This archived content originally appeared at, the predecessor site to, and does not carry the sponsorship of the Cato Institute.

An Austere Recovery

As I pointed out in a previous post, in the course the BBC/LSE “Hayek versus Keynes” debate the Keynesian side made some claims to which I had no opportunity to respond. My earlier post addressed some of them, but left another alone. This was Lord Skidelsky’s claim, aimed at Great Britain’s current riot-provoking austerity campaign, that no government has ever achieved a speedy recovery from a recession by clamping down on its spending or reducing its indebtedness.

But there is at least one instance of economic recovery–and hardly a trivial one–that contradicts, or at least very much appears to contradict, Lord Skidelsky’s claim. This is the United States’ rapid recovery from the deep recession into which it sank in the last half of 1920.

In many respects the boom-bust cycle that started in April 1919 was typically “Hayekian”: during the boom year ending in April 1920 the Fed held its rediscount rate at 4 percent despite rising money market rates. Commercial banks took advantage of the low rate–as they’d actually been encouraged to do by the Fed–by borrowing from the Fed in order to re-lend at a profit, causing bank loans and investments to increase by just over 25 percent. General prices, and prices of commodities and land and other factors of production especially, in turn rose more rapidly than they had since the Civil War, exacerbating a gold drain that had begun with the armistice. Under the circumstances a reversal was only a matter of time.

When it came, the reversal was both sudden and sharp. Commodity prices tumbled from an index value of 248 in May 1920 to one of just 141 the following August, while consumer prices witnessed their greatest rate of deflation ever. Businesses were unable to pay their bills, industrial production fell by an unheard of 30 percent, and almost 5 millions workers lost their jobs, bringing the unemployment rate, which had been less than 2 percent, to just below 12 percent. Yet by August 1921 recovery was well underway. What’s more, it was so swift that by the spring of 1923 unemployment had given way to a pronounced labor shortage, while industrial production reached a new peak.

Did the U.S. government hasten the recovery by means of deficit spending and other “stimulus” programs? Not in the least. instead, it stuck to conducting business as usual which, in those naive days before Keynes revealed that prudence and thrift were shopworn Victorian shibboleths, meant reverting to its prewar budget and retiring its wartime debt. In other words, it followed what would today be called an “austerity” policy, and did so to a degree that makes recent austerity measures in Great Britain and the U.S. seem downright profligate. Instead of spending more than it had been, the Harding administration steadily cut expenditures, exclusive of debt retirement, from just over $6.4 billion in fiscal 1920 to just under $3.3 billion in fiscal 1923–a whopping 45 percent! As a percentage of GNP, Randy Holcombe shows, Federal outlays fell from just over 7 percent to well under 4 percent.* And although its revenues also declined over the same period, from about $6.7 billion to about $4 billion, the government nevertheless devoted a large share–almost $1 billion–to reducing its indebtedness. As Benjamin Anderson, who was at the time an economist employed by Chase National Bank, observes in Economics and the Public Welfare (1949),

The idea that an unbalanced budget with vast pump-priming government expenditure is a necessary means of getting out of a depression received no consideration at all. It was not regarded as the function of the government to provide money to make business activity. It was rather the business of the United States Treasury to look after the solvency of the government, and the most important relief that the government felt that it could afford to business was to reduce as much as possible the amount of government expenditure, which had risen to great heights during the war; to reduce taxes–but not much; and to reduce public debt.*

Turning to monetary policy, although easy money did contribute somewhat to the recovery, the contribution was minor and largely unintended. Thus while the Fed banks gradually lowered their discount rate from 7 to 4 percent between 1921 and 1922, 4 percent was not especially low in light of the rapid deflation then in progress. And despite the rate lowering commercial bank rediscounts declined, as banks preferred reducing their indebtedness to the Fed to taking advantage (as they regretted having done earlier) of opportunities to re-lend borrowed funds at a profit. The Fed also made what was at the time an unusually large open-market purchase of government securities. But this only served to further reduce commercial bank rediscounts, and was moreover done, not with any intent of stimulating recovery, but solely so that the Fed could earn enough revenue to cover its expenses and pay promised dividends to its commercial-bank shareholders.

