Some of the pages at Caplan’s personal site are now annoying to read, and it’s split up into three parts, so I’ve complied it all into one page here.
(Originally written by Bryan Caplan to Tyler Cowen.)
Hi. I just finished reading the third volume of Schlesinger’s _Age of Roosevelt_ (I think he gave up on later volumes, so I guess I’m done!). And then I re-read Rothbard’s _America’s Great Depression_ while my regressions were running.
AGD now seems to me to be a mixed bag. On the one hand, the theory chapters at the opening are very weak. It seems to me that he could have written a much clearer book if he just clearly explained the classical view that unemployment is caused by excessive real wages and used _that_ as the theoretical underpinning for the rest of the work. To his great credit, he did explain this point, but it deserved much more attention.
Instead he focused on the Austrian theory of the business cycle, which makes less and less sense the more I think about it. I think Jeff Hummel wrote a critique where he said that it is really a theory of sectoral shifts rather than depression. If, as in the Austrian theory, initial consumption/investment preferences “re-assert themselves,” why don’t the consumption goods industries enjoy a huge boom? After all, if the capital goods factors have had their prices bid too high, have not the consumption goods factors had their prices bid too low?
Moreover, I can’t figure out why Rothbard thinks businessmen are so incompetent at forecasting government policy. He credits them with entrepreneurial foresight about all market-generated conditions, but curiously finds them unable to forecast government policy, or even to avoid falling prey to simple accounting illusions generated by inflation and deflation. Even if simple businessmen just use current market interest rates in a completely robotic way, why doesn’t arbitrage by the credit-market insiders make long-term interest rates a reasonable prediction of actual policies? The problem is supposed to be that businessmen just look at current interest rates, figure out the PDV of possible investments, and due to artificially low interest rates (which can’t persist foreover) they wind up making malinvestments. But why couldn’t they just use the money market’s long-term interest rates for forecasting profitability instead of stupidly looking at current short-term rates?
I also can’t overlook his bizarre claim that inflation prevents the dispersion of productivity gains throughout the economy. It may lead to a _different_ dispersion that would otherwise occur, but that is a very different matter.
On the other hand, I still think his historical treatment of the Hoover presidency is just great. And interestingly, all of the specific facts Schlesinger describes support Rothbard’s interpretation of Hoover as proto-New-Dealer, which is why Schlesinger is forced to “interpret” the facts such a large proportion of the time rather than simply stating them.
I think you wrote a Critical Review piece in which you criticized Rothbard’s emphasis on Hoover’s efforts to keep up wages. But I still think it makes a lot of sense. Actually, reading Rothbard in tandem with Friedman’s monetary history gives two surprisingly consistent perspectives. Rothbard shows the harm of wage rigidity, while Friedman shows the effects of monetary contraction given wage rigidity. Rothbard’s point is really necessary for Friedman’s to make sense, because a massive monetary contraction and velocity decline created only a brief recession in 1920-21 (as Friedman himself notes in a somewhat puzzled manner).
And while Rothbard grossly underrates the negative effects of the monetary contraction given wage rigidity, Friedman seems to seriously underestimate the extent to which the Fed _did_ try to counter-act the monetary contraction.
I’m embroiled in my empirical work now. It’s coming along, and seems like it may produce some interesting results. I’ll keep you informed.
From email@example.com Mon Jul 1 19:49:35 1996
Received: from mail5 (firstname.lastname@example.org [18.104.22.168]) by ponyexpress.Princeton.EDU (8.6.12/8.6.12) with ESMTP id TAA04471 for ; Mon, 1 Jul 1996 19:49:33 -0400
Received: from ernie.ucop.edu (ernie.ucop.edu [22.214.171.124]) by mail5 (8.6.13/Netcom)
id QAA15782; Mon, 1 Jul 1996 16:47:14 -0700
Received: from AHUPUAA (ahupuaa.ucop.edu [126.96.36.199]) by ernie.ucop.edu (8.6.11/8.6.11) with SMTP id QAA33287 for ; Mon, 1 Jul 1996 16:46:55 -0700
Date: Mon, 1 Jul 1996 16:46:55 -0700
Content-Type: text/plain; charset=”us-ascii”
Subject: Re: Rothbard’s _America’s Great Depression_
Bryan Caplan wrote:
>Instead he focused on the Austrian theory of the business cycle, which
>makes less and less sense the more I think about it. I think Jeff
>Hummel wrote a critique where he said that it is really a theory of
>sectoral shifts rather than depression. If, as in the Austrian theory,
>initial consumption/investment preferences “re-assert themselves,”
>why don’t the consumption goods industries enjoy a huge boom?
As a matter of fact they do, if by “boom” in consumer industries you mean an increase in consumer demand relative to demand for producer’s goods. When the crunch comes, the prices of consumer goods do rise relative to prices of producer goods. In the 1929-type of situation, where the crunch is coupled with a sharp decline in the money supply, both consumer and producer prices decline in absolute terms, but the decline in capital prices is much sharper than for consumer prices (In AGD, Rothbard does document this phenomena occuring during the Great Depression). Under the fiat money system we have now, the money supply is not allowed to fall as the credit contracts, and can even increase while the depression is underway. Thus, it is possible for a depression to take place even in the midst of sharp increases in consumer prices (as in the stagflations of the 1970s). No non-Austrian theory even begins to coherently explain the existence of inflationary depressions, which is part of the reason why Austrianism regained intellectual respectability during the 1970s.
The key point to keep in mind here is that it is the relative, not absolute, demands for producer and consumer goods that are associated with the trade cycle phenomena. A “boom” or a “bust” is indeed a sectoral shift where the capital structure first deviates from the market time preferences and then returns. The empirical data for the U.S. suggests that booms and busts are not uniform economy-wide, but are concentrated specifically in the most capital-intensive sectors of the economy precisely as Austrian theory predicts. Any attempt to define “depression” apart from sectoral shifts misses one of the most crucial aspects of real-world trade cycles.
