April 2014

Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30      
Blog powered by Typepad

« A Milton Friedman on the Horizon? | Main | The People’s Republic of Berkeley »

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d83451eb0069e20120a58e693f970c

Listed below are links to weblogs that reference Hoover and the Great Depression, redux:

Comments

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

Hoover himself kept insisting that he was no laissez-fairist. In his acceptance speech for the GOP nomination in 1932 he laid out the many things he did to combat the depression. People should read it.

Anyway, the difficulty for interventionists is that if they admit Hoover did many of the things Roosevelt did or wanted to do, then they will have to distinguish forms of intervention. Thus they might have to know something instead of simply *feeling* something.

They can't admit that Hoover was an interventionist because if they did, they can't claim that FDR saved America. If they admit that Hoover did intervene, then it would prove their point wrong. Intervention is the statist weapon and if Hoover did intervene then they have just lost their argument for their only weapon.

It's the same in the current environment. Remember, it was the deregulation of and lack of intervention in the free market in the G. W. Bush's term that caused the economic crisis. You can point to all the examples of State intervention you want, but they will shift the argument to needing more intervention. I am sure with "better" technocrats.

Roosevelt is a "hero" because he expanded the _federal_ welfare state. Federal government policies aimed at taking care of the unemployed, widows and orphans, and the elderly.

Hoover had the Federal government provide some aid to state and local government for relief, but that sort of temporary piecemeal effort is little more than charity, rather than a recognition of a Federal responsibility to enforce individual welfare rights.

Also, Hoover claimed that price fixing was in the interests of the business class. The market process, in his view, amounted to cutting their own throats. (As in, cut throat competition.) Doesn't populist rhetoric about protecting the little guy against exploitation by big business sound so much more "heroic?" Lambast the rich and have government provide social services to everyone. Isn't that all that is worth considering?

What is a little cartelization aimed at protecting the status quo? Unless the policies are aimed at transfering wealth away from the rich through a federal welfare state, it is really nothing different from laissez-faire. Who understands all that business about cartellizations and competition?

You do realize that in the 1929-1933 period when Hoover was supposedly keeping wages high that manufacturing wages fell 21.4% as compared to a 12.7% decline in the 1922 recession when the governments non-intervention policies supposedly allowed the markets to adjust.

Or, are you completely ignorant about these "so called" facts.

Are you referring to nominal wages or real wages Spencer? Are you accounting for changes in productivity as well? The point is that even if real wages fell by 21.4%, given that the price level was experiencing close to a 30% decline and that productivity was declining as well, wages should have fallen even more than they did over the period.

The best evidence that real wages were still too high, thanks to Hoover's policies (including Davis-Bacon and Smoot-Hawley), is that unemployment continued to climb to levels never before seen. Unemployment rates in the mid 20% range are, shall we say, pretty suggestive of real wages being too high.

Or are you completely ignorant about the level of unemployment and its relationship to the real wage?

Oops, I should have said "even if nominal wages fell by 21.4%...."

I'm well aware of the points you are making.

I'm just simply making the point that wages actually fell sharply when this paper is claiming they rose.

You will also note that as soon as the economy bottomed in 1933 wages started rising.

The evidence is overwhelming that your simplistic introductory economics model of wages has no support in the data.

Show me a single real life example of falling wages leading to greater employment.

When demand falls as it does in a recession the marginal product of excess labor falls to zero and with a marginal product of zero no amount of wage cutting will lead to business hiring employees they do not need.

Steve,

Your answer to spencer assumes what you and Ohanian are trying to prove, that the rising unemployment was due to the insufficiently rapidly falling wages.

There are lots of problems with this paper by Ohanian, including some I think any decent Austrian economist would be all too aware of, starting with its assumptions of general equilibrium and rational expectations throughout. Was not this period one of massive disequilibrium all over the place with expectations totally out of whack? Nobody has had a worse record in analyzing what is going on now in the US economy than this pathetic DSGE gang.

Some other points.

Real wages fell during the 1870s depression, but that did not keep unemployment from sharply rising. There was no pro-union president then, and unions were practically non-existent, and very weak to the extent they existed. Roughly similar conditions held in the 1890s as well, although labor activism was greater. On this matter see Clarence Dickinson Long's _Wages and Earnings in the United States 1860-1890_, 1975 (at least for the 1870s period).

Despite the hilarious effort by various folks to turn Hoover into an FDR, there is no question that the power of unions was increased far more under FDR than under Hoover (Wagner Act, 1935), along with lots of other interventions, far more than Hoover's (some of his probably damaging the recovery, I would agree). If this sort of thing explained what was going on, then why did we not see a worsening of conditions after FDR came in?

Somehow Ohanian fails to notice that nominal debts were not reduced, which meant that real indebtedness rose with the deflation. Wages declining even more rapidly would worsen even further bankruptcies and pressure on wage earners from rising real indebtedness. This is a point made by many, starting with Irving Fisher, going through Milton Friedman, and including many others. Would we expect employment stabilization now if all salaries were cut 25% while mortgage payments and other debt payments remained constant?

Finally, there is the little matter that if one looks at the international data on real wage changes and changes in output during the GD, there is no relation whatsoever. Not there. None. Brad DeLong provides a nice figure on this from a classic paper on the subject at
http://delong.typepad.com/sdj/2009/08/herbert-hoover-a-working-class-hero-is-something.html.

Bottom line: anybody who pushes this Ohanian story in a too-uncritical way is going to end up looking kind of silly, especially someone who is not a knee-jerk believer in the now-thoroughly discredited DSGE approach.

First interesting thing here is that recovery started 6 months before Roosevelt took office, and probably about a year before any of Roosevelt's policy measures took effect. In what regard we can credit Roosevelt for that (weak) recovery? Hoover “liquidationist”, if anyone, should be credited.

Second, De Long's argument is quite silly. According to him, everything would be even worse if wages were less rigid, i.e. allowed to fell to market clearing level, which is to say even more than they actually fell. But, by the same token we could say Hoover should urge employers to increase wages three times, and depression would be even milder than it was! If De Long is right, everything that we need to avoid depression is thus - strong unions and rigid wages!

