is that the macro variables have behaved this way, despite the fact that the money supply increased substantially (apparently offsetting a fall in velocity.) So other than the money supply increase, I still think the Japanese case is relevant to the Austrian model. But I also know there are different versions of ABCT, so obviously I can’t say it’s relevant to all versions.

    Bill, I think that is a good argument and I should have talked about it. I do understand why investment mis-allocation might make RGDP fall during a transition phase, but as you say it should look different from what we observe. You should see booming industries drawing resources from declining industries, not a general lack of demand in almost all industries. Perhaps there is an explanation, but like you I don’t see it.

  20. Gravatar of saifedean saifedean
    13. June 2009 at 09:22


    You’ve hit at the crux of our disagreement and laid it bare. Thanks!

    Before I get into a detailed reply to your points, allow me to begin with the general context. Mises, in 1912, made the most important (and most ignored) observation in monetary economics: any money supply is enough, and there is never a need to increase the money supply. I know Chicagoans have an aversion to reading anything by Austrians, but please humour me for a minute and try to entertain Ludwig’s point:

    “The services money renders are conditioned by the height of its purchasing power. Nobody wants to have in his cash holding a definite number of pieces of money or a definite weight of money; he wants to keep a cash holding of a definite amount of purchasing power. As the operation of the market tends to determine the final state of money’s purchasing power at a height at which the supply of and the demand for money coincide, there can never be an excess or a deficiency of money. Each individual and all individuals together always enjoy fully the advantages which they can derive from indirect exchange and the use of money, no matter whether the total quantity of money is great or small. Changes in money’s purchasing power generate changes in the disposition of wealth among the various members of society. From the point of view of people eager to be enriched by such changes, the supply of money may be called insufficient or excessive, and the appetite for such gains may result in policies designed to bring about cash- induced alterations in purchasing power. However, the services which money renders can be neither improved nor impaired by changing the supply of money…. The quantity of money available in the whole economy is always sufficient to secure for everybody all that money does and can do.”

    A bit more from Rothabrd:

    “With respect to the supply of money, Mises returned to the basic Ricardian insight that an increase in the supply of money never confers any general benefit upon society. For money is fundamentally different from consumers’ and producers’ goods in at least one vital respect. Other things being equal, an increase in the supply of consumers’ goods benefits society since one or more consumers will be better off. The same is true of an increase in the supply of producers’ goods, which will be eventually transformed into an increased supply of consumers’ goods; for production itself is the process of transforming natural resources into new forms and locations desired by consumers for direct use. But money is very different: money is not used directly in consumption or production but is exchanged for such directly usable goods. Yet, once any commodity or object is established as a money, it performs the maximum exchange work of which it is capable. An increase in the supply of money causes no increase whatever in the exchange service of money; all that happens is that the purchasing power of each unit of money is diluted by the increased supply of units. Hence there is never a social need for increasing the supply of money, either because of an increased supply of goods or because of an increase in population. People can acquire an increased proportion of cash balances with a fixed supply of money by spending less and thereby increasing the purchasing power of their cash balances, thus raising their real cash balances overall.”

    “A world of constant money supply would be one similar to that of much of the eighteenth and nineteenth centuries, marked by the successful flowering of the Industrial Revolution with increased capital investment increasing the supply of goods and with falling prices for those goods as well as falling costs of production. As demonstrated by the notable Austrian theory of the business cycle, even an inflationary expansion of money and credit merely offsetting the secular fall in prices will create the distortions of production that bring about the business cycle.”

    [both quotes from The Austrian Theory of Money, by Murray Rothbard

    Now think about it for a second. The amount of money in an economy is a purely irrelevant statistic, precisely as meaningful as the length of an inch. The whole of the American economy can function identically with a money supply of $10 Trillion, $30 Trillion, $5,000 Trillion, $17 Million, $543.78, $12 or $0.000001. This is truly just a number. The amount of economic activity will be divided in the same proportion to the total money supply in order to give the differing price levels. If the money supply was $10T, prices would be a third of prices if money supply was $30T. Adding a zero next to every bill would increase the money supply tenfold, but also increase prices tenfold. Turkey woke up one day and decided to knock 6 zeros off of its currency and it went about its life normally.

    Any increase or decrease in money supply, could never, under any circumstance, confer any benefit on society as a whole. America would be no better off with triple its money supply or a third of it anymore than its citizens wouldn’t be taller if the inch was a third of its current length or triple it. If the increase or decrease in money supply is done completely proportionately (a la Turkey’s zero-cide) then it has no benefits, except for making calculations easy and saving people the hassle of knowing what a billion billion actually is.

    In effect an increase in money supply is simply a transfer of wealth from all of society to those who receive the new money first. The only benefits of an increase of money supply always and forever occur to the people who first receive the new money, for they simply get this money’s purchasing power, while everyone else’s money’s purchasing power decreases. People who are on fixed wages (the poor) are the ones who suffer the most as inflation takes hold, raises prices, and their wages remain the same.

    Now, bearing this in mind, here’s how I’d answer your points. My general problem, as an Austrian, is a methodological one. You set up macro aggregates that in my opinion do not mean anything in the real world and then assume there are scientific relations governing them, and that economic theory uncovers these relations. Austrians work from the micro up. Please read Hayek’s Nobel Prize acceptance speech for the best statement of this point.

    So for instance, you say: “An increase in the production of goods and services, in aggregate, raises income (real income,) not wealth.”
    No. An increase in production, with a constant money supply, does not increase income, it keeps income constant, but increases the purchasing power of income.

    “The increase in income raises the demands for most everthing.”
    There is no such a thing as aggregate demand. This is just Keynsian/Fischerite smoke and mirrors. You insert aggregate demand into the analysis, and it allows you to come up with whatever result you want.

    In the real world, where laws of physics and conservation of matter apply: a rise in production leads to a rise in purchasing power of money, and people buy more stuff and have more needs met, and everyone is happier. There are more goods out there being bought with the same amount of money. That’s it.

    We are so used to the idea of inflation that we cannot even get ourselves to question that prices continue to rise. They shouldn’t. If there was no central bank constantly increasing the money supply, prices would simply fall constantly. Economic growth would happen by people gaining more purchasing power to their money, not by having more zeros next to their money.

    “Anyway, the only reasson why growth causes lower money prices is if the demand for money rises with income.”
    No one ever wants money for its own sake, what people want is the purchasing power of money. With economic growth and deflation, this purchasing power increases. End of story. No need to increase the money; the money increases its purchasing power.

    “Given the nominal quantity of money, the price level must fall until the real quantity of money rises to meet the growing demand for real money balances.”
    There simply is no need for the quantity of money to rise! The price just falls and everyone is happier. What exactly is so horrible about the fact that your computer today is twice as good and half as expensive as your computer from 2 years ago? It hasn’t destroyed the world, has it? Stuff gets better and cheaper, and that’s a good thing. There’s no need to print money to make it expensive again!

    “Always, the question is how is a shortage or surplus of money cleared up by the market economic system.”
    There is never such a thing as a shortage or surplus of money. Any quantity will do. What matters is relative prices and purchasing power. What you are saying is simply Irving Fischer rehashed. Hasn’t the spectacular failure of everything Fischer believed in taught Chicago anything? His attempt at scientifically managing prices in the 1920’s led to the stock market crash. He was under the illusion that he needed to increase and manipulate the money supply to make sure it is sufficient and that the prices work properly. For 90 years Austrians have been saying that this is futile, needless, impossible and will lead to disaster, because:
    1- Any supply of money is enough
    2- The price mechanism is a self-coordinating emergent process that will determine relative prices on its own, not with a central planner.
    3- The price mechanism can only work without any form of central planning.
    4- Any attempt at central planning of prices will inevitably fail, for the very same reason that central planning of potatoes, cars or substance abuse will fail.

    “The market process by which a given quantity of money is cleared up requires all of those specializing in the prices nad production of a few products to take on a second job-determining the demand to hold cash balances through out hte economy.”
    You must read Hayek’s The Use of Knowledge in Society. The price mechanism does NOT work through any one person making it his job to determine the demand to hold cash balances! It works precisely because it is no one’s job to make it work!! Please, please, please re-read that paper. Everyone Chicagoan/Keynesian I know has (at best) skimed it in a freshman seminar, and continues to believe that they understand it, when they patently don’t. It’s the most important paper on prices in economics, and every economist needs to understand it.

    “And that is why it is better to have a monetary institution where the nominal quanity and/or yield on money adjusts so that it accomodates changes in the demand to hold money. Those who make these adjustments can specialize, and leave others to concentrate on the proper relative prices and production in particular sectors of the economy.”
    You must realize here that you are under a strong illusion: you formulate some false Fischerite theory (based on meaningless aggregates) about how money behaves, and then you posit that that central planner is needed. Then, you take the evidence of the failure of the central planning of money as evidence for the need for more central planning of money, and the unsuitability of free exchange to determine how money works. I bet that for you, the 1929 crash illustrates a market failure that shows the need for better smarter monetary management. You fail to notice how it was precisely “better smarter scientific money management that led to the crash and the depression!