The apparent asymmetry between booms and busts is that in the short-run most of us like booms a lot better than we like busts. From the point-of-view of the typical wage worker, it’s rather nice to have employers suddenly flush with cheap credit bid more for your services. The harmful effects of diverting capital and labor to more time-consuming lines of production become evident only at a later point in time, when consumer goods seem to be getting more expensive and/or investments seem to be earning paltry rates of return and/or rates of loan defaults start to go up. As the errors of the inflationary malinvestment are revealed in the marketplace, entrepreneurs shift their diminished demand for producer goods back to the shorter production processes. For this to occur, the capital-intensive industries that previously bid up wages, initiated lengthy/higher stage production processes, etc. must now cut their losses by dismissing their workers and perhaps abandoning some of their capital. There is a real cost associated with the waste of capital and productive effort on longer-term production projects at the expense of more highly desired shorter-term projects. These costs cannot be avoided by “fine tuning” or “soft-landings” or other manipulations of money and credit by the Fed Open Market Committee; sooner or later the malinvestments have to be accounted for. Sectoral shifts always have a cost, and shifts caused by money and credit manipulations create severe economy-wide costs. Unlike shifts induced by market forces, however, the widespread shifts induced by an inflationary credit expansion and the subsequent correction have no offsetting benefits in terms of increasing long-term productivity.
The trade cycle is by no means the only factor in explaining the Great Depression, but in fairness to Rothbard it should be acknowledged that he does go into great detail explaining how other government policies of the time served to severely aggravate and prolong the economic downturn. Indeed, AGD is as much an indictment of non-monetary forms of interventionism as it is of central bank manipulation of credit. Rothbard argues at length in AGD, for example, that pre-1930s downturns did not have such disasterous consequences as the 1929 crash because of the anti-Laissez Faire policies introduced by Hoover and other statist Republicans in response to the crash. The theoretical focus on the trade cycle theory in the early part of the book is primarily a means for explaining how such an economic calamity could have occured been initiated at the end of the apparently stable and productive 1920s, but is by no means the sole focus of Rothbard’s analysis.
>Moreover, I can’t figure out why Rothbard thinks businessmen are so
>incompetent at forecasting government policy. He credits them with
>entrepreneurial foresight about all market-generated conditions, but
>curiously finds them unable to forecast government policy, or even
>to avoid falling prey to simple accounting illusions generated by
>inflation and deflation. Even if simple businessmen just use
>current market interest rates in a completely robotic way, why doesn’t
>arbitrage by the credit-market insiders make long-term interest
>rates a reasonable prediction of actual policies? The problem is
>supposed to be that businessmen just look at current interest rates,
>figure out the PDV of possible investments, and due to artificially
>low interest rates (which can’t persist foreover) they wind up
>making malinvestments. But why couldn’t they just use the money
>market’s long-term interest rates for forecasting profitability instead
>of stupidly looking at current short-term rates?
Four reasons. First, the credit expansion affects all interest rates, whether they be long or short term, so arbitrage between them doesn’t “predict” what the inflation premium should be. Arbitrage simply takes some of the funds that have been injected into the short term market (or the Treasury bond market, or the international bond market, or the Warburg trade acceptances, or the Florida real estate market, etc.), and redistributes them to other types of investments. Second, the mind of a Benjamin Strong or an Alan Greenspan is not a predictable phenomenon. Making something as important as money and credit dependent on an elite few renders investor anticipations of future monetary phenomena highly uncertain and vulnerable to deliberate attempts by the insiders to fool the markets (in AGD, Rothbard presents evidence of purely political motives and even outright corruption as spurring the expansionary credit policy of the 1920s).
Third, credit markets and money are so pervasive throughout the economy that any cluster of entrepreneurial errors are bound to be far more serious here than anywhere else. It is precisely the robotic investor who only watches interest rates and doesn’t pay attention to the possibility of a credit contraction harming his investments who gets burned the worst by a credit crunch. How could a passive price-taker know when interest rates were too high or too low, unless he could independently estimate what the equilibrium rates should be (including a calculation of the default and inflation risks)? Fourth, few investors (or economists for that matter) understand the trade cycle well enough to profit from it. Presumably, a cluster of errors could be avoided *if* many investors were fearful of artificial credit expansions *and* they had timely information as to when they were taking place. If you read the financial press, however, you’ll find that such consciousness of credit and monetary issues is lamentably missing and really useful information on the activities of the Fed absent. Even with direct access to a financial newspaper wire service with an archival search engine (such as I have), one has to look awfully hard to find any useful information for an Austrian trade cycle theorist to take advantage of. Likewise, Austrian trade cycle theory is not exactly standard fare in Econ. 101 at most universities.
The Fed does publish its statistics (yes, I’ve used their web pages also), but what use are they? Not only are the numbers not very timely, one has to be a real expert to make any sense out of them. Their numbers obfuscate critical information about bank reserves and credit and about components of the money supply figures (which is necessary to eliminate the credit instruments that clutter up M2 and get at the real money supply numbers). Moreover, Fed numbers say nothing about overseas banks and other foreign holders of dollars and dollar-denominated accounts. When you consider that the Bank of Japan is holding 10% of the Federal debt and has something like $100 billion in U.S. dollar accounts, and the various major Japanese banks control far more assets than banks of other industrialized countries, one can begin to see that the lack of information on foreign holdings is fatal to any worthwhile analysis of the monetary situation using traditional Fed data.
It is also worth remembering that investors of the 1920s had far less macroeconomic information available to them than they do today. Ironically, it is the government’s mania for collecting statistics beginning in the New Deal era that has made the money illusion less effective because of the “rational expectations” effect. As lousy as some of the government numbers are, investors (particularly since the big dollar depreciation of the 1970s) watch them like hawks and thus shorten the time horizon for government exploitation of the people via money supply increases. Of course, it shouldn’t come as a suprise if the government were to try taking advantage of the numbers junkies by cooking the numbers. The recent attempt to emasculate the consumer price index such serve as a cautionary warning not to blindly trust government statisticians.
>I also can’t overlook his bizarre claim that inflation prevents
>the dispersion of productivity gains throughout the economy. It
>may lead to a _different_ dispersion that would otherwise occur,
>but that is a very different matter.
I don’t recall Rothbard saying that. I recall him saying that productivity gains would normally result in falling prices for consumer goods, and that an attempt to stabilize consumer prices via inflation would offset at least some of the productivity gains by causing the malinvestment of capital and other capital consumption effects.
>And while Rothbard grossly underrates the negative effects of the
>monetary contraction given wage rigidity, Friedman seems to
>seriously underestimate the extent to which the Fed _did_ try
>to counter-act the monetary contraction.