And third, but not least important, we have forgotten depression of 1920/21. Prices fell more steeply than during Great Depression, government actually slashed spending instead to increase it, and recovery was almost immediate. Government did not abandon gold standard, as Friedmanite inflationists require, nor it increased spending, as Keyensian inflationists require. Why then we need those same “arguments” (abandoning gold and increased spending) as "explanations" of Great Depression? Occam's Razor!

Nikolaj,

All the data I have seen shows the bottom of GDP and the peak of unemployment at about the time of FDR coming into office. The stock market hit bottom in June, 1932, but not much else that I am aware of did prior to March, 1933.

As I noted on another thread here, the 1920-21 recession was a postwar readjustment that had none of the financial market distortions and collapses that we see with the really serious recessions and depressions, such as the one in the 1870s, when policy pretty much resembled those in the early 1920s, and even had falling real wages, as well as the Great Depression and our current unpleasant situation.

Barkley,

I actually haven't read the full paper yet, so I'll reserve comment. My point in the original post was not to defend Ohanian per se, but to respond to the response to his paper that it just *can't* be true that Hoover was in favor of activist government. (Yes, less active than FDR, no argument there.)

In re-reading my original post, I make no effort to defend O's paper, rather my focus is only on the question of Hoover's beliefs. I wisely reserved judgment on his paper itself until I read the whole thing.

You might want to reserve judgment on my take on Hoover, Roosevelt, and the GD until the EJW piece is out (in which, I will note, I attack a DSGE model).

Steve,

OK. Point well taken.

Another point about Herbie Hoover is that, in the 1920s, he paved the way for New Deal intervention in the housing market. See Janet Hutchison, "Shaping Housing and Enhancing Consumption: Hoover's Interwar Housing Policy," in John F. Bauman, Roger Biles and Kristin M. Szylvian, _From Tenements to the Taylor Homes: In Search of an Urban Housing Policy in Twentieth-Century America_.
A real class act, Hoover. Worse than Dick Milhous, contra Krugman, even if Dick did do Section 8 vouchers.

Barkley,

It is not clear what means "postwar readjustment". What kind of specific theory of recession is that? Whatever reasons for recessions in general, 1920&221 recession according to conventional wisdom, should be much worse than Great Depression, because government let prices to decline sharply and slashed spending during the bust. Conventional wisdom, supported both by Friedmanites and Keynesians, is that specific depth and severity of GD should be ascribed to government allowing prices to decline so sharply. Simple question - why 1920/21 recession was so short and less severe, when price deflation was even worse than during GD? And government decreased its spending instead to increase it.

But, it is even more amusing. You rightly point out that recessions during 19th century were milder than GD, and in all of them general price level fell and government policy resembled to that of early 1920s. But, whence then severity and longevity of Great Depression? Something is obviously wrong on logical level here, if we insists on one circumstance that is common to all depressions (falling price level), to explain specific features of one of them. Isn't it?

Nikolaj,
I do not know exactly what caused the 1920-21 recession, but it did not involve the sorts of finanical market dislocations and crashing bubbles as the worst ones.
Calling the 19th century ones "mild" is a joke. The 1870s one was second to the GD in severity and the 1890s was third, by the reckoning of most economic historians. The one in the 1870s went on for most of the decade.
I just did some checking and found a source for what happened with wages in 1920-21, curiously a source pushing the Ohanian story, one Gerald Jackson at http://www.brookesnews.com/092601obama1930s.html. So what numbers does he tout? During the same time that prices fell 45%, nominal wages only fell 11%, in short, real wages rose substantially. So much for that wage flexibility story being the key to what happened.
There is also a paper by Bryan Caplan on all this, in which he deals with the Bernanke-Cary model. Aggregate data shows the same degree of wage flexibilitiy in the early 1920s recession as in the Great Depression. He then uses a disaggraged by industry data set that finds greater wage flexibility in the early 1920s, which I find completely unsurprising, except that he does not do the same thing for the 1930s to make a proper comparison. Instead he argues how of course there was greater wage flexibility in the early 1920s because there was the more rapid rebound, thus of course assuming the outcome that needs be proven in order to prove the outcome.

This is all pretty weak and pathetic.

Ohanian:

"This analysis also provides a theory for why low nominal spending – what some economists refer to as deficient aggregate demand - generated such a large depression in the 1930s, but not in the early 1920s, which was a period of comparable deflation and monetary contraction, but when firms cut nominal wages considerably".

Exactly. Now, Ohanian might be right or wrong in ascribing severity of GD to (only) high wages during Hoover's term. But one thing is obvious - Friedman-Schwartz conventional "falling price level" theory is logically clearly without the merit.

Barkley Rosser: During the same time that prices fell 45%, nominal wages only fell 11%, in short, real wages rose substantially. So much for that wage flexibility story being the key to what happened.

Well, real wages rose substantially and guess what? So did the unemployment rate. In 1919 the unemployment rate was 1.4 percent and by 1921 the unemployment rate was 11.7 percent. It sounds like wage flexibility or the lack there of is key to what happened to the unemployment rate. But unlike during the GD, where the government was trying to pursue a policy to keep wages from falling, the unemployment rate fell back down to 2.4 percent by 1923.

Barkley - that Brookesnews link is broken.

Nikolaj,

No, the falling price theory is not disproven. What happened in the GD was due to the fact that prices fell over an extended period of time, thus allowing them to get embedded into expectations. The price decline in 1921 was over so quickly it barely had any time to register on anybody.

Tom,

The question is the impact of policy about wages on the economy. Your argument that because real wages rose that means that "wage flexibility or the lack thereof is key to what happened." How so? We have all these people arguing that wages were flexible in 1921 but not in 1929, thus explaining the difference, but in fact they do not seem to differ much in that flexibility, with, if anything, the increase in real wages being greater in 1921 than in the later period. I also note we have had much less downward wage flexibility since WW II, but, gosh, nothing like either of those episodes until now.

Sinclair.

Hmmm. You are right. I just found it again by googling "US wage rates 19210 1921 1922," and it was the 18th item. Showed up fine with the url I gave, but trying the url itself does not seem to work. Sorry about that.

As Mises noted,

Since they are all complex, "Every historical experience is open to various interpretations and is in fact interpreted in different ways. History can neither prove nor disprove any general statement."