    In a free market, money and credit would be just like any other good: determined by supply and demand. Money would be constant, with only a bit of new mining increasing the supply (to keep our coins from becoming too small). Credit would be sold and bought like any good, through people’s preferences for investment, saving and consumption. If a lot of people want to loan money, the price of credit (interest rate) drops; if more people want to borrow, the interest rate rises. Credit simply is another market that can only clear without any intervention.

  21. Gravatar of saifedean saifedean
    13. June 2009 at 11:54


    I am not sure we understand each other on Japan. We talk a different language. My main problem with your language is that Japan exposes the utter fatuity of things like GDP, RGDP, NGDP. You can torture these numbers to tell you anything.

    But here’s an alternative, Austrian, way of looking at Japan, and I’d love to get your thoughts on it:

    After the housing and stock market bubbles burst, the malinvestments were to be liquidated, and the money supply was dropping. An Austrian would tell you you should leave things alone because that’s the best way to end the recession and have a healthy recovery. Unfortunately for Japan, they’re even more Keynesian than America. So the government goes and builds roads whose only purpose is to lead to new places where one can build new roads. This is a complete waste because it is an increase in the money supply, it is inflation, and it is wasteful production that no one would have paid for willingly in a free market. But it is government-mandated: Soviet Keynesianism at its best.

    The problem with GDP is that bridges to nowhere count as an increase in GDP. The reality, of course, is that we do not care about make-work; we care about increased valuable stuff for real-life consumers. When the government takes resources and wastes them on worthless garbage, that is not good; that is bad. But it still registers as a growth in GDP and politicians get re-elected, even as people are getting poorer.

    So there have been three forces working for the past few years: 1- the deflationary liquidation of malinvestments from the stock market and housing bubbles (deflationary, lowers NGDP, lowers RGDP necessary and good). 2- The inflationary Keynesian bridges to nowhere (inflationary, raises NGDP, raises RGDP, wasteful and bad). 3- Real actual good productive economic activity raising productivity, increasing supply, decreasing prices (deflationary, raises RGDP, raises NGDP, productive and good).

    Because of these three countervailing forces, the Japanese economy has been in this malaise for a while. Spending increases nominal GDP in the short-run, while hurting the economy in the long-run. Deflationary adjustment decrease GDP in the short-run, but increases it in the long-run. Economic growth increases real GDP. Add these three up and you get: The lost decades.

    If the government would stop listening to Keynesians, there would be a short sharp recession, a period of adjustment and then a healthy recovery would ensue. As long as it continues to waste precious resources on wasteful bridges to nowhere, it will continue to stall all attempts at recovery.

    This is the best illustration of the iatrogenic nature of economic planning: so long as the government listens to the Keynesians, NGDP and RGDP do not fall and politicians feel happy about themselves, but the economy is shackled and recovery cannot happen. If policymakers stop listening to Keynesians, there will be a drop in NGDP and RGDP, but that would soon be followed by a healthy recovery. Unfortunately for Japan, Keynesianism looks set to continue.

    Not surprisingly, when American Keynesians are asked about Japan, their only reply is that MORE spending was needed, and the reason the economy hasn’t recovered is because Keynesianism wasn’t tried hard enough. It’s exactly like the witch-doctor who tells you your problem is that you did not take enough snake oil!

  22. Gravatar of
    13. June 2009 at 13:18


    I have read both Mises and Rothbard on monetary theory.

    I read Mises first and then Rothbard. I only read Keynes
    and Milton Friedman much later. If you actually want to understand money and banking, try Yeager.

    Everyone pretty much understands the point about the any quantity of money is adequate. However, you are sadly in error if you think that means there can never be a shortage or surplus of money.

    The market process by which a shortage or surplus is corrected is through changes in the money prices of all goods and services so that the real quantity of money adjusts to the demand to hold real money balances. Both Mises and Rothbard explain this process. They recognize that there can be shortages or surpluses of money if the purchasing power of money is away from equilibrium.

    In microeconomics, there are surpluses or shortages of particular products when their prices are away from equilibrium.

    This however, doesn’t mean that changes in the nominal quantity of money or the yield on money aren’t a more appropriate method of adjusting the real quantity of money to the demand to hold it. Rather than requiring everyone in the economy to adjust their money prices to change the real quantity of money, the nominal quantity of money or its yield can be adjusted. Firms that specializing in issuing money and in setting the yield on money can do this, and there is no need for anyone else to try to judge what is happening to money demand so that they can set their prices accordingly.

    I have, of course, read Hayek’s essay on the uses on knowledge in society. I have no idea what you are trying to say about it. Nothing Hayek wrote is inconsistent with the notion that entrepreneurs specialize and set prices, purchase capital goods, and determine production.

    My _point_ was that imbalances between a given quantity of money and the demand to hold it do not allow this sort of specialized entrepreneurship.

    Hayek’s point was about relative prices and the allocation of resoruces–the way that prices communicate tacit knowledge about activites remote to a particular entrerepreneur.

    Hayek, of course, advocated changing the money supply to offset changes in velocity, and towards the end of his life, came to favor a privatized monetary system whem those entrepreneurs specializing in issuing money would adjust the nominal quanity of money to meet the demand for money.

  23. Gravatar of
    13. June 2009 at 13:39


    As far as I am aware, neither Rothbard nor Mises deny that an increase in production leads to an increase in income. It is just “real” income that increases necessarily. You, for some reason, uderstand “income” to mean nominal income. So, if nominal income remains the same, and prices fall, then there is an increase in “real” income.

    It is conventional to speak of use the term income to refer to income, unless otherwise specified.

    So, the flow of the production of goods and services over time–consumer goods, capital goods, etc, is matched by real incomes earned by those contributing resources to that production.

    By saving, individuals accumulate assets. By investment, firms production and accumulate capital goods. That is wealth.

    So, increased production creates more income. If it is saved and invested, then it adds to wealth.

    Aggregate demand exists. It is the sum of the nominal demands for all the output. But, I didn’t actually say anything about aggregate demand. I said that the demands for most things are positively related to real income. The demand for normal goods rises when income rises–basic microeconomics.

    The demand for Wheat is made up of the demands of a bunch of individual’s who buy wheat. While it is generally a quantity of wheat, the demand relationship allows a calculation of a nominal demand for wheat (price times quantity demanded.) And, that can be added up with other kinds of demands to get an aggegate, if you want.

    And, if fact, shortages or surpluses of money impact nominal expenditure.

    Firms will only produce what they can sell. They can only sell what people will buy. The total production in the economy is not going to be any more than the total amount of things poeple want to buy.

    What is true, however, is that real volume of purchases can rise without the nominal valume rising. It just requires lower prices. But the other side of that coin is that the real supply of money is rising to meet the real demand to hold money.

    Suppose their is a general glut of goods. What do Rothbard and Mises say? Well, prices should be lower. Well, the lower prices results in a larger real volume of purchases. It increases the real supply of money. But there is an increase in aggregate demand.

    I think you should steer away from any thought that increased production leads to lower prices because of something to do with physics.

    The price level or purchasing power is supply and demand determined. Rothbard and Mises both recognized this. Now, it is true that Rothbard, especially, peopled that changes in the demand for money should always accomodated by a change in the price level. This is the same thing as saying that the real volume of purchases should adust through changes in the price level.

    My view (which comes from Yeager) is that this process is painful. And that changes in the nominal quantity of money solve the problem in a better fashion.

  24. Gravatar of Bill Woosley Bill Woosley
    14. June 2009 at 04:22

    Wow! Sorry about all the typos.

    Anyway, income is generally used to mean “real income” unless otherwise specified as nominal income.

    For normal goods, demand is positively related to real income. And the relative prices of normal goods rise when real income rises.

    There is no way that the growth process can work if people don’t buy more goods and services as more are produced. And the relative prices of everything cannot possibly fall. It is, of course, possible for money prices to fall and for real expenditures to rise to match an increase in aggregate production.

    The real quantity of money is rising to match the real demand to hold money, and at the same time, the real volume of expenditures is rising to match the real income that resource owners earn with full employment of resources.

  25. Gravatar of ssumner ssumner
    14. June 2009 at 05:59

    saifedean, If you want to communicate with economists on an economics blog, you have to learn to speak our language. I am not going to learn a completely new language because one commenter doesn’t use economics terms the way all other economists do. Here is an example. You started by talking about what was happening to Japanese aggregates like real GDP, nominal GDP, the money supply, etc. Then when I told you that you got the facts all wrong, you told me that these aggregates don’t mean anything anyway. Well if they don’t mean anything, you should say why; don’t use aggregates you don’t have knowledge about. Here is another example, you say in reply to Bill:

    “So for instance, you say: “An increase in the production of goods and services, in aggregate, raises income (real income,) not wealth.”
    No. An increase in production, with a constant money supply, does not increase income, it keeps income constant, but increases the purchasing power of income.”