Rothbard fully acknowledges the problems caused by inflexible wages in AGD. He goes through the calculation of how real wage levels were increasing through the early part of the Depression and even goes into an extended discussion of how various stabilizationists and industrialists foolishly advocated wage rigidity as an anti-Depression measure. But linking the problems of wage rigidity solely to monetary contraction is not necessarily a valid interpretation of the Great Depression in that the money supply didn’t contract until the depression was well underway. The Friedmanite position fails to consider that the monetary contraction was *preceded* in time by the downward pressures on the credit markets and the upsurge in unemployment. Indeed, there were some parts of the economy (notably in the agricultural sector) in deep trouble well before the big stock market crash. To fully explain this sequence of events, the Austrian trade cycle theory is indispensible.
A Friedmanite would typically argue that the Fed should have done more to counteract the contraction, but they fail to grasp that the contraction and the eventual gold devaluation and demonetization of 1933 was the result of a failure of the credit system that was backing the bank and thrift accounts. People didn’t repay their loans because the loans were unproductive (having been used to purchase producer goods that were malinvested), not because there was suddenly no money in circulation to enable borrowers to repay the loans. Only at a later time did the deflation begin to significantly compound the credit defaults via reductions in borrower income. But what caused the initial wave of defaults and bank failures? It is putting the cart before the horse to say that the monetary contraction caused the credit collapse.
From email@example.com Mon Jul 1 17:00:22 1996
Received: from mail.netcom.com (firstname.lastname@example.org [188.8.131.52]) by ponyexpress.Princeton.EDU (8.6.12/8.6.12) with ESMTP id RAA00284 for ; Mon, 1 Jul 1996 17:00:20 -0400
Received: from liberty.uc.wlu.edu (liberty.uc.wlu.edu [184.108.40.206]) by mail.netcom.com (8.6.13/Netcom)
id NAA17994; Mon, 1 Jul 1996 13:52:57 -0700
Received: by liberty.uc.wlu.edu
(220.127.116.11/16.2) id AA135714431; Mon, 1 Jul 1996 16:53:51 -0400
Date: Mon, 1 Jul 1996 16:53:51 -0400 (EDT)
From: An Claidheamh Ceilteach
Subject: Re: Rothbard’s AGD
Content-Type: TEXT/PLAIN; charset=US-ASCII
To Bryan Caplan (mostly)–
You seem to be denying that the Fed causes systematic malinvestment, and claiming rather that only sudden, unexpected changes in policy can hurt the economy. Historically, however, the business cycle has been systematic and regular, implying that whatever policies are the cause have not been sudden erratic shifts but rather continuous, institutionalized (if you will) interference with the market.
Doesn’t your theory rely on the notion that the Fed’s interest rates correlate to the hypothetical market interest rate? For example, if money demand goes up, the interest rate will rise. But interest rates, as a form of price control, are a very blunt weapon. Sometimes money demand will not go up, but cost-push inflation occurs, and the Fed foolishly raises interest rates to forestall the inflation. The result is a surplus of funds for lending, because interest rates are above the market level. Don’t surpluses and shortages of money have the potential to cause serious adjustment problems in the economy?
Or maybe I was confusing your position on the economy as a whole with your position on the stock market only?
“So go ahead and kneel or bow
I will choose to stand, to be my own
Until the end”
From email@example.com Tue Jul 2 04:08:05 1996
Received: from netcom5.netcom.com (firstname.lastname@example.org [18.104.22.168]) by ponyexpress.Princeton.EDU (8.6.12/8.6.12) with ESMTP id EAA15005 for ; Tue, 2 Jul 1996 04:08:04 -0400
Received: (from timstarr@localhost) by netcom5.netcom.com (8.6.13/Netcom)
id BAA01138; Tue, 2 Jul 1996 01:07:14 -0700
Date: Tue, 2 Jul 1996 01:07:14 -0700
From: email@example.com (Tim Starr)
Subject: Caplan on Business Cycles
>From: “Bryan D. Caplan”
>Subject: Rothbard’s AGD
>I don’t think I want to get into a further discussion of the stock market
>right now. I will however point out the two big problems with Tim’s
>reply on Rothbard’s _America’s Great Depression_.
>Your remark about entrepreneurs’ time preference leading them to take
>advantage of a credit surplus against their long-term interests simply
>makes no sense at all.
Then we’re even, because your comments thus far have made no sense at all to me.
>Someone who can borrow and lend has no reason
>to take account of anyone’s time preference but the marginal time
>preference expressed in interest rates.
That’s nice, but completely irrelevant because most entrepreneurs don’t have the ability to borrow or lend. They’re usually one or the other. With a central bank that lends at interest rates below the market-clearing level, other lenders can either stick to the market-clearing level, in which case they won’t lend anything out ’cause borrowers will prefer the central bank’s loans, or they can match the central bank. In the latter case, they’ll do so even if they know that the low rates aren’t sustainable in the long run. Same goes for borrowers.
You seem to have a very strange view of how business operates. Imagine a corporate officer in a meeting trying to defend himself for refusing to borrow at the below-market rates offered by the Fed instead of the at-market rates offered by some private lender. He’d get removed in an instant. Or imagine if he was a lender trying to defend himself for refusing to lend at the below-market rates offered by the Fed, sticking to much higher rates. He wouldn’t be generating any assets, because he wouldn’t be making any loans. How do you expect him to keep his job?
“Yes, I know that my department’s been doing nothing, but that’s because the Fed lowered interest rates too much. We’ll just have to wait for it to raise ’em again, then we’ll start making some money again.”
>You make another remark which would be valid in combination with a number
>of extra assumptions:
>When the bust comes, capital shortages are caused because the
>capital that would’ve been available was malinvested in capital goods
>which have to be liquidated, & the consumer goods industry can’t boom
>because the workers laid off from the capital goods industries don’t have
>any more income to spend on consumer goods & also because the malinvestment
>prevented capital from being allocated to consumer goods for which there
>would’ve been real demand.
>In particular, the problem of laid-off workers would make sense if there
>were real wage rigidity.
Nope. It makes sense as long as wage level corrections take time. For as long as it takes for wage levels to correct themselves, there will be surplus labor (unemployment). Wage level corrections aren’t instantaneous.