Proponents of Keynesian pump-priming often berate the Hoover administration for its “liquidationist” strategy for dealing with the outbreak of the Great Depression–forgetting that it was Hoover himself who caricatured Andrew Mellon, his Secretary of the Treasury, as someone who wished to “liquidate” the stock market, farmers, real estate, and so forth, and who took pride in not having followed his advice. But Mellon was also Harding’s Secretary of the Treasury; and Harding, unlike Hoover, trusted him. It is one of the greater ironies of economic history that “liquidationist” policies, including government austerity, are blamed for prolonging a depression for which those policies were set aside, while being denied credit for perhaps helping to end one in which they really were put into practice.

*Sentences added 8/20/2011


On very rare occasions, when it regards a broadcast as having been particularly well-received, BBC Radio 4 rebroadcasts the same program for a third time. I’m very pleased to say that it has chosen the Hayek versus Keynes debate for this honor, and will therefore air it again, for the sake of regular listeners who missed it (and can’t be bothered with podcasts) on August 24th. Yo Hayek!

  •!/trollyball.mvp Scott Burns

    We talked briefly about this timeframe in my economic history class. My professor, who was very good albeit overtly liberal, showed us all the statistics about how steep the downturn was, only to summarize the recovery in two sentences: “But this was merely a post-war ‘refitting’ slump. By 1922 everything was back to normal.” Within 20 second, we had fastforwarded to 1929. No gold exchange standard. Nothing. Just ‘recession – mysteriously fast recovery – roaring 20s – Andrew Mellon = evil – then the stock collapse.

    I raised my hand and respectfully asked, “Given what we’ve learned about Keynes, shouldn’t the worldwide ‘Great Depression’ have occurred in 1920–not 1929? Surely the collapse of government spending across the world at the end of the war as “austerity” measures to repay debts combined with drastic deflation, a reimposition of the ‘gold standard,’ and massive technological changes would’ve been Keynes perfect formula for depression…not the comparatively benign events of 1929.”

    To his credit, he took about 30 seconds to think and then humbly responded: “That’s a good point. But I think most historians contend Keyne’s ‘General Theory’ only applied to the English situation.” I disagree, as I think Keynes would when he wrote foreign leaders with advice. Besides, what’s so “general” about his theory if it only applies to the Uk?

    Later on in the class he came back to my question and said: “You know what, I forgot to mention prices were less sticky in 1921 than they were in 1929.” Better, but not what exactly I’d call a slam dunk argument.

    • George Selgin

      I wonder why anyone should imagine the structural adjustments following a major war–which, after all, devotes many resources to doing things that aren’t done at all in peacetime–should be easier than those following a peacetime boom that merely involves relative malinvestment. I mean, is it easier for a returning soldier who has spent his last years acquiring no human capital save that necessary for killing people, to find a job than someone laid off from from constructing homes? I bet some vets could answer that one, and emphatically, too.

      •!/trollyball.mvp Scott Burns

        In regards to WW1, I don’t think there’s any good rebuttal for Krugmanite Keynesians. Government spending actually decline and there was actually a budget surplus.

        In regards to WW2, I’d imagine most Keynesians would respond: “But we did have a fairly noticeable decline in Y after the war. Luckily, the G.I. bill prevented there from being a “glut” of idle workers by paying for many soldiers to go to college, and we still kept many troops on the payroll in Korea, etc. Most importantly, people finally had an outlet to spend their war-time savings as consumer production rose. Hence, recovery was quick.”

  • Mike Sproul

    Has Skidelsky never heard of Sargent’s “The Ends of Four Big Inflations”? In every case the countries were suffering not just from inflation, but from recession, and fiscal reforms (spend less, tax more) brought about recovery from both inflation and recession.