    The purchasing power of income is identical to real income. So your reply is a complete non-sequitor. You need to use terms like ‘real income’ the way other economists do, otherwise people won’t know what you are talking about. Another example: You spend a lot of time trying to show that in the long run the nominal quantity of money doesn’t matter. All economists (Austrian and non-Austrian) agree on that point. And almost all economists (Austrians and non-Austrians) agree that changes in the nominal amount of money can have short run real effects. Indeed almost all economists agree that money shocks can make the economy less stable. So you are preaching to the converted on that point. The fact that you assume we don’t know that, makes me wonder if you understand conventional economics well enough to dismiss it as inferior to Austrian economics.

    I am still waiting for any of the many Austrian economists who visit my blog to provide a single example of an important macro phenomenon that Austrian economics can explain but conventional macro cannot. And no one has come forth with one.

  26. Gravatar of saifedean saifedean
    14. June 2009 at 07:59


    Thanks for the thorough reply.

    The reason that I bring up prices is because it is precisely the price mechanism that is the way that the quantity and price of money is adjusted to meet supply and demand constraints. You seem to view the money supply as being different from everything else on the market whose.

    The whole point of prices is that they work towards bringing about equilibrium. When the demand for tomatoes is high, prices rise, tending people towards consuming less tomatoes, and incentivizing producers to produce more. The same would be true for money. Whenever there is a shortage or surplus of money, prices will work to equilibrate this through raising and lowering the prices.

    The key point here is to realize that money is like any good. Supply and demand will be equalized through the price of money. The same is true of credit. The key insight is to think of Hayek’s price mechanism as working for money as well as for other goods.

    “Rather than requiring everyone in the economy to adjust their money prices to change the real quantity of money, the nominal quantity of money or its yield can be adjusted. Firms that specializing in issuing money and in setting the yield on money can do this, and there is no need for anyone else to try to judge what is happening to money demand so that they can set their prices accordingly.”

    You continue to think of money as being a centrally-planned phenomenon, and therefore you find that “imbalances between a given quantity of money and the demand to hold it do not allow this sort of specialized entrepreneurship.”

    I see no reason why that is the case. Mainly because I see the “specialized entrepreneurship” as a bit of a red herring. Everyone is an entrepreneur in tomatoes, in a sense, because everyone’s consumption and production of tomatoes is determined by (and determines) the price of tomatoes. Similarly with money. Imagine the case where there is a constant supply of a hard money. All goods are traded with each other and with money. There is no clear separation between what constitutes a money and what doesn’t. Money is simply that which is a store of value””so gold, silver, jewelry and lots of things would in a way be money. Supply and demand determine prices of all goods relative to each other and relative to all moneys. There is no need for anyone to manage anything about prices. This is why I brought up Hayek’s paper. You need to think of its message on prices as a coordination mechanism as applying not just to prices of tin and tomatoes, but also to the price and supply of money.

    This point is perhaps better explicated in Hayek’s Monetary Theory and the Trade Cycle. In that book, Hayek discusses how the messing with the money supply is what leads to dislocations and distortions of productions through its impact on the price mechanism.

    And as for Hayek’s support for free banking””I see nothing in this that is contradictory to my viewpoint, on the contrary, it supports it. Under free banking and free supply of money, consumer choice acting through prices determines what get used as money, what price exists for it and how it relates to other goods. That’s all I’m trying to say.

    And thanks for the recommendation of Yeager. I have not read him yet but plan on doing so soon.

  27. Gravatar of saifedean saifedean
    14. June 2009 at 09:28


    I mention the point about the nominal quantity of money not mattering to emphasize the very important and completely ignored point that there is no need to increase the quantity of money. Ever. And that any increase in the quantity of money is:

    1- redistribution of the purchasing power of the old money to the recipients of the new money
    2- Distorts the price mechanism, which is the coordination mechanism of the economy
    3- Results in business cycles, booms and busts that leave the economy worse off.

    But I’m glad you ask me “to provide a single example of an important macro phenomenon that Austrian economics can explain but conventional macro cannot. And no one has come forth with one.”

    I’ll propose one: The Great Depression.

    I am currently reading Friedman and Schwartz’s The Great Contraction. While they talk at long length about the importance of the Great Contraction in causing the depression after the stock market collapsed and the bank runs started, they do not offer an explanation of how the collapse came about in the first place. What was it, exactly, in the neoclassical explanation that caused:
    1- The stock market bubble
    2- The run on the banks afterwards

    Before we can move on to discuss the contraction by the Fed, must we not explain that?

    I know Keynesians like Krugman pride themselves on not bothering with the explanation of the causes of the business cycle and instead moving on to the practical and important matters of what can be done about it””public works, employment schemes, etc… I have looked around for neoclassical explanations and they seem all to begin AFTER the crash, and they all talk about how the Fed should have injected liquidity and prevented the insolvency of banks.

    But these are all prescriptions without a sound diagnosis. Doesn’t that worry you? Would you go to a doctor who tells you “never mind the diagnosis, take this pill”? And if you continue to go to him and to take this pill, and you continue to get worse, wouldn’t you ask yourself if maybe that pill was more harm than good.

    Here, on the other hand, is the Austrian explanation of the 1929 crash and the ensuing depression:

    European and American central banks abused the gold standard and expanded their money supply in order to fight WWI. Since this is an expansion of the money supply, as Austrian Business Cycle Theory would tell us, it will lead to a transfer of purchasing power from everyone with money to those who receive the new money (the governments who used it to fight the war). Also, once the expansion of the monetary supply stops, this will lead to a corrective deflationary recession that liquidates the malinvestments of the expanded money, drops prices, and brings currencies back to a stable price ratio with gold. This is key: monetary expansion, as soon as it starts, is destined to bring about an artificial boom followed by a bust as soon as the monetary expansion stops. You can keep expanding the money supply for a long time, but you are just delaying the inevitable bust and recession.

    So, Germany expanded its money supply most (for reparations) and so had hyper-inflation. America stopped expanding after WWI and so it had the 1920 recession, which was short and sweet because the government did nothing about it.

    But Britain wasn’t as lucky. Having printed a lot more money, their currency continued to suffer. But Britain was also unwilling to undergo the necessary corrective deflationary recession to bring its currency down. So it continued to expand the money supply. But this only devalued the pound further, and meant that people could exchange their pounds for gold and ship it out, and thereby deprive the British government of more gold. So gold started leaving Britain and going to other European countries and the US””countries where the inflation had stopped, or was less than Britain’s.

    At this point, Montagu Norman, the British central banker, started leaning on Benjamin Strong and the European banks to expand their money supply as well, and in some cases, to buy pounds as reserve currencies. If they did that, then that would lower the pressure on the British pound, since gold’s price would drop everywhere else, and not just in Britain, and so the British inflation would not result in gold leaving Britain. It’s a nice idea, but completely pernicious: it would just export Britain’s problem to the rest of the world, turning the inevitable recession that Britain had been trying to avoid into a global depression. The French weren’t very willing to be swindled by Norman, but Strong was.

    The result was that Strong and the Fed expanded the money supply immensely throughout the 1920’s. To make matters worse, Irving Fischer thought that this expansion would be great, because he could “scientifically” manage the price level of the economy and coordinate production to make sure inflation wouldn’t happen. From the 1920’s, Austrians said that this could not work, that the inflation would happen anyway, that it would lead to a boom-and-bust cycle, and that it would be very harmful for the economy.

    But the American government didn’t listen to any Austrians and it continued to inflate and tried to get Fischer to manage the economy and the price level.

    The mainstream story says that there was no inflation in the 1920’s, because the price level didn’t rise. This is wrong because:
    1- The 1920’s had a lot of real economic growth which was deflationary. So expanding the money supply is inflationary, even if the price level remains more or less constant. This is another important Austrian point: inflation does not mean a rising price level; if the price level is naturally dropping due to growth, then you could have inflation with a constant price level, or even with a dropping price level (if the deflationary growth is more than the inflation).
    2- Fischer’s management of the price level was never in any way accurate or scientific, simply because it cannot be done.

    Here is where Austrian theory comes in: this inflation, as Hayek would tell us, led to a bubble. This bubble was in the stock market. Once the inflation stopped, the bubble had to burst. And that’s what happened, as the stock market collapsed.

    There is no better illustration of the superiority of the Austrian explanation than Fischer continuing to argue, up until AFTER the crash, that things were fine and the stocks will still rise. This is not the flawed mistake of a genius, this is simply the repudiation of Fischer’s fantasies of price management, and an exposition of his misunderstanding of inflation and bubbles.