But economics can. It proves that there cannot be chronic, massive unemployment in a free market, that it can only arise from interference with the market.

Barkley Rosser said: "No, the falling price theory is not disproven. What happened in the GD was due to the fact that prices fell over an extended period of time, thus allowing them to get embedded into expectations. The price decline in 1921 was over so quickly it barely had any time to register on anybody."

This is not explanation, but circular reasoning. Fact that prolonged fall in prices accompanied GD is not an explanation for it, but the most important fact that has to be explained. Why prices fell over prolonged period of time only during GD while ending quickly in 1921 and in many other recessions? They should not, according to monetarist and Keynesian theories. Government 1920-21 SLASHED spending and ALLOWED quantity of money to decrease as much as during GD. Simple question is - how was it possible that recession 1921 ends so quickly without abandoning gold standard and without any Keynesian fiscal expansion? You do not offer any explanation for that. To credit Roosevelt's abandoning gold standard for recovery is additionally illogical, because you agreed with me that recovery started well before Roosevelt took office.

So, falling price theory is pretty much disproven as explanation for GD, that is unless it provides some specific reason why deflation produced so disastrous results over so long period of time only 1929-32 and not in any other similar situation. Morevoer, in some periods, deflation was accompanied by decades of strong economic growth (last third of 19th century in America for example). To merely repeat fact of long deflation as a "cause" of economic decline is not an explanation, but petitio principii.

"In most situations writers are not much interested in carefully doing justice to the character of rival alternatives. Their purposes are instead pedagogic or argumentative. Their aim is to get the reader to the point where he or she has taken on the writer's own particular understanding of a thing. Or the aim is to propel an audience to accept the writer's conclusion about some contention. These pedagogic and argumentative aims transform the narratives we get of the explanatory alternatives which are available within a discipline, and very often they obliterate the character of the alternatives that existed in the past. Philosopher and science historian Thomas Kuhn has provided us with a number of instructive examples showing how classic scientific texts and standard textbook presentations tend to obscure the past and mislead us about the role played by such things as measurement and conceptual change in the process of the development of physical science.

The philosopher of biology David Hull goes further, showing how outright myths have served the pedagogic and argumentative purposes of science studies writers. Among other examples, Hull recounts the venerable myth which has it that the elder John D. Rockefeller, Sr. sought to vindicate the virtue of 19th century American capitalism by appeal to `Darwin's' theory of the survival of the fittest, which was taken to imply that those who were financially successful in the American economy outcompeted others due to their superior fitness as human beings. The usefulness of this myth in advancing the pedagogic or argumentative purposes of many writers is hard to miss. It is also easy to understand why the myth of QWERTY as an inferior technology for typewriter keyboards seems not to die in the path dependence literature, despite the thorough debunking this myth has been given by Liebowitz and Margolis. In order to advance a number of particularly treasured pedagogic or argumentative goals some myths are just too good to give up, no matter how far off the mark these myths may be from the truth."

From my unpublished paper "The Significance of Myth and Misunderstanding in Social Science Narrative: Opening Access to Hayek's Copernican Revolution in Economics"

Ohanian is a good paper because its focus is why employment fell rapidly in manufacturing when real wages were stable or rising and agricultural employment was stable with large falls in real wages.

His and Rothbard’s explanation is cartelisation with Hoover as the ringmaster. Rothbard treated Hoover as the ringmaster of employer cartels and rather quickly dismissed the threat of unions. Ohanian included protection from unions as Hoover’s bargaining chip over hold-outs.

Most responses to Ohanian say how dare he! The correct response is can he make his case?

Cartel explanations of the early years of the great depression face the problem of collusive instability due to cheating and new entry.

The Rothbard and Ohanian cartels were informal, and compliance and new entry were not policed by law. Cheating is more likely in these cases because the chances of detection and punishment are less. Non-price competition also builds with time within all collusive arrangements as the history of all regulated industries show.

A special problem for Austrian analysis is the protection and dispensations offered in both the Hoover and Roosevelt New Deals from antitrust laws. If antitrust laws are a burden on competition, removing them would increase market efficiency in the 1930s.

Antitrust laws have complex special interest dynamics. Large corporations in the 1930s would want to weaken antitrust laws to allow them to undertake cost cutting mergers and compete with greater vigour with smaller rivals. As Coase wrote somewhere in early the 1970s, antitrust laws presume price cutting to be predatory, prices rises to be monopolistic, and stable prices are collusive.

Ohanian has made a good case for Hoover as a cartel ringmaster. There is a need to explain the selective collapse of manufacturing but not agricultural employment, but there still unanswered questions. I think it was John McGee who wrote that the history of cartels is the history of double-crossing.

Having now read the Ohanian paper, I can say a few more things. Jim Rose has made the key point though. :)

The first 27 pages of the paper are simply GOOD economic and intellectual history. Not an equation in sight. Just some historical data on wages and some very good work with Hoover's memoirs and some secondary sources documenting Hoover's long-standing desire to maintain wages and the deal he struck with manufacturing firms to call off organized labor if they didn't cut nominal wages. I think it's an excellent piece that adds to the understanding of this period we have from other historians and economists, including Rothbard.

The second half of the paper is his DSGE model that tries to estimate just how strong the effect was. I should note: to suggest as Barkley did earlier that it begs the question to claim that maintaining nominal wages in the face of falling prices and productivity is an important explanation of the high unemployment of the Hoover years seems to me to suggest that demand curves don't slope down. I plead guilty to thinking that they do and I also point to Vedder and Gallaway for an attempt to provide the empirics Barkley seems to want.

Barkley quite rightly criticizes the model in the paper, which has several problematic theoretical assumptions from an Austrian perspective. However, unlike the Eggertsson DSGE model that I critique in EJW, Ohanian has at least done his historical homework and constructs a problematic model for estimating the size of the effect on GOOD historical evidence about policies that were in place. Eggertsson buys the Myth of Hoover (as well as some Myths of Roosevelt) that Ohanian rightly rejects.

Bottom line: Ohanian may be wrong about how much Hoover's labor market interventions mattered, but THAT they mattered and THAT they were to some (I'd say "significant") degree responsible for the depth, if not the length, of the GD is correct, and this paper provides more evidence for that claim.