    • George Selgin

      I don’t know the answer here; but of course he would be free to argue that because they clearly didn’t involve any collapse of aggregate demand, but quite the contrary, the episodes in question are not the sort of recessions that he has in mind. For my part, I would reject the suggestion that “fiscal reform” alone ever suffices to combat inflation. Monetary restraint is what’s crucial; though in the absence of fiscal restraint the likelihood of monetary expansion tends to increase.

  • David Stinson

    Am I correct in thinking that the 1920/21 downturn was perhaps sharper and longer than it could have been had monetary policy followed a less-than-zero rule (i.e., assuming that they still created the boom but figured things out coincident with the bust) ? Would they have been able to do that under whatever form of the gold standard existed at the time?

    • George Selgin

      Hard to say, David. The fact is that while commodity and other prices appeared remarkably flexible downward at the time, hourly wages didn’t adjust downward all that much–a matter of a little bit more than 10 percent–between 1920 and 1922. One theory is that a reduced flow of immigrants dating from immigration restrictions imposed during the war had resulted in a rising equilibrium real wage rates. In any case, the fact that nominal wage rates didn’t end up adjusting down all that much suggests that monetary policy may not have been all far off from a “productivity norm” ideal.

      • David Stinson

        Interesting. Thanks for your response. But is nominal wage stickiness not implied by the high unemployment rates or do you attribute the unemployment to sorting out the malinvestments (i.e., structural reasons)?

        The motivation for the question was whether the recession might even have been shorter had monetary policy been ideal, notwithstanding the apparently pro-cyclical (in Keynesian terms, but perhaps in reality counter-cyclical) fiscal policy.

        • George Selgin

          There certainly appears to have been a temporary money shortage, which an “ideal” policy might have averted–this is evidenced by statistics showing a sharp drop in nominal GNP. But rapid downward adjustments in prices generally, and to a lesser extent in wages, seem to have sufficed to bring rapid recovery, in part by reviving investment spending. This isn’t to deny that stable spending all along would have been better, and it certainly isn’t to suggest that there was nothing wrong with the Great Contraction of the following decade that more “austere” policies couldn’t have made up for. What the 1920-22 episode seems to suggest is not that monetary stability is unimportant, but rather than fiscal profligacy isn’t essential–if indeed it isn’t counterproductive–when it comes to coming out of a bust.

          • David Stinson

            Agreed. Thanks for your response, George.

  • Robert Dell

    I believe the period between 1945 and 1947, when federal spending was cut from 42 percent of GDP to 15 percent, is another good counter Keynesian example. The unemployment rate over that period stayed below 3.6 percent and real GDP grew by 9.6 percent. According to David Henderson, “the postwar bust that so many Keynesians expected to happen never did.” Historian Arthur Herman found that the biggest contributor to growth during this period was a steady rise in private capital investment, which jumped from $11 billion in 1945 to $41 billion in 1948. His conclusion is that “the root of prosperity was the same then as now: private capital formation and investment, not consumer or government spending.”

    • George Selgin

      Yes and no, Robert. In fact, despite what Scott (playing Devil’s advocate) observes, there was a very severe decline in government spending after the second World War ended, and the episode does show that you can have such a decline and yet not have a deep and protracted recession or a depression–a very important point. But it isn’t quite so clear a case of “austere” policies being followed after a steep crash, with recovery nonetheless occurring rapidly. That was what Professor Skidelsky suggested never happened. It happens, though, that he also asserted that renewed New-Deal type policies helped after WWII, a point which (as I noted in my earlier post lined above) is not consistent with Truman’s actual policies, which actually took a marked pro-business turn after the war, as one element of Cold War reaction to the Communist threat (or bogeyman, if you prefer).

  • Paul Marks

    The perils of reading one’s e.mails backwards (most recent first – oldest last).

    I just wrote out a comparison of 1921 and 1929 – and you had already done it.

  • Izzy


    Here is the full Hayek v Keynes debate on youtube