    So, in the Austrian story, the crash was caused by the 1920’s inflation. Just like this current crisis was caused by the Greenspan/Bernanke inflation. And just like the 1870’s was caused by the printing of the greenbacks. This then necessitated a large corrective deflationary recession for the economy to recover from all the harmful inflationary pressure, for the malinvestments to be liquidated, for prices and wages to drop, and for production to get restarted on sober terms with prices that are not distorted by Fischer and the Fed. If the Fed were to expand credit again, as Friedman suggests, then that would simply lead to reflating the bubble, causing it to collapse again, probably in a worse way.

    This is where the neoclassical story starts. By ignoring the real cause of the crash and depression, the neoclassical story only sees the trigger (contraction) as the deep cause. Yes, NGDP did fall as the depression set in. But that is not the cause, it’s a symptom of the real cause: the inflation of the 1920’s. The real problem with the neoclassical story, then, is that it recommends, as a way of fighting the results of monetary expansion, more monetary expansion. Still, in all credit to neoclassical, at least they don’t blame it on “animal spirits”.

    So how do the Austrians explain the persistence of the Depression from 1929 till 1945? It was caused by the disastrous statist policies that Hoover (yes, Hoover) and FDR employed that prevented the recovery from ever taking place, and instead turned the economy into a centrally-planned hell-hole. Price-controls and wage-controls were among the worst of these, as were the trade tariffs, increased public spending financed by taxation, the repeal of the Gold standard, government coordination of production and consumption. If you were to institute any of these things anywhere, they would lead to massive negative economic consequences. Instituting them in America, after the stock market crash, where the economy needed to badly readjust, is what led to the Great Depression.

    The best book to read on this is Rothbard’s America’s Great Depression, available in full here:

    What do you think? How does this Austrian account compare to the conventional macro accounts?

  28. Gravatar of TGGP TGGP
    14. June 2009 at 13:53

    I am still waiting for any of the many Austrian economists who visit my blog to provide a single example of an important macro phenomenon that Austrian economics can explain but conventional macro cannot.
    Wasn’t Popper’s objection to Freud/Adler/Marx that they could explain too much? Austrians of course are going to reject Popper’s emphasis on empirical falsification, but you’re not an Austrian (although I believe you’ve also downplayed the importance of empirical evidence in economics).

    Peter Boettke has a reply to Robin Hanson’s “If not data, what?” post at

  29. Gravatar of Bill Woolsey Bill Woolsey
    15. June 2009 at 04:24


    The Mises blog is not a good source for discovering what neo-classical or Keynesian economists think.

    If you will review this blog, you will see that Scott has quite a bit to say about the Great Depression.

    For what it is worth, the downturn in 1929 started in August and the stock market crashed in October.

    The convetional monetarist explanation of the start of the depression is that the Fed restricted money growth becaue they thought that stock prices were overvalued. This started a recession. And stock prices fell.

    Whether or not that is the correct interpretation, there has also already been substantial discussion of the purported malinvestment created by a policy of price level stability in the twenties.

    I think you are mistaken in your assertions about the impact of money creation. Further, you are very much mistaken in your assertion about the ease of price adjustments.

    I remain baffled by your claims that I see money as something centrally planned.

    The process by which changes in the purchasing power of money causes its real supply to adjust to the demand to hold it is difficult and painful.

    Changes in the quantity of money that offset changes in the demand for money don’t cause disruptions. The allocation of resources generated by stable growth in nominal income isn’t “wrong” and that generated by deflation “right.” They may be different, but not wrong or right.

    Anyway, I started economic study by adopting Rothbardian dogma from top to bottom. Keep studying.

  30. Gravatar of ssumner ssumner
    15. June 2009 at 05:25

    Saifedean, I have just complete writing a book on the Great Depression, and should be revising it right now. Let me just say that you are wrong in your facts about the Great Depression. I don’t know where my Austrian commenters get the idea that Strong “expanded the money supply immensely” during the 1920s.

    Here is the data for the monetary base:

    January 1920 – $6.906 billion.
    December 1929 – $6.978 billion.

    The Austrian view seems to be based on false facts.

    The stock market boomed because the economy was doing very well in the late 1920s, arguably better than in any other period of American history.

    I agree with you on the disastrous statist policies, as do many other conservative non-Austrian economists. Indeed I have published a number of papers criticizing those policies, without ever using Austrian analysis. I am not saying the Austrians aren’t right about some things, just that I don’t see useful new ideas in their approach.

    TGGP, I agree that it is hard to falsify in economics, and that that is a problem. I do think empirical evidence (not just VAR studies) is important, but I also think that other arguments are important as well.

    I wish I had more time to follow that “data” debate. I tend to agree with the view that empirical studies rarely convince economists to change their minds. But there must be a few examples. In general the process of mind-changing is much more subtle and complex, and even I don’t understand it very well.

  31. Gravatar of saifedean saifedean
    15. June 2009 at 08:55


    I’m glad you’re still revising the book, because it seems like it needs it.

    You’re making a mistake which Rothbard dedicates 3 entire chapters (chapters 4, 5 and 6) to debunking and taking apart.

    You are looking at the monetary base, which indeed did not increase. Rothbard shows that, but then analyzes the TOTAL money supply. Surely, everyone would agree that the total money supply is a far more relevant metric here than the monetary base. The total money supply expanded from $45.30b to $73.26b in the period you mentioned””a 61.8% increase.

    It was this expansion of the TOTAL money supply that generated the excess fake credit which then caused the bubble in the stock market. As soon as the Fed began to tighten its money in 1928-9, the bubble began to unravel.

    Chapters 4, 5 and 6 in Rothbard’s America’s Great Depression are specifically dedicated to showing exactly how the monetary expansion happened, and where it happened. It frankly astonishes me you’d say there was no expansion of the money supply.

    It is true that none of this is mentioned in Keynesian and Neoclassical histories of the Depression, but that does not make this false. You can read those three short chapters and judge for yourself. The book is available in its entirety online here:

    I strongly suggest you read his book before publishing yours. It’ll take you 20 minutes to skim through these chapters online, please do it!

    From AGD, p.93:

    “Over the entire period of the boom, we find that the money supply increased by $28.0 billion, a
    61.8 percent increase over the eight-year period. This is an average annual increase of 7.7 percent, a very sizable degree of inflation. Total bank deposits increased by 51.1 percent, savings and loan shares by 224.3 percent, and net life insurance policy reserves by 113.8 percent. The major increases took place in 1922-1923, late 1924, late 1925, and late 1927. The abrupt leveling off occurred precisely when we would expect””in the first half of 1929, when bank deposits declined and the total money supply remained almost constant. To generate the business cycle, inflation must take place via loans to business, and the 1920s fit the specifications. No expansion took place in currency in circulation, which totaled $3.68 billion at the beginning, and $3.64 billion at the end, of the period. The entire monetary expansion took place in money-substitutes, which are products of credit expansion. Only a negligible amount of this expansion resulted from purchases of government securities: the vast bulk represented private loans and investments.”

  32. Gravatar of saifedean saifedean
    15. June 2009 at 09:06


    You keep talking about “changes in the quantity of money that offset changes in demand for money”. Where do these changes come from? If you believe that there should be a monetary authority, a national mint, a central bank, or any form of centralized monopoly that provides this increase, then you are arguing for central direction of money.

    My point, simply, is that if money were left without any government intervention or centralized direction it would be a spontaneously emergent self-coordinating process that does not have anything painful about it. Prices and the money supply rise and drop according to market forces and nothing necessitates any intervention on behalf of anyone.

    You argue that changes in the quantity of money (by a central bank, presumably) do not cause dislocations or disruptions. For me this is just central planning that will inevitably cause disruptions.

    They key point, I repeat, is thinking of the price mechanism as not just working through prices and money, but also working to coordinate the money supply itself. If you can make this leap in understanding, you can see that there is no need whatsoever for anything like a monetary authority anymore than there is a need for a potato authority. And you will find that any failure in the money supply and money market will be the result of this centralized direction of money””as you would expect if you had centralized direction of potatoes.

    The problem with much mainstream monetary economics is that it has made the mistake of ignoring the power of the price mechanism and free exchange to coordinate the money side of the economy. I still find Mises’s argument from 1912 to be completely convincing. On a free market, good money would emerge and it would be a spontaneously ordered process. Any argument that you could contrive to say that money is different could equally be contrived for any other good on the market to argue for the need for its central planning.

    On the GD, see my comments to Scott.

  33. Gravatar of saifedean saifedean
    15. June 2009 at 15:47


    You said: “The convetional monetarist explanation of the start of the depression is that the Fed restricted money growth becaue they thought that stock prices were overvalued. This started a recession. And stock prices fell.”

    But that would precisely be the Austrian explanation, except, again, it misses the previous and most important part of the story: the monetary expansion.

    The Austrian story is that monetary expansion causes an unsustainable boom that must end either with the destruction of the currency in a stagflationary nightmare, or with the end of the monetary expansion, which bursts the bubble and brings about a recession. So, yes, it was the Fed’s stopping the monetary growth that ’caused’ the crash, in that it triggered it, but the real root cause is the monetary expansion in the first place.