>The problem of laid-off workers lacking income would make sense if the
>still-employed workers did not receive a compensating _increase_ in
Why should “still-employed workers” get any “compensating increase” in income in the bust part of the business cycle? The whole reason for the bust in the first place is that there’s been malinvestment in capital goods, with the corresponding misallocation of wages to labor. But, of course, you don’t believe in malinvestment. Sorry, I forgot.
>…This makes sense only if there has been an increase in money
>demand (which may be identified with a decline in money velocity).
Or a decrease in money supply, which is what the bust part of the business cycle is all about.
Why am I explaining elementary econ concepts to a PhD econ student?
>Of course, a combination of real-wage ridigity and an increase in money
>demand are themselves _sufficient_ to generate a depression.
Translation: a decrease in money supply will lead to above-market wages, which will lead to unemployment, which will lead to the need for a downwards wage level correction until market-clearing levels are reached again. If this downwards correction is prevented, then you’ll have unemployment that won’t go away. Congratulations! You’ve just reinvented the Austrian theory of the business cycle! Too bad that’s what you were trying to refute in the first place.
Tim Starr – Renaissance Now! Think Universally, Act Selfishly
Assistant Editor: Freedom Network News, the newsletter of ISIL,
The International Society for Individual Liberty,
1800 Market St., San Francisco, CA 94102
(415) 864-0952; FAX: (415) 864-7506; firstname.lastname@example.org
Liberty is the Best Policy – email@example.com
From firstname.lastname@example.org Wed Jul 3 19:26:52 1996
Received: from mail6.netcom.com (email@example.com [22.214.171.124]) by ponyexpress.Princeton.EDU (8.6.12/8.6.12) with ESMTP id TAA18648 for ; Wed, 3 Jul 1996 19:26:50 -0400
Received: from ernie.ucop.edu (ernie.ucop.edu [126.96.36.199]) by mail6.netcom.com (8.6.13/Netcom)
id QAA28567; Wed, 3 Jul 1996 16:24:57 -0700
Received: from AHUPUAA (ahupuaa.ucop.edu [188.8.131.52]) by ernie.ucop.edu (8.6.11/8.6.11) with SMTP id QAA47174 for ; Wed, 3 Jul 1996 16:24:51 -0700
Date: Wed, 3 Jul 1996 16:24:51 -0700
Content-Type: text/plain; charset=”us-ascii”
Subject: Business Cycle Theory
Bryan Caplan writes:
>Reply to Vincent Cook:
>By “boom,” I mean a period of unusually high output, not “an increase in
>consumer demand relative to the demand for producer’s goods.” Perhaps I
>don’t understand your definition, but it would appear to be the case
>that consumer goods output and prices could fall 30%, producer goods
>output and prices could fall 50%, and under your definition there would
>be a “boom” in consumer goods.
It appears that an explanation of booms and busts are in order, so we can understand how Austrians define them. The key to the Austrian theory is the idea that artificial credit expansions cause a malinvestment of producer goods, which ultimately impairs the output of consumer goods and leads to a period of correction where producer goods are shifted back to their optimal configuration. The period of initial distortion is a boom, and the subsequent correction is a bust.
Note that the artificially-stimulated demand for producer goods in the boom phase does *not* mean that producer goods have somehow magically multiplied in quantity as an immediate consequence of the credit expansion. The critical immediate consequence of a credit expansion is that non-specific forms of labor, natural resources, and capital are shifted from stages of production that are less remote from the consumer to stages that are more remote, and that a substitution of technologies occurs where stages of production take longer but yield greater output. The artificial decline in interest rates fools entrepreneurs into thinking that longer-term projects are now more profitable, when in fact the long-term consumer time preferences will not support such time-intensive modes of production.
In the correction, or depression phase, producer goods are shifted back to shorter production processes. Shifting as such is not a costless process, either in the boom or the bust phase. However, from a psychological standpoint, the falsification of accounting and the increased demand for labor during the boom phase make most of us feel more prosperous even if in fact we are squandering our capital and generally wasting more of our productive efforts. The trauma of the bust, on the other hand, is compounded by the wave of bankruptcies as the errors are revealed, useless forms of capital are written off, and wage rigidity in hampered labor markets brings about massive unemployment.
>You are certainly wrong when you write that “No non-Austrian theory even
>begins to coherently explain the existence of inflationary depressions.”
>Here’s a few:
>a. Natural resource shocks, e.g. oil (reduces supply, raising price and
>b. Workers wake up from their real/nominal wage confusion and demand a
>raise to compensate for inflation (again, reduces supply, raising price
>and reducing output). Lucas won the last Nobel prize for his work on
>c. Technology shocks (again, reduces supply, raising price and reducing
>output). The theory which attributes business cycles to technology
>shocks, known as Real Business Cycle Theory (RBC), has been the hottest
>topic in macro theory for a decade.
Various supply-side shocks (you forgot to mention Smoot-Hawley, which also had the same effect) can account for price increases and declines in consumer goods output, but they do not explain the existence of inflation (which is classicly defined as an increase in money supply, not an increase in prices), nor the existence of a boom-phase preceding the bust, nor the concentration of boom/bust phenomena in capital-intensive sectors of the economy. Moreover, non-Austrian theories hold that demand-side stimulus ought to spur more production and thus prevent depressions and mass unemployment, albeit at the cost of price increases. The real puzzle of the 1970s to non-Austrians was that stimulatory policies, while predictably injurious to the value of the dollar, did *not* prevent a depression. In the 1973 to 1975 period in particular, there was a very nasty economic downturn and a raging credit expansion/inflation occuring simultaneously. The infamous stagflation did not fit the conventional paradigm, but it did the Austrian paradigm. Hayek, it should be noted, won the Nobel Prize in 1974 for his contributions to Austrian trade cycle theory.
The story of the early 1930s is complicated by the fact that supply shocks (especially Smoot-Hawley and the dust bowl) and severe Hooveresque interventions did occur in addition to the intitial credit collapse. But explaining 1929 is at the heart of our argument, and there is no obvious supply shock you can point to that accounts for the start of the depression.
>If you want other explanations for why capital goods industries are hurt
>_worse_ than consumer goods industries, there is also a simple
>alternative view which fits the facts better. One interesting business
>cycle fact is that DURABLE consumer goods suffer about as much as other
>capital goods industries. Very hard for the Austrian theory to explain.
