The New Deal and the Great Depression

Like the folks writing at Mahalanobis, Marginal Revolution and Free Exchange, I was rather surprised to see Berkeley Professor Brad DeLong claim, “A normal person would not argue that the New Deal prolonged the Great Depression.” Since Brad is a smart guy, I think it might be time for me to acknowledge my freakiness.

I divide the U.S. Great Depression (1929-1939) into three episodes: (1) the initial downturn (1929-30), catastrophic free fall (1931-32), and slow recovery (1933-39). I believe there were different factors in play in each.

The initial downturn does not strike me as all that remarkable. Had the economy begun to recover in the middle of 1930, as the steady appreciation in stock prices from December 1929 through April 1930 suggested that it might, that episode would not look that different from any of a number of other historical business downturns. The same kinds of forces that produce a typical economic downturn can offer a quite satisfactory account of what happened in 1929-30. Among these, a monetary contraction in 1928-29 likely contributed to that downturn, as did an exogenous drop in consumption and investment spending.

What really distinguished this episode from a typical downturn was not the severity of the initial decline, but the fact that, unlike a typical business cycle, things began to deteriorate quite dramatically after 1930. In my opinion, one of the reasons for that is the collapse of the money supply, which led the economy into a ferocious deflation. I’ve argued that the gold standard contributed to that, and certainly each country is observed to begin to recover from the Great Depression as soon as it suspended gold convertibility. I also believe that the bank panics played an important role in propagating that second phase, and am quite happy to grant Daniel Gross that New Deal financial innovations such as the FDIC and FSLIC were unambiguously helpful in correcting some of those problems.

After 1933, the economy began to grow again, with real GDP increasing at an average compound rate of 6.7% per year from 1933-1939. However, although growth over this period was strong, the unemployment rate stayed high, and there surely remained substantial excess production capabilities. So, in my mind, the third part of the question of “What caused the Great Depression?” is to explain why did the recovery take as long as it did after the contractionary monetary forces were removed?

What is supposed to help the economy recover is that a substantial pool of unemployed workers should result in a fall in wages and prices that would restore equilibrium in the labor market, as long as the government just keeps the money supply from falling (which, as just noted, the Fed failed most spectacularly at doing during 1931-32). And yet, in the midst of quite significant unemployment, between 1933 and 1934, average hourly earnings increased by over 25% in sectors such as iron and steel, furniture, and cement, and over 50% at lumber mills. How could that be?

Those numbers, by the way, come from a paper by UCLA Professors Harold Cole and Lee Ohanian that appeared in the Journal of Political Economy in 2004. And their answer to the question is that the persistence of so many unemployed workers was due in no small part to the National Industrial Recovery Act of 1933 and the National Labor Relations Act of 1935.

Cole and Ohanian noted that many in the Roosevelt Administration believed that the severity of the Depression was due to excessive business competition that led to wages and prices that were too low. I actually agree, in a perverse sense, with part of that diagnosis– I see the rapid deflation of 1929-33 as quite destabilizing. But I’m inclined to believe that the way to fix that would have been through a monetary and fiscal expansion rather than trying to lift nominal wages and prices back up by sheer government fiat.

The purpose of the NIRA and NLRA was to promote labor and trade practice provisions so as to limit the extent of competition between firms and competition between workers. Among the NIRA codes that Cole and Ohanian highlight include minimum prices below which firms were not allowed to sell their products, restrictions on productive capacity and the amount that could be produced, and limitations on the workweek. Cole and Ohanian concluded on the basis of model simulations that these kinds of New Deal policies might have accounted for 60% of the persistence in the output gap.

Outside of manufacturing, the Agricultural Adjustment Act of 1933 and Soil Conservation and Domestic Allotment Act of 1936 — paying farmers not to grow wheat– were designed with the specific goal of reducing agricultural production in order to raise agricultural prices. State regulatory commissions like the Texas Railroad Commission ordered oil producers to cut back production in an effort to increase oil prices.

The notion that if we can just create more monopoly power for every single sector of the economy, encouraging every sector to produce less so they can raise their wages and prices, that we will then somehow make everybody richer, is so spectacularly wrong-headed that I would be just as dumbfounded to find that Brad De Long believes it as he seems to be by those of us who maintain that some aspects of New Deal policy surely did make the recovery from the Great Depression slower.