    The best way to understand monetary expansion is as a crack addiction. Once you take the first hit, you get an unnatural high, and you soon want more. From then on, if you get more crack, you get happier, but get worse withdrawal symptoms afterward; whenever you do not get crack, you suffer immensely from withdrawal symptoms. The addict will tell himself that the reason he is suffering is that he can’t get crack, and therefore would demand more crack. It would work–temporarily, but it would only mean that the next bout of withdrawal will be harder. From then on, there is no happy ending possible: whenever the addiction is stopped, it will lead to awful withdrawal symptoms. But the sooner the addiction is stopped, the better. If the addiction is continuously indulged, it will lead to death. It is the same with monetary expansion, once you do it, then you have doomed yourself to a recession that will happen once you stop. If the monetary expansion is stopped immediately, it will be a short and fast recession. The longer it is continued, the longer and harder the recession. If it is continuously indulged, it will lead to Zimbabwe. It really is that straightforward.

    If we ignore the original hit of crack, then we can make perfectly good “science” that associates withdrawal symptoms with the lack of crack hits, and therefore recommends more crack as the treatment of withdrawal symptoms. This is mainstream macro in a nutshell.

    Mainstream macro refuses to acknowledge the original monetary and fiscal expansion, and so continues to forever recommend monetary and fiscal expansion as a way of treating the ill-effects of monetary and fiscal expansion.

    Once one recognizes that the monetary expansion happened (which it clearly did, see my comment to Scott), then the neoclassical account fits perfectly within the Austrian account, though the neoclassical remains woefully incomplete.

    And the real problem with the neoclassical, of course, is that it recommends, as a solution, the monetary expansion which was the root of the problem.

    As Hayek put it: “To combat the depression by a forced credit expansion is to attempt to cure the evil by the very means which brought it about; because we are suffering from a misdirection of production, we want to create further misdirection “” a procedure which can only lead to a much more severe crisis as soon as the credit expansion comes to an end. It would not be the first experiment of this kind which has been made. We should merely be repeating, on a much larger scale, the course followed by the Federal Reserve system in 1927, an experiment which Mr. A. C. Miller, the only economist on the Federal Reserve Board and, at the same time, its oldest member, has rightly characterized as ‘the greatest and boldest operation ever undertaken by the Federal reserve system’, an operation which ‘resulted in one of the most costly errors committed by it or any other banking system in the last 75 years’. It is probably to this experiment, together with the attempts to prevent liquidation once the crisis had come, that we owe the exceptional severity and duration of the depression. We must not forget that, for the last six or eight years, monetary policy all over the world has followed the advice of the stabilizers. It is high time that their influence, which has already done harm enough, should be overthrown.”

    From Monetary Theory and the Trade Cycle

  34. Gravatar of Current Current
    16. June 2009 at 00:06

    I mostly agree with saifedean on the monetary expansion in the 1920s. This happened and it happened in money substitutes like bank accounts.

    Larry White has shown that the Scottish free banking system was capable of increasing the amount of fiduciary media in circulation without needing a corresponding increase in the gold base. Year by year the banks became better at understanding how their reserves would move and requiring less of them. And, these were banks not protected by a central bank.

    The Fed allowed for even more of an increase since it protected it’s members by being a lender of last resort.

    I think central banks and commercial banks still do this to some extent. M3 and M4 increase faster than the base. And they may increase while the base stays steady.

  35. Gravatar of Bill Woolsey Bill Woolsey
    16. June 2009 at 03:02


    Your faith in the market mechanism is touching, however, it is no substitute for actual anaylsis. Your notion that central bank monetary policy is central planning is wrong. If the auto industry was nationalized, then setting the price and quantity of cars would not amount to centrally planning the economy. Try to grasp that notion, without immediatiately thinking of all the reasons why nationalizing the car industry is a bad idea. (I think it is a bad idea, but not because it amounts to centrally planning the economy.)

    Base money production is nationalized. Setting the quantity of base money or its yield isn’t the same thing as centrally planning the economy. If nominal income were stablized, the allocation of resources would continue to be determined by the market process. Relative prices would still adjust. Profits and losses in different areas of the economy would still create incentives to produce more or less of various particular goods and services. Savings and investment would still determine real interest rates and coordinate the allocation of resources between present and future consumption. The notion that the central bank could possibly plan all of those activities or somehow impact them all by ajusting the quantity of base money, or its yield, is absurd. All that the central bank would be doing is controlling the “price” and quantity of one good–the monetary base. It is nothing like central planning of the economy.

    What it does mean is that the problems made inevitable by the existence of money become the problems of the central bank–and these problems materialize whever money is used–more or less, everywhere.

    Anyway,the concepts of the quantity of money and the demand to hold money have nothing to do with support of the status quo of a government monopoly on the issue of base money. I think your ideological concerns (to show that government should not be involved in money) are corrupting your analysis.

    I think evolved metalic monetary systems were better than barter. That doesn’t mean that price changes smoothly adjust the real quantity of money to the demand to hold it. When the nominal price of the good used as medium of account can no longer be adjusted by specialized entrepreneurs in that market (for example, how gold prices are set today,) but instead is set by changes in the prices of all the other goods in the econmy, by entrepreneurs speciallizing in all of those other specific marekts, the market for the monetary metal is less effective, which, sadly, creates problems in every other market. All our experience suggests that it works poorly.

    What is the signal that entrepreneurs thoughout the economy receive that there is a shortage of money and they need to lower their prices to increase the real quanity of money? It is lower sales at current prices. However, generally, lower sales at current prices means that their particular product is overproduced and too many resources have been devoted to its production. Those resources would be better used to produce other things. And so, lower prices are usually appropriate, but also reduced production–freeing up resources to produce other things.

    Usually, when people lose their jobs at current wages, this means that their current areas of endevor are less valuable uses of their time and they should do something else. They should find a new trade, or, less traumatically, find a more sucessful firm. Sadly, this process takes time.

    However, if the reduced sales and employment at current prices are simply a signal of an increase in the demand to hold money, and what needs to happen is an increae in the real quantity of money, all that should happen is that everyone lower their prices and wages, and pretty much continue with the same levels of production and employment. There is no need to realocate resources away from the production of everything, but that is the signal that is being received.

    Of course, there may be some shift in resource allocations. Some prices might fall more than others because people reduce their pruchases of particular goods to accumulate money. But perversely, those areas of the economy that need to expand receive lower prices too. The reverse of the usual signal. The normal process is that higher prices (including higher wages) signal a need to expand production of some particular thing. Too many resources are currently devoted to the wrong products and they should shift to producing those productes with higher prices. But, in the case where prices must fall to increase ithe real supply of money, the increase in relative prices is masked by lower money prices.

    The problem is that all the other prices in the economy are not only being used to signal the proper allocation of resources, but also the market conditions for the good used as money. It should be no surprise that this works badly, and it must have economy wide impact in order to signal the proper price adjustments in everyone else.

    All of this analysis applies both to a fiat currency and to a market evolved metatlic money. It may well be that it is just tough. Money is better than barter, and money means we have these problems.

    It could be that having the central bank try to adjust to quantity of money to meet the demand so that these price level changes are unnecessary (which doesn’t mean that relative prices don’t need to change–they do,) would cause other problems and so it is better to just leave the nominal quantity of money unchanged and suffer these problems.

    By the way, I favor free banking.

  36. Gravatar of Bill Woolsey Bill Woolsey
    16. June 2009 at 03:17


    What is M3 and M4?

    In the U.S., we have M1, M2, and MZM.

    You need to understand, that some of these measures of the money supply include items that have their own prices and yields and whose quantities are adjusted through an ordinary supply and demand process. People would want them, buy them. People who sell them, put them on offer. Shortages or surpluses lead to changes in their yields that bring quantity supplied and demanded together.

    There are good reasons to believe that changes in their equilibrium quantities may be related to the demand to hold money proper–things like currency with a fixed nominal price and no nominal yield, as well as checkable deposits that can be spent into existence by the issuer–accepted by those who don’t want to hold them except to spend them away.

    From an hyperempirical perspective–it doesn’t matter whether we count them as part of the quantity of money or else a subsitute asset that impacts the demand for some narrower, theoretically consistent conception of money. The ability to use them to predict nominal income might be better. (If for some reason, you think that the proper measure of money is the one that best predicts nominal income.) Or, to the degree that central bank policy has an impact of these quantities through some mechanism, maybe watching them and using it as a feed back for open market operations or something, might better help control nominal income. (If that is what you want to do.)

    But, I hope you can see that maybe these measures of the “money supply” are pretty pointless if you give up on the implicit assumption that the demand to hold money should be unchanging or that targetting some measure of the money supply is a sensible policy.

    Personally, I think base money is a good measure of the quantity of money. Something like M1 (currency plus checkable deposits) would be useful, but because of sweep accounts, the U.S. measure is worthless. Measured checkable deposits doesn’t include all checkable deposits, maybe no more than half. A substantial portion of checkable deposits are measured as savings accounts–maybe as much as half. But many savings accounts are not part of the medium of exhange, put perhaps a close subsitute.