Not at all. While Rothbard doesn’t focus much on durable consumer goods when talking about trade cycle theory, other Austrians have. I would recommend Mark Skousen’s _The Structure of Production_ for a more comprehensive treatment of the subject. In brief, durables are as much subject to time-based calculations as are producer goods remote from the consumer. In _Man, Economy, and State_, Rothbard does discuss how durables are capitalized over time according to the general rate of time preference. It is a relatively straightforward application of time preference theory to treat deferred consumption services provided by a consumer durable as equivalent to a final stage of production. The only complication is that the price of the durable reflects a bundle of multiple services distributed over time, not a service provided at a single time.
>One simple explanation is that _any_ durable good purchase, whether
>durable capital goods or durable consumer goods, is going to be much
>more sensitive to changes in income or profitability than non-durable
>purchases. In any period buyers of durable goods are both replenishing
>their stock to account for depreciation, PLUS adjusting their desired
>total stock depending upon new information about profitability (for
>firms) or permanent income (for individuals). The arrival of a
>depression causes both forecasts to be adjusted downwards; often this
>means that there is no point even making up for depreciation, since
>natural wear-and-tear simply moves you closer to your new, lower total
>stock. Thus, mainstream neoclassical economics has a perfectly clear
>and simple explanation for the relative price changes you discuss, which
>has the added virtue of explaining the decline in durable consumer goods
Whether purchasing a consumer durable or some form of fixed capital, one calculates the present value of a good by discounting all the future services provided by the good. If interest rates are being manipulated, the rate of discount changes over the course of the trade cycle. In the boom phase, low real interest rates imply a lower rate of discount on future services, while higher real interest rates during the bust mean a steeper rate of discount. Durables, relative to non-durables, undergo an increase in demand during the boom and a decrease in demand during the bust because of this interest rate sensitivity of their demand.
The deferred purchases phenomena is not unknown to Austrian theory, though there is no particular _a priori_ reason why consumer durables should be more affected than consumer non-durables. If one were afraid, say, of losing their job, one might want to reduce all kinds of consumption and not just defer durables purchases.
>Your attempt to describe the asymmetry of the sectoral shift from
>consumer goods to capital goods makes very little sense. Wage workers
>in capital goods industries are unhappy when old time preferences re-
>assert themselves. But wage workers in consumer goods industries should
>be overjoyed. The former suffer while they find a new job; the latter
>enjoy a period of high wages.
But workers in the lower order industries suddenly have a lot of new competitors for their jobs. To the extent that labor is non-specific to the lower and higher order industries, a uniform wage is going to prevail. The rise and fall in aggregate labor demand is due to the rise and fall in labor demand by the higher order industries. There is a minor mystery as to how the higher order industry was able to pay for its increased demand during the boom, but the answer is simple enough. The banking system created credit out of thin air and loaned it to the most time-intensive industries, such that investment expenditures (which were paid to owners of various factors of production, including labor) were able to increase without a concurrent restriction in consumer expenditures.
If unemployment increases because people
>are moving between sectors, then unemployment should also rise during
>the boom as well as the bust.
In a hampered labor market, wage rigidity is asymmetric. Wages are usually free to rise, but don’t fall easily when all sorts of interventionist devices are used on labor’s behalf. Nobody (except in Randian fiction) organizes the employers to go on strike, nor does the government compel workers to give employers special benefits or restrict the demand for labor.
>a. Arbitrage won’t work. Yes, it will; I’m talking about inter-
>temporal arbitrage, and I’ll tell you precisely how to make money.
>Suppose that cheap money has lowered interest rates this year to 1%. I
>know that this is not sustainable, and that next year interest rates
>will be back to 10%. But entrepreneurs, ignorant of the Austrian
>theory, stupidly expect the 1% interest rate to prevail forever, ad make
>business decisions based on a 1% interest rate. If the long rate is 1%
>too (reflecting the erroneous expectation of a permanent 1% rate), it’s
>easy to make money. Just borrow long. The first year put the money in
>a CD and each 1%. When the CD matures, interest rates have risen to
>10%, so you loan out the money. At the end of the year, you pay back
>the original 1%/year bond, and pocket the extra 9%. This strategy is
>sure to make you money unless the long rate is 11% over a two year
>period. (Thus, if you have to pay back $1000 in 2 years, then the issue
>value must be $900.09, and $909.09 a year later.)
If a single investor knows the truth and profits from it, their is little discernable macroeconomic effect. If, on the other hand, lots of investors begin to realize that a credit crunch is coming, something very interesting happens. Everybody tries to dump their 1% investments and become borrowers. Guess what- you get a crash in the financial markets! You wake up one morning and see the Dow dropping a hundred points every half hour and Alan Greenspan holding press conferences assuring the establishment that the Fed stands ready to bail them out. Entrepreneurs begin to recalculate how to make a profit at the new interest rate, and quickly conclude that they must disinvest from those businesses with the highest interest rate exposure. The high-cap industries, suddenly cut off from credit, begin to discharge workers and welch on their obligations. It sure sounds a lot like the start of a depression to me.
The point here is that a cluster of entrepreneurial errors, such as those caused by the distortion of the structure of production brought about by bank credit expansions, are corrected by new entrepreneurial information. But arbitrage is only as good as the information that entrepreneurs possess. If markets are initially ignorant of the distortion in the credit markets and the production structure, the trade cycle will inevitably be set in motion. Ignorance will always prevail to some extent, so the credit manipulations will always have some impact. Perhaps after a few iterations of the boom/bust cycle, some investors will learn to watch for advance signs of credit expansions and time their investments accordingly. The relevant information needed for such a strategy to work is agonizingly difficult to come by (and may not exist at all), but when it is available it can be extremely profitable. One wonders if the SEC has ever investigated employees of the Fed for trading on inside information . . .
>The whole point of the Austrian theory is supposed to be that low
>interest rates mislead entrepreneurs; they discount future income
>streams at the current, low rate of interest, and wind up holding the
>unprofitable bag when interest rates rise. What I am saying is that due
>to intertemporal arbitrage, which is going on all the time all over the
>world, this will never happen. Even a moron knows that you don’t use
>the current short-term interest rate to discount all future income
>flows. The existence of inter-temporal arbitrage on longer-term debt
>provides a simple and accurate alternative way to discount future income
>streams rationally so the “malinvestment” problem need not arise.