One more for Barkley:

The flexibility of prices matters most in the face of an insufficient supply of money.

In 20-21, as I understand it, we were in a bust that arose from inflation of the war years and late teens. The federal government stayed out of the adjustment process and the decently high degree of price and wage flexibility (it ain't perfect!) enabled that recession to be fairly deep (unemployment rates were briefly above 10%) but short. From what I know, there was no secondary deflation concern, mostly because as you rightly note, there was not any crisis/panic in the financial markets.

1929-33 is different on two grounds. First, as this thread is discussing, Hoover (and Roosevelt later) dramatically decreased the flexibility of wages (and prices to a lesser degree). This became a problem when combined with the Fed's failure to stem the fall in the money supply associated with the problems in the banking system. All of that together gave us a decade of unemployment rates above 10%.

Since the GD, we certainly have less flexible wages than we did in 20-21 (though perhaps more so than in the Hoover years). What we haven't seen is the massive monetary deflation that would make inflexible wages a real problem.

Perhaps until now. Although in the current case, I wonder whether the high rates of unemployment we're seeing are not so much the result of inflexible wages, though I think that's a potential problem, but other labor market rigidities as well as other policy induced distortions that are simply preventing the structural adjustments necessary to re-employ that labor.

If the lost jobs are in sectors such as finance and construction, where it takes time to re-adjust human capital to where demand is now, then it's not surprising we'd see lingering unemployment while those adjustments take place. If the distortions of the boom were serious enough - drawing lots of labor into areas that were unsustainable - then it might well take significant time to adjust.

Still being in the middle of it, it's hard to say. We need some historical distance to know better.

While excessively high real wages might cause a surplus of labor and unemployment, focusing on high real wages is a mistake when describing the consequences of a shortage of money and a general glut of goods. While a careful reading of Ohanian's paper shows that he understands the central role of monetary disequilibrium, less careful readers may be confused.

Unemployment looks like a surplus of labor. Basic supply and demand analysis implies that a surplus of any good exists when its price is above equilibrium. A lower price of the good raises quantity demanded, reduces quantity supplied, and clears up the surplus. And so, a lower price is the solution to a surplus.

In the labor market, the wage is the price. Unemployment looks to be the result of wages above equilibrium. A lower wage rate should increase the quantity of labor demanded, decrease the quantity of labor supplied, and clear up the surplus. Isn't it obvious that lower wages are the solution to unemployment?

Basic supply and demand analysis is based up relative prices, the ratios of the money prices of different goods. A surplus exists for a good when its relative price is above equilibrium, and a lower relative price clears it up. In labor markets, the parallel concept is the real wage--the money wage relative to the money prices of goods and services. A surplus of labor (and apparently, unemployment,) would be due to real wages being above equilibrium. Wages are high relative to the prices of products. Real wages need to fall, clearing up the unemployment.

However, if the quantity of money is less than the demand to hold money, the result will be a net surplus of other goods, including currently-produced goods. Since firms will not produce goods they cannot sell, output declines. And with less output to be produced, firms demand less labor. A surplus of labor can exist without there being any change in real wages.

The problem, however, is money prices of all sorts, including resource prices like labor, are too high. A drop in all money prices, including money wages, will result in a higher real quantity of money. As the real quantity of money rises to meet the demand to hold money, the shrinking shortage of money is matched by a reduced surplus of goods. The growing demand for goods and services should result in firms producing more and hiring more labor. While both money prices and money wages need to fall, they need to fall at least roughly in proportion. Real wages are not too high and do not need to fall.

There is, however, a realistic disequilibrium process where excessively high real wages develop and appear to be the cause of unemployment. When there is a surplus of a good, each firm is motivated to lower money prices. And while firms are also motivated to cut wages when there is a surplus of labor, goods prices may adjust more rapidly than wages. If this occurs, then surpluses of products and labor will result in money prices falling more than money wages, so real wages rise.

The sticky adjustment of money wages slows the decrease in the prices of goods and services. From the point of view of each firm, it cannot continue to lower prices if costs do not fall as well. Production must be reduced. It is possible to conceive on an "equilibrium" where the firms are maximizing profit (or minimizing loss,) output matches sales (at a depressed level) and there is a surplus of labor, with continuing downward pressure on money wages.

In this "equilibrium," as the surplus of labor continues to result in falling wages, cost fall, and supply rises. The profit maximizing (or loss minimizing) level of output is higher. Firms lower prices less than in proportion to the decrease in costs, and real wages fall.

However, the key element of the process is that the lower prices are increasing the real quantity of money and clearing up the underlying shortage of money. This is what increases the demand for goods and services, and so increases the demand for labor as well.

The emphasis on the changes in real wages is just a red herring. The problem is that sticky prices are hindering the adjustment of the real quantity of money to the demand for money. Some prices can be more sticky than others, and the adjustment process involves increases in the relative prices of goods whose money prices are more sticky and decreases in the relative prices of the goods whose prices are more flexible.

Because unexpected changes in the price level (including wages,) involve shifts between debtors and creditors, and because deflation of prices increases the real rate of return on zero interest currency, there are advantages to some monetary institution that corrects for a shortage of money some other way. From that perspective, complaining about excessively high real wages during recession is counterproductive. Still, it is important to recognize that the market process that corrects for a shortage of money is a generalized deflation of all prices, including the prices of resources like labor. If some prices fail to adjust, that adds to the disruption.

DG,

Your statement of what economics can "prove" looks like a theological asertion to me.

Nikolaj,

I must confess I have little use for people who misrepresent what I say. I most definitely did not "agree" that the turnaround started prior to FDR getting in office. The only thing that turned around was the stock market. The stock market has risen nearly 50% this year in the US, but the economy itself has not yet turned around, although maybe it will soon (hopefully).