    M2 is awful because it includes CDs. And, incredibly, with a break of of 100k.

    MZM isn’t bad, but it includes all savings accounts.

    My understanding is that the Fed cares nothing about measures of the money supply–they bought the Taylor rule (with subjective adjustments) hook, line, and sinker. Inteest rate targetting.

  37. Gravatar of Current Current
    16. June 2009 at 03:33

    Regarding M3 and M4 – I’m British and I’ve read some old books.

    I think you miss my point. In this instance I’m not talking about using money supply as a policy tool. I’m talking about history. I don’t think the demand to hold money is a constant either. I know some of the money substitutes come from fairly free markets.

    Earlier in the discussion Scott Sumner claimed that the 1920s boom could not be money fueled and he gave base money figures as evidence. My point is that these don’t tell the whole story.

  38. Gravatar of saifedean saifedean
    16. June 2009 at 05:10

    “If the auto industry was nationalized, then setting the price and quantity of cars would not amount to centrally planning the economy.”

    Yes, it would. And if that happened, soon enough, the auto industry would be in as bad a shape as the banking industry: failing continuously, usurping taxpayer money and irredeemably corrupt. Without a doubt, you would then look at it as a market failure, invent some model where you argue that that the reason it failed is because it wasn’t centralized but because markets are irrational or some such… and then call for more nationalization.

    I don’t have a faith in the market process. I have a very thorough opposition to any form of central planning, especially the central planning that dupes people into thinking it isn’t central planning.

    There is no need for anyone to set the quantity of cars, potatoes, or computers in the economy, because people acting on their own, through their preferences coordinated by the price mechanism takes care of that in a better way than could any car, potato, or computer czar. History is unequivocal in its validation of this. The same is true of money.

  39. Gravatar of Current Current
    16. June 2009 at 05:44


    One point Bill Woolsey is making is that interventions must be thought of separately to central planning. I agree with him here. Surely they pose similar problems, but they are not the same problems.

    The Fed does not set the price and quantity of cars. Rather, it’s actions are similar to setting the price of steel or some commodity so as to affect the price of cars. Friedman made that point using the same analogy.

    Mises says something similar in “A Critique of Interventionism” where he treats interventionism separately to nationalization. This is how we should look at the Federal Reserve system. Part of the banking industry is nationalized and intervenes in the commercial banking industry. These interventions have subsequent effects on the rest of the economy, that is the deliberate purpose of the exercise.

  40. Gravatar of Scott Sumner Scott Sumner
    16. June 2009 at 09:50

    saifedean, I use the MB definition of money, because that is the money actually created by the Fed. The Fed does not create life insurance reserves and all those other categories. And how can you say that only Rothbard looks at the broader definition? Friedman and Schwartz did the same.

    The 1927-29 expansion was the only expansion in modern history without any inflation. And you say that the Great Depression was caused by an overly expansionary monetary policy? Why aren’t decades associated with far more money creation and far more inflation followed by even Greater Depressions? The theory seems to have no explantory power. On the other hand mine does. Thus the roaring 60s weren’t followed by deflationary Fed polices, and hence there was no depression in the 1970s, just a mild recession in 1970 and a medium one in 1974. The roaring 1980s with much more inflation than the 1920s were not followed by a Great Depression of the 1990s (as one famous book had predicted.) Why, because the Great Depression was not caused by easy money in the 1920s, it was caused by tight money in the 1930s.

    BTW, Hayek later changed his mind and realized the Depression should have been cured with a more expansionary monetary policy.

    Current, A few months ago I had an argument with another Austrian who claimed M2 rose faster than the base. Not true, in recent decades the MB has grown much faster than M1 or M2. And this is even true if you exclude the unusual recent rise in the base. The relative amounts of MB and M2 are mostly determined by public preferences, although reserves requirements also play some role.

    Current and Saifedean, Let me make this distinction. The MB is the actual money created by the Fed. If you want to say it is not informative, I’ll gladly agree. But if we talk about the Fed doing something, or doing nothing, most people refer to the MB. Some Austrians want the Fed to freeze the base. My point is the Fed didn’t do that much. Now if you want to argue they should target something else like M4, that’s fine. But if we are going to move to consider options of targeting private sector aggregates, then let’s target NGDP. Indeed I think NGDP rose at about 3% during much of the 1920s, roughly the rate that Bill advocates. So I still say monetary policy was pretty sound, even if you move away from the MB.

  41. Gravatar of Current Current
    16. June 2009 at 10:34

    Scott: “A few months ago I had an argument with another Austrian who claimed M2 rose faster than the base. Not true, in recent decades the MB has grown much faster than M1 or M2. And this is even true if you exclude the unusual recent rise in the base. The relative amounts of MB and M2 are mostly determined by public preferences, although reserves requirements also play some role.”

    In this case I wasn’t talking about recent decades. Besides, although it hasn’t happened in the US other forms of money have grown faster than the base in other countries. My point is that looking only at base money doesn’t show the whole picture. I’m sure you know this in general, that generality applies here.

    Scott: “The MB is the actual money created by the Fed. If you want to say it is not informative, I’ll gladly agree. But if we talk about the Fed doing something, or doing nothing, most people refer to the MB. Some Austrians want the Fed to freeze the base. My point is the Fed didn’t do that much.”

    I don’t think that the monetary base is always informative either. The effects of the Fed go beyond the monetary base. In particular the Fed are the lender of last resort. How they perform that role or how they will perform that role can affect other things. If they indicate privately to banks that they will offer good terms then those banks will prefer to take more risk than they would have done otherwise.

    Scott: “BTW, Hayek later changed his mind and realized the Depression should have been cured with a more expansionary monetary policy.”

    Hayek thought that he had been wrong to oppose reflationary policies once the depression had begun. He disagreed with the quote saifedean gives. I think he may have been wrong. However, he never changed his view on how recessions come about from his position in the 1940s.

  42. Gravatar of saifedean saifedean
    16. June 2009 at 11:45


    You’re far more of an Austrian than you think; you just need to start reading some Austrians to realize why!

    You say: “I use the MB definition of money, because that is the money actually created by the Fed. The Fed does not create life insurance reserves and all those other categories.”
    The whole problem with central banking is that everyone and their dog can create new money that increases the money supply, while being guaranteed by the Fed. This is bad, but it is true. And it is true because all the money creation happening in the economy is effectively under the watch of the Fed, because the Fed guarantees all of that money, deposits, and money substitutes of the financial industry.

    Money created by these other means is also money and it counts towards the money supply. You can’t decide to ignore it because it doesn’t fit your theory. But your shifting of the goalposts and your moving on to other periods shows that you concede that the TOTAL money supply was increased. Good. Please make sure you mention this in your book.

    Now, please read Rothbard’s AGD to see exactly how the total money supply was increased BY THE FED. It was the Fed that was openly and by its own admission engaging in open market operations with the stated purpose of stopping the drain of gold from Europe. They bought sterling, shipped gold to France and Germany, lowered the interest rate, increased bank reserves and, in the immortal words of Strong, gave “a little coup de whiskey to the stock market.” This all increased the money supply. All of this is thoroughly detailed in AGD. That you refuse to read this book does not mean that none of this happened. That there was no rise in the price level does not mean that none of this happened, because there was deflationary economic growth going on at the same time!

    Note that if you agree the total money supply was increased, then you basically agree with the Austrian theory in everything that matters. I come now to your objections, which change nothing in the validity of the theory. You must realize that your ignorance of Austrian theory is not an argument against Austrian theory; it is an argument for you to finally pick up a book about it.

    First on inflation: The root of the business cycle is NOT inflation; it is an INCREASE IN THE MONEY SUPPLY. I’m not shifting the goalposts, this is what the theory says, and you can find this in Hayek’s books from the 1920’s. Yours is a very common misconception.

    Now, economic growth is deflationary, it leads to a drop in prices. Increasing the money supply is inflationary and leads to a rise in prices. What would happen if you had economic growth and an increase in the money supply? You would get two countervailing forces: economic growth pushes prices down, monetary expansion pushes prices up. What happens to the price level is a matter of the magnitude of the two factors, and their differing effects on different products.

    What happened in the 1920’s, therefore, was massive economic growth, along with massive monetary expansion. These two cancelled each other out to roughly keep prices constant. BUT THE MONEY SUPPLY WAS INCREASED, which is what causes the business cycle.

    Now, you move on to the other episodes of inflation, and here, I must agree with you on the kernel of truth in the neoclassical story: if the Fed had indeed expanded the money supply after the crash in 1929, there would not have been a contractionary depression. But here’s what the neoclassical story doesn’t tell you: you would’ve had massive inflation and then another crash. And then more crashes and more depressions. The problem, as I mentioned in my above comments to Bill, is that you simply ignore the original cause of expanded monetary supply; you are like a doctor treating a crack patient with more crack.