There seems to be a serious confusion here between long/short loan arbitrage and arbitrage between current and future rates of return on production projects. The spread between input costs (like wages, royalties, and rents) and output income that I get by engaging in a production process over time is independent of how I financed the purchase of the inputs for the process. Thus, my demand for credit is going to depend on my calculation of what those spreads are. Whether I sell short-term instruments, long-term instruments, or shares of ownership of the business, the overall rate of return on the initial expenditures for inputs is strictly a function of the income generated by the production process. The issuance of long or short debt merely gives creditors a fixed portion of this income.
Since business debtors typically are not trying to be charitable to creditors, they will not bid more for either a long or short loan than is justified by the natural rate of return. This means that in an unhampered market, the yield curve on loan instruments should be flat except for differences in risks on the various instruments. The yield on stocks or other evidences of ownership of productive assets likewise should conform to the natural rate of interest. The long/short type of arbitrage merely serves to flatten out the yield on all types of investment securities. Some structural unflattening of the yield curve is possible if the central bank prefers dealing in short-term instruments, but there is no inherent link between the yield curve and credit expansion. Because of long/short arbitrage, a credit expansion injected into the short-term market will depress long-term interest rates as well as short-term interest rates. A passive follower of long-rates such as was described by Mr. Caplan in his previous post is not immune to the falsification of capital accounting caused by the credit expansion.
Anticipated changes in the natural rate of return, however, should affect all yields. Expected changes in the spread between input expenditures and output income affects all securities markets, including rates of return on all forms of debt and equity. It is silly to accuse Austrians of holding a qualitative credit doctrine (which Rothbard in particular goes into great detail to refute), because forecasts of the natural rate of return have nothing to do with the differences between short-term and long-term loan instruments in the Austrian trade cycle theory. Rothbard and other Austrians argue that a producer could just as easily track the long-term rate instead of the short-term rate and still get burned by a credit expansion/credit crunch sequence.
>b. People (or at least politicians) aren’t predictable. Sure they are.
>Not perfectly. But quite well. This can be shown by looking at stock
>market movements immediately upon the announcement of new policies.
>There are often sharp changes – representing the market’s correction of
>its forecast error – but they are incredibly small as a percent of the
>value of the shares traded. It is a plus or minus 1% change, not a plus
>or minus 50% change. Thus, in most cases the market’s forecasts of
>government behavior are quite accurate. At least as accurate as its
>forecasts of world oil supply, gold discoveries, and a million other
The phrase “This can be shown” in the above context is highly doubtful. Have the positivist economists finally found some way to conduct a controlled experiment on human beings? Moreover, I can distinctly remember market moves that were substantially greater than 1%. Less than 10 years ago, there was a stock market crash every bit as dramatic as what happened in 1929 (losing 25% in one trading session as I recall), and the economy did indeed slump into a downturn, though the subsequent response by the government was mercifully less interventionist than in the Hoover/Roosevelt scenario.
Am I really supposed to take seriously the notion that markets are always nearly perfectly informed? Even the most casual glance at stock market charts discloses a great uncertainty in the value of most companies, with highly-leveraged cyclical companies in particular showing great flucuations in share price. A perfectly prescient market would not show such variability in share prices, given that the sum of the discounted future earnings of these companies historically have been much less variable.
>c. How could a passive price-taker know when interest rates were too
>high or too low? Easy. If the long rate is above the short rate, a
>robot could determine that rates are temporarily low. The bigger the
>disparity, the bigger the divergence.
See my comments above about the difference between external rates of return on long vs. short financial isntruments and the internal or natural rate of return on production processes.
>d. Investors and economists are ignorant of economic theory. Well,
>journalists sure are. But big mutual funds have very smart people
>working for them who are not ignorant. I see no evidence that Austrian-
>influenced funds do particularly better than just buying an index fund.
>Gold bugs must have been kicking themselves all through 1995 as gold
>stocks made almost no money while the rest of the stock market gained
Austrians would be the first to tell you that economists make lousy forecasters, and even Skousen is not a gold bug all the time. What matters in outperforming the market is having access to better information and cashing in on it in a timely manner. Having a sound grasp of economic theory, or alternatively, having enough experience in the markets to have good intuitions about cyclical phenomena and their effects on prices, is a necessary but not sufficient condition for being a successful speculator.
>Your remarks on problems with government statistics are all probably
>right, but that hardly prevents forecasting from working. It is just
>less accurate than it otherwise would be. Similarly, statistics were
>less ample and quickly available in the 20’s, but there was still plenty
If you cannot perform controlled experiments (where you systematically vary one factor at a time), you can’t really know what the future holds. The only reason why economic statistics have any forecasting value at all is that humans tend to have the same circumstances and values tomorrow that they had yesterday, and that there is a body of a priori theory that helps us understand which variables are meaningful and what the direction of causality is among these variables. But we know humans, both individually and en masse, can undergo substantial and even radical changes in circumstances and values in an unpredictable fashion. How many people do you know who predicted the fall of the Soviet Union and the establishment of a somewhat-capitalist Russia? Ten years ago, how many people were investing big bucks in the Internet? How many of the statistical wizards in charge of the mutual funds, etc. do you know who profited from the aforementioned stock market crash?
>Finally, you wisely wonder what caused the bank failures and defaults in
>the first place. There were several factors at work. First, an
>increase in money demand, which tends to create defaults throughout the
>economy. Second, wage rigidity encouraged by Hoover created more
>defaults. And finally, deflation and bank failure can and did form a
>vicious cycle: deflation makes it impossible for people to repay their
>loans, causing banks to fail, which further reduces the broader money
>supply and worsens deflation.
(1) So entrepreneurs are smart enough to figure out all the sudden twists and turns of Fed policy, but they can’t react to gradual changes in monetary demand? Since the exchange demand for money (the goods and services offered in exchange for money) was relatively stable during the late 1920s, this hypothesis supposes that there was a cluster of entrepreneurial error caused by a sudden jump in the reservation demand for money (the desire by cash holders to retain their balances instead of spending them). This would imply that a fall in consumer expenditures and incomes preceded the decline in money supply and the implosion of the credit structure. But where is the evidence for this? Incomes did not start falling catastrophically until 1931. Also, it wasn’t until 1931/32 that a real run on the banking system took place. But the challenge here is to explain what happened in 1929. Why, at the peak of aggregate incomes and consumer expenditures during the whole period, did it suddenly become so difficult for certain industries to make a profit? Why did the collapse in prices for capital goods, as represented by stocks, begin their dive in the midst of such prosperity?