Regarding the early 20s recession, I would agree with Steve H. that it was mostly a readjustment from the wartime inflation. I am not sure exactly how it was triggered, but it looks like a classic "inventory adjustment" sort of recession, much like the one in 1983, with Jim Morley of Washington University recently arguing that these sorts of recessions have the characteristic that the harder and sharper they fall, the more rapidly and sharply they bounce back. We went down more rapidly in 1921 than 1929-33, and without all the mess in the financial and monetary system that was around in 1929-33, so we bounced back easily and rapidly (and, for those for whom all this really about today, today looks a lot more like 1929 than 1921, although Morley is the leading advocate of the idea that we might have a "V" type bounceback).

Greg,

Rockefeller may not have made such an argument, but Herbert Spencer did in a general way when he cooked up Social Darwinism earlier.

I would say that there remains debate about QWERTY and path dependence. Liebowitz and Margolis sharpened up the debate very much, but Paul David and others have come back with counterarguments. This is not a case of "thorough debunking." And, there are other examples around of such path dependence, with an example being given in a paper by an Austrian economist, Young-bak Choi, that I published in JEBO, on how it took so long (centuries) for the much more efficient Hang-gul alphabet to be adopted in Korea in place of the Chinese one.

Rose and Horwitz,

Actually, Ohanian's supposed facts are off, even though he does provide some interesting material on Hoover. This idea of him as this "ringmaster of cartels" is hilarious. FDR was far more of a ringmaster, with me agreeing that his policies of cartelization, later thrown out by the Supreme Court, probably did slow the recovery (and also were the parts of his policies most resembling "liberal" corporatist fascism, as argued by Johah Goldberg).

Part of the problem is that unions were pathetically weak in the 1929-30 and there was plenty of wage flexibility. Nominal wages were not rigid, but went down substantially during this period. Ohanian is simply wrong on this. Union membership was barely above 5% of the work force in 1930, this only rising (and very rapidly) after FDR's 1935 Wagner Act, going above 20% before 1940. Indeed, as noted above, wages were flexible downwards in the early GD and fell at nearly the rate as in 1921, but prices fell more, leading to rising real wages. But the disjuncture was greater in 1921, with prices falling much further than did nominal wages and with real wages rising much higher.

So, you guys and Ohanian need to get the historical facts straight. The Ohanian piece looks good, but is seriously flawed on its basic facts and story line, quite aside from the flaws of the DSGE model.

Rosser:

Hoover "jawboned" firms to maintain nominal wages in the face of a falling quantity of money and nominal expenditure. This hampered the market process (which I think works terribly) by which the real quantity of money will rise again to meet the demand.

Roosevelt left the gold standard and so reversed the shortage of money and began to create growing nominal expenditure. The primary impact was reduced money demand due to expecations of inflation, but the quantity of money could and did grow. Then, he promoted all sorts of cartelization, including support for unions, and so dampened the increase in output and employment generated by the growing nominal expenditures.

That is my understanding of the situation. While I can imagine using rising price floors to generate expecations of inflation and so lower money demand and cause inflation, I think there are better approaches--approaches that don't disrupt production and competition for decades.

Horwitz still refers to "an insufficient supply of money" and Woolsey to "a shortage of money."

It looks like there's no killing that idea. Like Rasputin, you can stab it, strangle it, and drown it, and like Dracula, drive a stake through its heart, and it'll still keep popping up, mocking physiology and logic.

Woolsey:

But nominal wages did fall in 1929-33, in contrast to the "bounceback" recession of Reagan in 1983. Sorry, but all this focus on wage adjustments is just a waste of time. And none of you have addressed the crucial point of what Milton Friedman labeled the "unbalanced deflation," namely the serious problems that nominally fixed debts rose as prices fell, thereby generating much further havoc, and with falling nominal wages in connection with that leading to further bankruptcies and financial collapses. This is all beyond this ridiculous story of equilibrium wage adjustments in the midst of massive disequlibria.

Meant to say the "real value" of nominally fixed debts rose.

Barkley -- my examples are specific. The Rockefeller, Sr. myth IS a myth, one explained by David Hull. The myth that the "Dvorak" was more efficient than the QWERTY system is also myth, as recounted by Liebowitz and Margolis.

NEITHER example addresses ANY other literature or argument.

These are simply examples of false "facts" used for pedagogic reasons. There are a lot of examples, these are simply one's I thought my economist audience would appreciate.

Larger social science of economic theory issues are really beside the point, and have nothing to do with my uses of these myths here.

Barkley writes:

"Rockefeller may not have made such an argument, but Herbert Spencer did in a general way when he cooked up Social Darwinism earlier.

I would say that there remains debate about QWERTY and path dependence. Liebowitz and Margolis sharpened up the debate very much, but Paul David and others have come back with counterarguments. This is not a case of "thorough debunking." And, there are other examples around of such path dependence, with an example being given in a paper by an Austrian economist, Young-bak Choi, that I published in JEBO, on how it took so long (centuries) for the much more efficient Hang-gul alphabet to be adopted in Korea in place of the Chinese one."

Barkley -- this is actually an illustration of the "Godwin's Law" point, the issue at hand gets distracted from by the "red meat" of the issues of social darwinism and path dependence. I'm not addressing EITHER of those red meat issues, I'm talking about fake facts being used for pedagogic purposes -- when those fake facts are too good to let go of, despite the contrary truth of the matter.

I wrote:

"Barkley -- my examples are specific. The Rockefeller, Sr. myth IS a myth, one explained by David Hull. The myth that the "Dvorak" was more efficient than the QWERTY system is also myth, as recounted by Liebowitz and Margolis.

NEITHER example addresses ANY other literature or argument.

These are simply examples of false "facts" used for pedagogic reasons. There are a lot of examples, these are simply one's I thought my economist audience would appreciate.

Larger social science or economic theory issues are really beside the point, and have nothing to do with my uses of these myths here.

Barkley writes:

"Rockefeller may not have made such an argument, but Herbert Spencer did in a general way when he cooked up Social Darwinism earlier.

I would say that there remains debate about QWERTY and path dependence. Liebowitz and Margolis sharpened up the debate very much, but Paul David and others have come back with counterarguments. This is not a case of "thorough debunking." And, there are other examples around of such path dependence, with an example being given in a paper by an Austrian economist, Young-bak Choi, that I published in JEBO, on how it took so long (centuries) for the much more efficient Hang-gul alphabet to be adopted in Korea in place of the Chinese one."