    You say: “Thus the roaring 60s weren’t followed by deflationary Fed polices, and hence there was no depression in the 1970s,”
    The 1960’s had massive expansionary monetary policy (to finance Viet Nam and the Great Society) and so they were followed by the inflationary 1970’s. Neoclassicals are right that if the Fed had contracted the money supply in the early 1970’s, it would’ve led to a massive recession. But instead, Nixon closed the window of convertibility of gold, expanded the money supply like crazy and caused a massive inflation that sent the whole world economy into a tailspin. (Most amusing, of course, were the mountains of economics papers that then blamed this inflation on oil embargoes, crop failures, cost-push inflation and other such nonsense! But I digress.)

    Remember, once the initial monetary expansion happens, there is no happy ending. Either it is stopped early, and a small recession ensues; or it is stopped late; and a massive recession ensues; or it is not stopped and Zimbabew ensues. As Hayek says, holding a tiger by the tail cannot end well.

    You say: “The roaring 1980s with much more inflation than the 1920s were not followed by a Great Depression of the 1990s”
    You continue to confuse monetary expansion with inflation. The 1980’s inflation was a result of all the massive monetary expansion of the 1970’s. They key point to understand is that when the monetary supply is first expanded, it goes into a specific bubble(s) that then burst causing a recession. Then, there is pressure to withdraw money, if it were to be met with more expansion as the neoclassicals recommend, you would then get that money spreading to the rest of the economy and causing inflation. That is what happened in the 1980’s. If it were to not be met with more expansion, it would cause a recession, but then there can be recovery. So the 1970’s was one big bad expansion of money that begat more expansion of money. When Volcker finally pushed the breaks and stopped increasing the money supply, the economy had the recession it needed, and then recovery started on a healthy, relatively non-inflationary track.

    “BTW, Hayek later changed his mind and realized the Depression should have been cured with a more expansionary monetary policy.”
    The problem here is that we can get bogged down into endless debates about what Hayek believed and what he thought politically tractable. As Current says, Hayek never changed his mind on the fact that recessions are caused by monetary expansion””that is Hayek’s life’s work after all. What is to be done after that recession has happened is something that’s not easy to pin down as Hayek was possibly changing his mind, but more likely, trying to appeal to politically feasible solutions. Ideally, Hayek would want government out of the money and banking business entirely. You can argue with the whole idea of him offering advice to an institution he does not believe should exist, and you would be on the same page with Rothbard. But you can’t continue to use Hayek’s politically-calculated messages to bludgeon his academic arguments.

    Anyways, let’s have our arguments over ideas, not authorities!

  43. Gravatar of Current Current
    17. June 2009 at 04:19

    Saifedean:”Now, economic growth is deflationary, it leads to a drop in prices. Increasing the money supply is inflationary and leads to a rise in prices. What would happen if you had economic growth and an increase in the money supply? You would get two countervailing forces: economic growth pushes prices down, monetary expansion pushes prices up. What happens to the price level is a matter of the magnitude of the two factors, and their differing effects on different products.”

    That is true. However there may be another explanation. The “real” aggregate figures that we derive from the nominal figures may not be meaningful.

    The output GDP equation is:
    GDP = C + I + G + (X − M)

    The income GDP equation is:
    GDP = R + I + P + W
    where R = rents
    I = interests
    P = profits
    W = wages

    When nominal GDP rises it is deflated to make real GDP. However, this process cannot take the artificial boom that Austrian economists talk about into account.

    Looked at from the point of view of the outputs equation we don’t know if the value of the investments I is what it appears to be. Recent experience from before monetary expansion or from when the rate of expansion was lower can’t deflate it.

    Also, that investment I is sunk into the capital aggregate K. This way of thinking allows no place for the structure of production. It doesn’t tell us if it was sunk into a sustainable or unsustainable production decision.

    Looked at from the point of view of the incomes equation we don’t know if the interest rate accurately represents time-preference decisions.

    Hayek and Von Mises both made this point, though I can’t remember where.

    Also, folks should understand that there are two sorts of ABCT theorists. Those who think reflation is a good idea and those who don’t. Of the modern ones Garrison, Horowitz, White & Selgin fall into the former category. Hulsmann, DeSoto, Hoppe and Salerno into the latter category.

  44. Gravatar of ssumner ssumner
    17. June 2009 at 05:14

    Current, I basically agree with you first comment here,

    Saifedean, I have read a good bit of Rothbard, and am not impressed. His theories don’t predict the post-war period well at all. We should have had multiple Great Contractions by now. Where are the post-war Great Contractions? Why hasn’t NGDP fallen 50% after each inflationary boom.

    BTW, The Fed does not guarantee private “money,” FDIC does. And we didn’t have that “problem” in the 1920s, nobody guaranteed bank deposits. That’s why 1000s of banks failed, and depositors lost their money.

    You seem to assume that in 1929 the Fed had no choice other than massive depression or hyperinflation. How about continuing to have 3% NGDP growth and stable prices, why not a happy medium?

    I do agree with you that the 1970s inflation was monetary.

    You describe the 1980s as “a healthy non-inflationary track” This is odd. Inflation was much higher in the 1980s than the 1920s, and so was money growth.

  45. Gravatar of Current Current
    17. June 2009 at 06:54

    We’re now reaching a place in debate where it’s different to talk about the view “of Austrian Economists” since they hold different views.

    Rothbard, DeSoto, Hoppe, Block and Hulsmann sit on one side the “Full Reserve Free bankers”. Selgin, White, Garrison, Horowitz and Hayek on the other the “Fractional Reserve Free bankers”. (Everyone is forever arguing about what Mises thought).

    The former group hold the view that Scott and (I think) Saifedean have read. Monetary expansion always leads to misallocation, a bubble and a crash. The degree of substitutability affects this. Only if there is a homogeneous capital good would there be no effect. However more closely substitutable capital goods lessens the effect.

    The latter group are more pragmatic. They hold the point about substitution. Also they think that monetary expansion may not cause a bubble and subsequent crash. They hold to a sort of monetary equilibrium theory. If the populace become more worried and hold more money then issuing more money will not necessarily cause price inflation. If the uncertainty is removed then banks will act to reduce the money stock.

    Some of both groups hold that a part of expectations theory is relevant. That can be used as an explanation for what happened in the 80s. Businesses in high-order goods industries may over time come to ignore sudden spikes in their business, they may learn not to respond to them.

    The problem with all of these theories is that the aggregate quantities are not so useful when talking about them. So, finding aggregate evidence is very difficult. The evidence Austrians point to (correctly in my view) is that the place to look for evidence is on the local level. The macroeconomic theory must be built up from microeconomics.

    Also, on this bit: “BTW, The Fed does not guarantee private “money,” FDIC does. And we didn’t have that “problem” in the 1920s, nobody guaranteed bank deposits. That’s why 1000s of banks failed, and depositors lost their money.”

    Saifedean is wrong that the Fed gaurantees private money subsitutes. But it does have an effect on this question.

    The Fed does not protect a bank from becoming bankrupt by making bad loans or other bad contracts. However it does provide a degree of protection that was not present before, it is the lender of last resort. Normal businesses must be able to make a profit and meet their balance sheet commitments. That was the case with commercial banks too before central banks. The central banks though provided a bank with a line of credit if it cannot obtain one elsewhere. It will protect it from balance sheet insolvency. Banks that would have been chucked out of the clearing system in the past may not once a central bank is in operation.

  46. Gravatar of Greg Ransom Greg Ransom
    17. June 2009 at 07:56

    Scott — In the early 30’s Hayek mostly merely argued against public works and other “expansionary” fiscal policies, and his arguments were directed at the British situation, a situation which had existed since Churchill returned Britain to the gold standard at the pre-war parity, instantly overpricing British labor in the international market.

    Hayek always acknowledged that inflation would have been a good political solution to the problem created by Churchill — with economic costs.

    I’ve seen no evidence that Hayek ever knew the empirical facts on the ground in America — and lets be honest, it wasn’t until the 1960s that even Americans began to get a clearer picture of the empirical facts on the ground.

    And Hayek _never_ endorsed the “real bills” doctrine that ruled the roost at the U.S. Fed.

    Scott wrote:

    “Hayek originally agreed with those who opposed monetary stimulus during the Great Depression. That policy error led directly to WWII. Later Hayek realized that he had been wrong, and argued that monetary policy should have been stimulative, to offset the fall in velocity.”