(2) The downward rigidity of wages, while it explains why workers would default in an environment where wages were falling, doesn’t explain why wages should fall in the first place. The equilibrium wage is at the point where the discounted marginal physical product of the marginal unit of labor times the price of the output equals the wage per unit of labor. Since output prices and physical productivity were relatively stable, what factor is left to account for the sudden decline in the value of labor in late 1929? The rate of discount, of course. It was the upward pressure on interest rates that was putting downward pressure on wage rates, particularly in the high-cap industries where the labor inputs were most remote from the consumer (magnifying the effect of the discount). It is true that Hoover didn’t allow wage rates to fall, thus creating mass unemployment and compounding the defaults. But Hoover didn’t create the artificially cheap credit that triggered the collapse in the first place. One has to read Rothbard’s AGD to discover who those villians were.
(3) But since most of the defaulted loans were secured by various collateral assets, there is the additional question of why the banks couldn’t recover their losses by selling the foreclosed collateral. Obviously, the collateral assets had to have declined in value first before loan defaults would cause any significant losses to the bank (significant enough to render loan loss reserves inadequate). Just as changes in the rate of discount had caused downward pressure on wage rates, they also caused a strong downward movement in the price of capitalized goods used as loan collateral. At a later stage (1932), rural banks faced the additional problem that armed force was used to prevent them from recovering and selling their property at market value (the infamous “penny sales”).
Reply to Jason Sorens:
You are quite correct that I was only describing the Fed’s effect on financial markets. Other markets, such as labor markets, do not appear to be nearly as quick to respond to policy changes. Probably because an individual arbitrager can make millions by noting that inflation will be 1% higher than expected, whereas for most workers it isn’t even worth thinking about in low-inflation regimes.
I don’t know what you mean when you say that business cycles are “continuous, institutionalized, and regular.” Can you predict when the next downturn will be, how severe it will be, and how long it will last? Clearly they aren’t regular in that sense. What sense do you mean?
I think you need to review the exact powers of the Fed more closely. The Fed does not control interest rates by imposing price controls, as you seem to imply. It controls interest rates by adjusting the rate of money supply growth. Thus, when the Fed “raises interest rates to forestall inflation,” there is nothing foolish in its behavior; this is just another way of saying that the Fed reduces money supply growth to forestall inflation, which is the only workable way to do it.
Reply to Vincent Cook:
By “boom,” I mean a period of unusually high output, not “an increase in consumer demand relative to the demand for producer’s goods.” Perhaps I don’t understand your definition, but it would appear to be the case that consumer goods output and prices could fall 30%, producer goods output and prices could fall 50%, and under your definition there would be a “boom” in consumer goods.
You are certainly wrong when you write that “No non-Austrian theory even begins to coherently explain the existence of inflationary depressions.” Here’s a few:
a. Natural resource shocks, e.g. oil (reduces supply, raising price and reducing output).
b. Workers wake up from their real/nominal wage confusion and demand a raise to compensate for inflation (again, reduces supply, raising price and reducing output). Lucas won the last Nobel prize for his work on this idea.
c. Technology shocks (again, reduces supply, raising price and reducing output). The theory which attributes business cycles to technology shocks, known as Real Business Cycle Theory (RBC), has been the hottest topic in macro theory for a decade.
If you want other explanations for why capital goods industries are hurt _worse_ than consumer goods industries, there is also a simple alternative view which fits the facts better. One interesting business cycle fact is that DURABLE consumer goods suffer about as much as other capital goods industries. Very hard for the Austrian theory to explain.
One simple explanation is that _any_ durable good purchase, whether durable capital goods or durable consumer goods, is going to be much more sensitive to changes in income or profitability than non-durable purchases. In any period buyers of durable goods are both replenishing their stock to account for depreciation, PLUS adjusting their desired total stock depending upon new information about profitability (for firms) or permanent income (for individuals). The arrival of a depression causes both forecasts to be adjusted downwards; often this means that there is no point even making up for depreciation, since natural wear-and-tear simply moves you closer to your new, lower total stock. Thus, mainstream neoclassical economics has a perfectly clear and simple explanation for the relative price changes you discuss, which has the added virtue of explaining the decline in durable consumer goods purchases.
Your attempt to describe the asymmetry of the sectoral shift from consumer goods to capital goods makes very little sense. Wage workers in capital goods industries are unhappy when old time preferences re-assert themselves. But wage workers in consumer goods industries should be overjoyed. The former suffer while they find a new job; the latter enjoy a period of high wages. If unemployment increases because people are moving between sectors, then unemployment should also rise during the boom as well as the bust. A mistaken sectoral shift does indeed have a real cost; I agree to that without complaint. But the Austrian explanation _by itself_ does nothing to explain why unemployment is high during the “bust” and low during the “boom.” It can’t even explain why output declines. Bohm-Bawerk’s capital theory, on which Rothbard wisely built his work, would suggest that actually the SHORT-run effect of switching to consumer goods production would be a period of _greater_ production, followed by a period in which production is less than it would otherwise have been if longer period products were tried instead.
Moving along. You give four reasons why it is harder to forecast and compensate for government policy.
a. Arbitrage won’t work. Yes, it will; I’m talking about inter-temporal arbitrage, and I’ll tell you precisely how to make money. Suppose that cheap money has lowered interest rates this year to 1%. I know that this is not sustainable, and that next year interest rates will be back to 10%. But entrepreneurs, ignorant of the Austrian theory, stupidly expect the 1% interest rate to prevail forever, ad make business decisions based on a 1% interest rate. If the long rate is 1% too (reflecting the erroneous expectation of a permanent 1% rate), it’s easy to make money. Just borrow long. The first year put the money in a CD and each 1%. When the CD matures, interest rates have risen to 10%, so you loan out the money. At the end of the year, you pay back the original 1%/year bond, and pocket the extra 9%. This strategy is sure to make you money unless the long rate is 11% over a two year period. (Thus, if you have to pay back $1000 in 2 years, then the issue value must be $900.09, and $909.09 a year later.)