Barkley,

You wrote,

"DG

Your statement of what economics can 'prove' looks like a theological asertion to me."

Economics without empirical verification is no more a religion than logic or mathematics without empirical verification.

Economics is not about visible events, but the invisible cause and effect relations between them? How could you observe them?

How could you observe the Invisible Hand?

Greg,

Oh, so you want to be literal. Sure there are myths, and the Rockefeller story is one. But in fact the claims of Liebowtz and Margolis turn out to be highly questionable to the point of themselves having become myths. Their data came from a man who personally hated Dvorak and destroyed his original sources. The fastest typist in the world uses Dvorak. So, be careful about what you call "myths."

Obviously the survey cited below isn't definitive proof, but it may suggest that wage deflation is here. From CNN Money, "Cut My Pay...Please!":

"With unemployment as high as 9.4% and job prospects scarce, job seekers are willing to accept as little as half of what they were making before, if it means finding a job.

In a recent survey, 65% of out-of-work respondents reported willingness to accept wages up to 30% lower than their previous compensation. And, 3% and 4%, respectively, said they would accept up to 40% and 50% of prior wages, according to the 2009 Annual Career Fair Survey released by Next Steps Career Solutions.

"In the old days people would expect to get at least a 10%-15% bump when they were making a transition from this job to the next," said Paul Bernard, an executive coach and career management adviser who runs his own firm. Now, "being asked to take cuts in the 20%+ range is pretty standard."

http://money.cnn.com/2009/08/28/news/economy/paycuts/index.htm?postversion=2009083113

I was about to ask why a group of Austrian economists were restricting themselves to considering the effects of wage flexibility *in the aggregate,* rather than considering the question of sectoral malinvestment, particularly construction and related industries. But then I found that Steve Horwitz had written:

"If the lost jobs are in sectors such as finance and construction, where it takes time to re-adjust human capital to where demand is now, then it's not surprising we'd see lingering unemployment while those adjustments take place. If the distortions of the boom were serious enough - drawing lots of labor into areas that were unsustainable - then it might well take significant time to adjust."

If workers are moving from "overheated" sector A into sector B, which has been more stable and has a sufficient supply of labor, then we would expect the marginal productivity of the displaced workers' labor to be somewhat lower in sector B. But as Steve suggests, even if workers like those in the CNN Money article are willing to take significantly lower pay, there simply may not be jobs for these people at all, because the malinvestments in residential construction and mortgage finance, say, have deprived other sectors of capital (reinforcing the lower marginal productivity of labor). Barkley Rosser mentioned the Bryan Caplan article which disaggregated the wage data for 1920-21, but not for 1929-33. Rather than wage flexibility alone, is it possible then, that there were more structural imbalances 1929-33 than in 1920-21 which would in part explain the more protracted adjustment period?

Rosser wrote - "And none of you have addressed the crucial point of what Milton Friedman labeled the "unbalanced deflation," namely the serious problems that nominally fixed debts rose as prices fell, thereby generating much further havoc, and with falling nominal wages in connection with that leading to further bankruptcies and financial collapses."

That debt was part of the problem to begin with. Much like with the current situation, the debt was unsustainable, fueled by asinine Fed Reserve policies that also fueled the "Great Bull Market" of the 1920's. People relied on too much credit back then, just as they did now.

The "unbalanced deflation" was part of the market restructuring process. They made bad bets and then needed to pay the piper, that is until Hoover tried, ineptly, for the State to move in for bailout....er I mean relief (Smoot-Hawley and Agricultural Marketing Act)
Prices couldn't fall fast enough, as they did in the 21 recession, Hoover saw the writing on the wall and jawboned industry to prop up his failed policies with wage freezes which caused more unemployment.

First, I fully agree with Steve’s praise of the quality of Ohanian historical research.

Second, Ohanian’s paper seeds of a major research programme for him and others.

Rothbard was on to something when he identified Hoover’s cartelisation attempts as a major factor in prolonging the 1929 depression.

Ohanian elaborates by identifying a specific and credible device that Hoover could use to police cheating within the high-wage cartels that he sponsored. Hoover had a bargaining chip against recalcitrant employers – no federal protection against unions. The higher real wages offered by the cartels, and threat of federal aid in the suppression of strikes, were his bargaining chips to quell union demands for more.

The challenge his research faces is well stated in a April 2007 preliminary draft where Ohanian said ‘An important aspect of understanding the Contraction is understanding the apparent dysfunction of the industrial labour market. The challenged for any theory of this dysfunction is that there was no legislated labour policy shift , as was the case with the NIRA and the NLRA, nor was there particularly extensive union strike activity during the early 1930s.’

Further historical research may ascertain how successfully the high wage cartels were policed by Hoover, and how and why these cartels unravelled in late 1931. Links to other parts of the Hoover New Deal such as the Reconstruction Finance Corporation and the prospect of higher tariffs as potential cartel policing devices are other lines of good inquiry.

The whole interventionist rationale is based on the premise that a recession or depression is a bad thing. But it is a good thing.

It is the withdrawal from the squandering malinvestments and overconsumption of the boom period, and from the markets under attack by rioters and looters in the legislatures and courts as well as the streets.

It is the reaccumulation and safeguarding of the capital that will be needed for reconstruction, when and if the opportunity for it arises.

So leave the recession or depression alone. It is just what we need.

And, once again, for the benefit of Woolsey and Horwitz,

from Mises, Human Action, 3rd Rev Ed.:

"The services which money renders can be neither improved nor repaired 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.” P 421

“It is possible by means of an increase in the quantity of money to delay or to interrupt this process of adjustment. It is impossible either to make it superfluous or less painful for those concerned.” P 431

A few other observations.

Woolsey continues to confuse an increase in the purchasing power of money with an increase in the quantity of money. They're not the same thing. If the total supply of money consists of 100 dollars, and the purchasing power of each one goes up, there are still just 100 dollars. There is an increase in the purchasing power but not in the quantity of money.

You cannot get away from the fact that the market wage is the equilibrium, full employment wage, and any wage higher than that a disequilibrium, unemployment wage.

So, while it is true that a contraction of the money supply would hurt those with fixed obligations, such as mortgages, unemployment, and, therefore, a higher than market, unemployment wage, would hurt them even more.