  47. Gravatar of saifedean saifedean
    17. June 2009 at 09:30


    During this thread, you have continued to shift the goalposts of your objections while maintaining your opposition to ABCT””even as I have shown you how each of your objections is factually wrong. First, you wanted “a single example of an important macro phenomenon that Austrian economics can explain but conventional macro cannot.” I gave you the Great Depression, and showed you how it is better explained using ABCT. You then objected because you said the money supply wasn’t increased in the 1920’s, citing the monetary base. I showed you that it was, citing the total money supply, which is surely more relevant. You then objected on the grounds that there was no price inflation, and I showed you that that’s irrelevant, as ABCT is not predicated on price inflation; it’s predicated on increased money supply. You then moved the goal-posts to discuss post-WWII examples. I went through these and gave you how they fit with the Austrian story, and now you come to me with the incredible line that for some reason ABCT must predicate contractions in the post-WWII world, and that since these contractions don’t exist, then ABCT is wrong. That objection, obviously, comes in no way to bear on the example of the Great Depression, but it, too, is a false objection, and here’s why:

    There is nothing in ABCT that necessarily predicts contractions””much less in the post-wars period. What ABCT says, as would mainstream macro, is that if after a bubble bursts the Fed doesn’t reflate, then you will have a contractionary and deflationary recession. But, ABCT also says that if the Fed reflates you’ll get inflation and more recessions in the future. Since WWII, the Fed has had absolutely no problem continuously inflating and reflating and re-reflating. So we have hardly had any contractions, and the ones we had were relatively short-lived. But we’ve had an enormous amount of inflation… as ABCT would predict (and monetarists wouldn’t)! The dollar has lost most of its value since the early 1970’s, and 95% of its value since 1914. That’s very inflationary, and it strikes me as conforming perfectly with what would happen to a currency issued by a central bank that has no hang-ups over using the printing presses, don’t you think?

    Which brings me to your “happy medium” argument. I don’t think that a happy medium exists, simply because the medium between bad and very bad is bad””not happy. Unfortunately, there is no free lunch. In spite of all that we’ve learned in macro, opportunity costs are real in government spending. So, monetary expansion will cause an artificial boom. Stopping the expansion will cause the boom to collapse with a recession to follow. Continuing expansion will make the boom bigger, and the collapse harder. Continuing forever will lead to Zimbabwe. Reflating is simply consolidating the monetary expansion of the boom into massive overall inflation. Reflation and a slow expansion of money supply is not a happy medium, it’s just a slower descent into the inevitable collapse. The solution is not this happy medium; the solution is not to inflate in the first place, so as to not have to reflate. The answer to crack addiction is not controlled a happy medium of controlled crack injections””it is to stop crack.

    As for the 1980’s, you’re right, there was money growth, but by the standards of the 1960s-70s, the 80’s were like a gold standard””Volcker knew how to (almost) cut off the crack. And the inflation in the 1980’s was really the unraveling of the oil, food and other bubbles that were generated by the 1970’s monetary expansion.

    Now, I want you to ask yourself a question: if your rejection of ABCT is constant, while your arguments for this rejection continue to shift, doesn’t that tell you that perhaps you’re being a tad unfair? If you were to consider ABCT and mainstream explanations equally, wouldn’t you be finding the ABCT explanations more plausible?

  48. Gravatar of saifedean saifedean
    17. June 2009 at 09:32


    Thanks a lot for your very helpful comments. I am aware of the distinction between the two groups, though obviously not as knowledgeable as you are about it. I am new to this whole Austrian business and have been mostly looking into ABCT, because I was fascinated by how useful it is to understand the current crisis””in contrast to mainstream macro, which, to put it charitably, isn’t. But I am particularly interested in seeing how any of those authors would deal with this question in light of expectations theory. Have any of them incorporated (or critiqued) Lucas’ work in any way? If you have any recommendations, please do share.

    As for the FDIC/Fed; a guarantee by FDIC is a guarantee by the Fed; there’s really no distinction in practice. The FDIC has nowhere near enough money to guarantee all the banks it guarantees, but the reason that fractional reserve banking doesn’t collapse into a giant nationwide bank-run is simply that the FDIC has immediate and complete access to Bernanke’s “technology called the printing press”. If this did not exist, no one would trust the FDIC, because the FDIC has nowhere near enough money to guarantee the banking sector.

    This current crisis has clearly shown that when push comes to shove, almost all implicit and explicit guarantees will be honored, and then some. That the Fed is injecting liquidity into Credit Default Swaps, of all things, tells you all you need to know. This, unfortunately, is a feature, not a bug, of having a lender of last resort that prints its own currency. Though I bet you Bagehot didn’t foresee CDS’s being bought by the Fed when he wrote Lombard Street!

  49. Gravatar of Current Current
    18. June 2009 at 03:36

    Saifedean: “But I am particularly interested in seeing how any of those authors would deal with this question in light of expectations theory. Have any of them incorporated (or critiqued) Lucas’ work in any way? If you have any recommendations, please do share.”

    They have, and I’ve read some of this. I can’t remember where though. I’ll reply in a couple of days when I’ve found that out.

  50. Gravatar of ssumner ssumner
    18. June 2009 at 06:36

    Greg, I have heard the argument about Hayek not knowing the facts, and have never understood it. What is there to know? Maybe there were no NGDP statistics, but everyone knew there was:

    1. Massive price deflation
    2. Massive output decline

    Wasn’t it obvious that NGDP plunged? The NYT was full of stories in the early 1930s that national income was estimated to have fallen in half. I can’t believe that Hayek didn’t even know we were in a depression. Did he think it was a supply shock? Then why deflation? What was his explanation for the massive drop in output?

    Fisher looked at the very same data, and correctly understood exactly what was going on, and what the Fed should have been doing. Had they followed Fisher’s advice there would have been no Great Depression.

    Saifedean, I haven’t moved the goal posts once. I have asked for a macro fact that Austrians explain better than conventional models, and you haven’t provided one. The conventional explanation of the GD is that policymakers let AD fall, that money was too tight. This explanation also explains the milder post-war recessions, in other words it is not ad hoc, like the Austrian explanation. You say easy money caused the GD, and then you say easy money explains why we haven’t had a depression since WW2. So easy money explains pretty much anything in your book. And I still have no idea what you are saying about the 1980s. Money was much easier in the 1980s than the 1920s. Sure it was looser than the 1970s, but so what?

    I am still waiting for something that conventional theory doesn’t explain as well as Austrian. Conventional theory tells why the Great Depression started in late 1929. Does Austrian theory tell us why that date was significant? All the Austrians can say is that for some reason the bubble finally burst in late 1929. Conventional theory gives a reason—tight money was adopted by the world’s major central banks in late 1929. That’s a far more powerful explanation.

    You are wrong about the FDIC, it is not backstopped by the Fed, it is backstopped by the Treasury. We saw that in 1990, when FSLIC went bankrupt. The Treasury rescued it, not the Fed.

    BTW, Next to the base, the M1 money supply is the most famous, and it only grew by about 13% during the 1920s, so the per capita M1 actually was stable. Your argument for easy money rests on rapid growth in other assets, beyond M1. The Fed has little or no direct control over those other assets. If they had sharply reduced the monetary base we would have had a depression in the 1920s, and other assets wouldn’t have grown as much, but a much better policy would have been price stability as Irving Fisher had advocated, or even better the 3% nominal GDP target favored by Bill Woolsey. That would have prevented the long run inflation you are so worried about, and also prevented the GD. I’ll take that over Rothbard’s deflationary views.

  51. Gravatar of Current Current
    18. June 2009 at 08:29

    In the more advanced Austrian theories there are explanations for what happened in the other times that Scott mentioned. However, I don’t think here is a place to go into them. It would take a lot of words, I’ve forgotten some of them and probably don’t properly understand others.

    Like most macroeconomic theories when we get into these more complicated theories things become overidentified. Just as a more advanced Austrian theory can explain lots of stuff so can more advanced theories of other sorts.

    However, I still think that Saifedean’s point about the Great Depression has merit. Scott points to monetary tightenings suggesting that they brought the GD about. I agree that after 1929 many silly things were done with money that had destructive effects.

    I don’t think though that the initial events of the Great Depression can be explained well by any other theory than ABCT. Scott, do you think that if tightening had not occurred then the boom could have just continued indefinitely? Why is it that after large and long booms economies become sensitive to every little change in inflation and a money supply? I’ve never read a satisfactory answer to this from a mainstream economist.

    To put the point another way…. You like most mainstream economist seem to use aggregate capital. The relative prices of portions of it don’t make a macroeconomic difference. In that case why not go for broke? Why target NGDP at 3% and run the risk that people will think you’ll miss. Why not target 300%?

  52. Gravatar of ssumner ssumner
    19. June 2009 at 04:32

    Current, Yes, I think the boom could have continued. Are there any depressions where nominal GDP rises at a steady rate? The closest example might be the 1974 oil shock. But there was no supply shock in 1929. If NGDP had kept growing, a Depression could only have occurred if output fell sharply as prices rose sharply. But without supply shocks that almost never occurs. So again, I think steadily rising NGDP would have prevented the Depression. Stock investors thought the same thing. So did many distinguished economists at the time.

    The second question is whether central banks can stabilize NGDP growth. I believe they can, or more specifically, I think they can stabilize expected NGDP growth, which is almost as good.

Leave a Reply