The whole point of the Austrian theory is supposed to be that low interest rates mislead entrepreneurs; they discount future income streams at the current, low rate of interest, and wind up holding the unprofitable bag when interest rates rise. What I am saying is that due to intertemporal arbitrage, which is going on all the time all over the world, this will never happen. Even a moron knows that you don’t use the current short-term interest rate to discount all future income flows. The existence of inter-temporal arbitrage on longer-term debt provides a simple and accurate alternative way to discount future income streams rationally so the “malinvestment” problem need not arise.
b. People (or at least politicians) aren’t predictable. Sure they are. Not perfectly. But quite well. This can be shown by looking at stock market movements immediately upon the announcement of new policies. There are often sharp changes – representing the market’s correction of its forecast error – but they are incredibly small as a percent of the value of the shares traded. It is a plus or minus 1% change, not a plus or minus 50% change. Thus, in most cases the market’s forecasts of government behavior are quite accurate. At least as accurate as its forecasts of world oil supply, gold discoveries, and a million other things.
c. How could a passive price-taker know when interest rates were too high or too low? Easy. If the long rate is above the short rate, a robot could determine that rates are temporarily low. The bigger the disparity, the bigger the divergence.
d. Investors and economists are ignorant of economic theory. Well, journalists sure are. But big mutual funds have very smart people working for them who are not ignorant. I see no evidence that Austrian-influenced funds do particularly better than just buying an index fund. Gold bugs must have been kicking themselves all through 1995 as gold stocks made almost no money while the rest of the stock market gained 35%.
Your remarks on problems with government statistics are all probably right, but that hardly prevents forecasting from working. It is just less accurate than it otherwise would be. Similarly, statistics were less ample and quickly available in the 20’s, but there was still plenty of information.
Finally, you wisely wonder what caused the bank failures and defaults in the first place. There were several factors at work. First, an increase in money demand, which tends to create defaults throughout the economy. Second, wage rigidity encouraged by Hoover created more defaults. And finally, deflation and bank failure can and did form a vicious cycle: deflation makes it impossible for people to repay their loans, causing banks to fail, which further reduces the broader money supply and worsens deflation.
Reply to Tim Starr:
First of all, I think we should at least keep this polite.
As for your substantive points:
1. It is true that small businesses cannot borrow and lend with equal ease at the quoted market interest rates. However, small businesses are a very small part of the U.S. economy; larger firms are in a very different situation. They have access to corporate bond and commercial paper markets, which allow precisely what I said — borrowing or lending at market rates with ease. Moreover, even small businesses can usually easily implicitly lend by repaying some of their debt, so they aren’t nearly as constrained as one might think.
2. I think you are confusing market-clearing interest rates and natural interest rates. Central bank intervention _lowers_ the market-clearing rate below the natural rate; it doesn’t reduce rates below the market-clearing rate. Of course, lenders have no choice but to lend at the lower rate if the central bank cuts its rates. But at my remarks to Vince indicate, there is no reason for long-term interest rates to be deceptive or cause malinvestment. If due to central bank intervention, interest rates are temporarily 1%, and (known only to Austrian economists) will have to rise back to 10% next year, then this year, lenders must take 1%. But if they lend out for 2 years, they won’t stupidly lend out at 1%/year for two years. Arbitrage prevents it. Otherwise, lenders could only buy short-term debt during the low-rate period, and then once rates rise they can be sure to earn the higher rates.
3. Thus, my view of business operations bears no relation to the story you tell. If rates are low now, but will be higher later, I wouldn’t tell my boss not to lend. I would suggest lending short, so we could take advantage of higher rates once they hit. I would however get fired if I presented an analysis of a project’s profitability, and in my calculations I stupidly used currently low short-term rates to discount all future cash flows. There is an easy way to avoid this error – look at long-term rates. And if I do so, there is no reason for any malinvestments to occur — no more than usual error would create, anyway.
4. One way that real wages can be rigid is if it takes time for them to change. Perhaps the phrase “real wage _inflexibility_” would make the meaning clearer.
5. I do believe in malinvestments. I just don’t think that the reason for clusters of malinvestments is central bank policy. Anytime someone calculates that a project will be profitable, and it isn’t, there is malinvestment. But intervention in short-term rates has no systematic effect on profitability calculations.
6. As for why still-employed workers would get a compensating increase. Simple. The whole Austrian theory is based on the idea that during the boom, there is a sectoral shift from consumer goods industries to capital goods industries. Hence, during a bust, the reverse shift occurs. It stands to reason that when any shift occurs, people already in the sector for which demand is increasing would benefit, just as people in the sector for which demand is decreasing would suffer.
7. You are quite correct to mention that a decline in money supply has the same effect as an increase in money demand. However, as Rothbard clearly points out, the bust occurs when the money supply increase _ceases_. On the Austrian view, the bust could occur even if the money supply remained constant at its higher level; monetary contraction is not an integral part of the theory (although as Rothbard points out in the pre-war period monetary contraction typically did occur). This is fortunate for the theory since there has not been an absolute decline in the money supply since WWII (with one debatable month in the late 40’s if I recall).
8. Hazlitt once remarked that Keynes said nothing both original and true. The same goes for the Austrian business cycle theory; insofar as it is true, it is not original, and insofar as it is original, it is not true. You amazingly claim that I have “reinvented the Austrian theory of the business cycle” by attributing unemployment to a combination of monetary contraction (or money demand increase) and rigid nominal wages. To the contrary, this explanation for unemployment has enjoyed the universal agreement of all academic macrotheorists of all schools since the 1920’s. It is not an Austrian view. Keynes repeatedly affirmed it in his _Treatise on Money_ and _General Theory_. The British economist Pigou used it in his _Theory of Unemployment_ [that title could be slightly off] published in the 20’s. The only difference is that Keynes accepted nominal wage rigidity as politically given, and tried to figure out ways around it. Stupid followers of Keynes failed to understand the microtheoretical underpining of the Keynesian model, but Keynes and his smarter followers such as Modigliani always knew that they were implicitly assuming nominal wage rigidity.
Since the 1970’s, even the stupidest Keynesian has been made aware of this fact; and essentially all academic macrotheory has built on this idea, with the exception of Real Business Cycle theory. Real Business Cycle also accepts the view that nominal wage rigidity _would_ cause unemployment, but they deny that there is any significant nominal wage rigidity.
Economic journalists may still be oblivious to this fact, but no academic macrotheorist is. They may however think that reducing nominal wage rigidity is so politically difficult that they don’t think the topic is very interesting.