A reflation would help them, and wouldn't hurt the rest of us, if it followed the deflation immediately. But once the market had adjusted to the deflation, a reflation would be more of the poison that sickened us in the first place.

And, whatever will cure us, that isn't it.

Barkely:

An unexpected change in the price level transfers funds between debtors and creditors.

With deflation, the transfer is from debtors to creditors. To the degree debtors continue to pay their debts, they spend less and their creditors spend more. To the degree they default, the former debtors spend more and the former creditors spend less. Presumably, this would just be a dampening of the first process. Default limits the tranfer.

Closing down operations during default greatly disrupts production. Reorganization or reallocation of resources is costly in a variety of ways.

The only way that the transfer between debtors and creditors will futher impact aggregate nominal expenditure is if either the nominal quantity of money is reduced or the demand for money is increased. There are plausible mechanisms for both. However, most of the mechanisms are really adverse impacts of expectation of default. Contracts were made with a low expecation of deflation, and now the deflation is eminent. Creditors flee to base money.

Using supply-side mechanisms to prevent deflation, however, doesn't help much. Jawboning firms to keep them from cutting wages (or prices,) much less imposing price or wage floors just results in the falling expenditure causing output to fall, which causes much the same problems with expecations of default.

I favor monetary institutions where the nominal quantity of money adjusts enough to meet the demand to hold money with nominal income growing at 3%. One reason is to avoid making every nominal contract a speculation on fluctuations in the demand to hold money. That doesn't mean that the market process that returns the market to equilibrium isn't decreaced prices and wages. Seeking to stop the symptom by voluntary or compulsorary price floors could leave the economy in permanent recession. If the only alternative is the deflationary wringer, then even if all debts must be defaulted and a base money only economy must recover through the mechanism of the Pigou effect, then that beats what.. centuries of depression?

Justin and Rose,

So, which is the problem, that prices did not adjust fast enough or that wages did not adjust fast enough in the Hoover years? It looks like prices fell more sharply in 1921 while nominal wages fell at nearly the same rate as after 1929. So, the bounceback came from real wages rising much more in 1921 than after 1929. Perhaps there should have been more nominal wage rigidity after 1929.

DG,

Oh, recession and depression are good things? I shall not play Godwin games again. I shall simply ask: are you employed? Are you in danger of losing your job? Have any of your family members or close friends been laid off and unable to find another job? If so, good luck telling then what a good thing this is. I am sure quotations from Mises will convince them.

Woolsey,

Creditors tend to be higher income than debtors. We would thus expect them to have lower mpc's (if I can dare to mention such a concept on this blog) than debtors. If so, then the increase in their spending from receiving transfers from the debtors can be expected not to offset the decline in spending by the debtors.

Bark,

Alright, for your benefit, rather than saying that the recession is a good thing, I'll say that it's a necessary evil.

Barkley, thanks for the update on Dvorak & QWERTY.

Facts matter, and knowing who cares about the facts also matters.

"Creditors tend to be higher income than debtors. We would thus expect them to have lower mpc's --"

Now now, let's not be silly and make completely unsustainable categorical claims about people. Would you, please, theorize on this a bit, and explain how it would hold universally that people with higher incomes tend to have a lower mpc than people with lower incomes.

I have always thought that the propensity to save is preference that has very little to do with one's 'class affiliation' or which income bracket one belongs to.

Regarding QWERTY and Dvorak,

The world record holder for the fastest typing on a normal keyboard is Barbara Blackburn who used a simplified Dvorak keyboard.

However this *isn't* the fastest type of keyboard. The fastest type are the 5 button chord keyboards used by stenographers. Operators of these may do 300wpm, twice as fast as the world record mentioned.

The whole issue is much more complicated than the proponents (and some detractors) of "path-dependency" like to claim. See the book "The Design of Everyday Things".

Santtu,

Sorry, but this is not a matter of theory, but one of empirics. I did not say that it was true all the time, but most of the empirical evidence I have seen leans to the story of its being true a lot more frequently than it is false.

Current,

Thanks for the further update. I know it is a complicated matter without definite or clear answers, which is why the initial round of articles on this on both sides were off base.

Barkly,

recession is good and healthy phenomenon, because recession eliminates misallocations of capital from the boom, spurred by easy money policy an artificial lowering of interest rates. The problem is not recession, but the boom that brings about malinvestment and distortions in capital structure. If you disagree, then you must agree with the notion that inflationary boom is a good thing (which is actually quite common thing in mainstream economics Wonderland where recessions are treated like some kind of natural disaster that must be attempted to avoid, overlooking any causal link between the boom and bust). Question "whether you were employed?" you posed to DG Lesvic is demagoguery that reminds me of similar complaints I used to listen from communists in Eastern Europe a while ago, who were so concerned with the people who are going to be unemployed because of privatization, not understanding that jobs in bankrupted state enterprise were not real, but fictive jobs. By the same token, jobs that exist in industries plagued by chronic malinvestment and structural distortions are not real jobs, and have to be liquidated sooner or later. Better (less expensive) sooner than later.

As for GD, conventional wisdom is that severity of it should be ascribed to Fed letting prices to fall. But, when we take into account Depression 1920-21 that wisdom, as we saw, falls apart. Now, you are trying desperately to salvage the thesis by shifting its definition. All of the sudden, price decline is not paramount, but different nature of two depressions, one "wartime", another one not, and that durability of price decline was defining feature of GD. But, that amounts to no definition at all. Like Moliers' definition of opium. GD was severe and long because it was severe and long :)

Ohanian's paper is an attempt to address this problem of logical inconsistency of traditional explanation of GD. Since every depression entails general price decline why GD was so specific? What was different in it, compared to other recessions and depressions, since price decline was a common feature of all of them? He proposes wage rigidity. I am not sure to what extent this solution holds water, but something must be said, some explanation must be found for specificity of GD. You are castigating him for noticing illogical character of Friedman's explanation of reasons for severity of GD. You are entitled to criticize validity of his solution, but not to pretend that the very problem he attempted to solve does not exist either.

Barkley:

Regardless of any difference in marginal propensities to consume between debtors and creditors, the only way that transfers between them can impact nominal income is if either the quantity of money falls, or else the demand to hold money rises. How much people save only impacts noominal income if it increases the demand for money without there being a matching increase in the quantity of money. While there are many possible processes that could lead from more saving to an excess demand for money, that is the key issue. Increased real debts due to deflation needs to cause an increase in the demand or decrease in the quantity of money if a lower price level itself pushes down the equilibrium price level.

My judgement is that the key process is that deflation creates an expectation of default and so an increase in the demaand for base money. Impacts on saving, and so the natural interest rate, and so the opportunity cost of holding base money, and so the long run demand for money seem less important to me. But, it is a possible avenue.

I for one am unimpressed by Ohanian's use of sources. He attributes a reduction in "steady state hours and capital stocks" to wage maintenance and work-sharing and justifies his inclusion of work-sharing on the basis of a feature of his model that ties capital input to the per capita length of the workweek and justifies that feature by reference to a source that's relevance is questionable but that is also atrociously misinformed about the extent and of empirical literature on productivity and working time.

see: http://econospeak.blogspot.com/2009/08/and-it-aint-shinola-ii.html

Well you know some people are scared about a resurgence of Austrian Economics when they start trolling Austrian blogs. Keep up the good work Steve.

Woolsey wrote,

"I favor monetary institutions where the nominal quantity of money adjusts enough to meet the demand to hold money with nominal income growing at 3%. One reason is to avoid making every nominal contract a speculation on fluctuations in the demand to hold money."

But after the market had factored the 3% inflation into its calculations, it'd be right back where it started, with all the same speculations, and complicated by the inflation.

"...money could not be pumped into the system to combat a supposed increase in the value of money without distorting the previous exchange values between the various goods." Rothbard, Man, Economy, and State, P 744:

You couldn’t stabilize the value of money without also stabilizing the value of everything bought and sold against it. So it wouldn't be enough to pump a little money into the system. You would need all-around price control.

There is no stabilization even in a graveyard.

The uncertainty of the future and speculative nature of all human action is a fact of life, and cannot be eliminated without eliminating human action and life itself.

Barkley, this is one reason people hand on to a useful myth, rather than concern themselves with the truth. The myth is just as useful -- and often moreso -- than the truth of the matter for the rhetorical purposes at hand.

My original discussion was about the role of myth in scientific explication. It wasn't about economics or typewriters.


Barkley wrote:

"Current,

Thanks for the further update. I know it is a complicated matter without definite or clear answers, which is why the initial round of articles on this on both sides were off base."

I see what you mean Greg, and I agree. There's lots of this in physical science teaching too. The myth serves a function like a moral fable, but it is something evolved rather than designed.

I'm afraid that Barkley Rosser is propagating some misinformation about the QWERTY debate.

Paul David has retired from the field of battle, apparently. The most recent thing I can find from him is a piece from 2000, which is simply a heavily edited version of a 1997 paper, in which he tries to change the subject with a bunch of double talk designed to obscure the fact that he'd called QWERTY evidence of market failure due to network effects. He's conceding he was wrong, by getting as far away from that as possible.

Indeed, David's research assistant at Stanford when he published his famous AER piece, 'Clio and the Economics of QWERTY', Doug Puffert, has, as of 2008, EXPLICITLY conceded that Liebowitz and Margolis were right that the QWERTY keyboard is not a market failure since the benefits, if any, of switching to it would never exceed the costs.

From Puffert's encyclopedia entry on 'Path Dependence' at the EH.Net site:

http://eh.net/encyclopedia/article/puffert.path.dependence

' For Liebowitz and Margolis, it was most important to show that the costs of switching to an alternative keyboard would outweigh any benefits, so that there is no market failure in remaining with the QWERTY standard. ***This claim appears to stand.*** David had made no explicit claim for market failure, but Liebowitz and Margolis — as well, indeed, as some supporters of David’s account — took that as the main issue at stake in David’s argument.'

[My emphasis in the above.]

That is total capitulation, albeit with a bit of face saving about David not having 'explicit'[ly] made such a claim.

As to Barkley's colorful claim about L&M getting their information from someone who hated Dvorak--which I take to be a reference to Earl Strong and his GSA study that concluded that QWERTY was probably better than DSK--that is misleading. Stan and Steve got SOME information from his study.

They also got information from an Australian Post Office study, an Oregon State Univ study, a Western Electric study, and several others. None of which support Lt. Cmdr. August Dvorak's claims of 20-40% improvement in speed with two weeks of training.

As for the reference to the world's fastest typist, I suppose that would be the late Barbara Blackburn of Salem, Oregon. Who, according to that scholarly powerhouse The Guinness Book of World Records, held the title by typing all of 8% faster than the man in second place who used a qwerty typewriter. Which is less than halfway to the lower bound of Dvorak's claims.

But, there is something interesting about Barbara Blackburn's story. She inadvertently refutes, in an interview, three explicit claims of Paul David (and repeated by Doug Puffert in the encyclopedia article linked to above) about the training of typists in the first half of the twentieth century in business firms in America.

Here's David (1986):

' Prior to the growth of the personal market for typewriters the purchasers of the hardware typically were business firms, and therefore distinct from the owners of typing skills. Few incentives existed at the time, or later, for any one business to invest in providing its employees with a form of general human capital which so readily could be taken elsewhere. (Notice that it was the wartime US Navy, not your typical employer, who undertook the experiment of retraining typists on the Dvorak keyboard.) The investment decisions of would-be touch typists, and the consequent supply of clerical and office workers with those skills, therefore remained largely beyond the individual control of the buyers in the market for typewriter hardware.'

The Blackburn interview which had been up at Dvorak Int'l's website has disappeared, along with Dvorak Int'l. However parts of it are excerpted here:

http://answers.yahoo.com/question/index?qid=20060702093109AAKLZ3S

Where we find that Blackburn, who had flunked typing in high school, redeemed herself in 1938 when a typewriter salesman offered her the chance to learn on his DSK machine, while she was studying at a business college (not in the wartime US Navy). She then purchased such a machine and carried it with her wherever she worked. Making an above average income by being in strong demand because of her productivity.

Obviously David is ill-informed about the investment decision makers of the time.

The comments to this entry are closed.