I openly confess to believing that government policies that were explicitly designed to limit manufacturing, agricultural, and mining output may indeed have had the effect of limiting manufacturing, agricultural, and mining output.

Technorati Tags: ,

34 thoughts on “The New Deal and the Great Depression

  1. Steven Vickers

    Ah, but Prof. Hamilton, you’re an economist, so by Gross’ logic, you aren’t qualified to speak on economic matters. Says Gross: “The argument that the New Deal’s efforts ‘perhaps had prolonged, the Depression,’ is likewise a canard. One would be very hard-pressed to find a serious professional historian–I mean a serious historian, not a think-tank wanker, not an economist, not a journalist–who believes that the New Deal prolonged the Depression.

  2. dryfly

    Please, the New Deal was a drop in the bucket.
    The Great Depression was the greatest economic plop since the panic of 37-43(US economy didn’t recover to the 1850s fully though that would probably spark some debate lol). The 3.5 year contraction the greatest of all time. You don’t recover over night.

  3. pgl

    One small footnote to the 1933 to 1939 period. Real GDP actually fell by 3.4% in 1938. Not quite sure why but the good news was that real GDP growth for the next couple of years averaged around 8%.

  4. pgl

    George Will starts his WaPo oped by noting the 1937-38 recession. Memo to self – read the links before commenting.

  5. algernon

    Prof. Hamilton,
    Persuasive posts. Didn’t Roosevelt increase taxes & isn’t that–as an added burden on productive effort–a negative aspect of the New Deal?

  6. ken melvin

    Lower wages lead to consumption of over production?? Someones dogma just overran their thinking.

  7. EconLog

    Roosevelt’s Law of Tooth and Claw

    Hamilton and Arnold are making a lot of sense. On the negatives of the New Deal, though, it’s also worth…

  8. Lord

    I remain very skeptical government policies had much of any effect. For the most part, these policies were mere enshrining of tendencies inherent in markets. Employers don’t like to cut wages and make their workforce unhappy and employees don’t like their wages cut which makes them unhappy. This stickiness is not the result but the cause of policy. That companies would not invest more in markets already flooded with product even if they could do it cheaper than existing suppliers is also common behavior. Competing on price is not considered the path to profits.

  9. Valuethinker

    There was a massive destocking of inventories in 1938. Optimism simply dissipated (it was at exactly this time that Roosevelt’s political programme ran into real problems, as I recall).
    1939 and 1940 there is no mystery. The world was gearing up for total war and the US was a leading supplier of the raw and intermediate materials for war. The US itself began to rearm, and US industry accordingly benefited.

  10. Dave Iverson

    I’m curious, James, as to why you begin your assessment with 1929, and work on two periods thereafter. I note that John Kenneth Galbraith begins his The Great Crash 1929 with the twenties run-up as stage stetting for catastrophe. Galbraith includes chapter titles: “‘Vision of Boundless Hope and Optimism’, Something Should Be Done? [including discussion of why various players felt dis-empowered to do anything], In Goldman, Sachs We Trust, The Twilight of Illusion,” all to predate his chapter on “The Crash,” Then follows with “Aftermath I,” and “Aftermath II”.
    Galbraith, like Hyman Minsky and some others believed that the stage-setting era was as important or more important to understanding the Great Depression as was the aftermath. It seems to me that too many economists seem blind to the run-up.

  11. JDH

    I have many disagreements with Galbraith on this, Dave, but really to do justice to that issue, I’d need another post.

    The short answer might be, my paper I mentioned on the monetary contraction makes the case that you really don’t need those stories to explain a typical downturn, which is what you would have had if things had turned up in the second half of 1930. I’m sympathetic to the idea that there was a contribution of speculation to the problems, but I think this has been much overemphasized by writers such as Galbraith.

  12. spencer

    I keep being told the New Deal weakened the recovery phase of the depression.
    But I’m never told what the norm or standard is that it was weak relative to.
    Weak is a comparative term. For example, my baby girl is six foot tall. Your first reaction may be to say she is tall. But she is the shortest person in our family, so by that norm or standard she is short. So what was the 1930s recovery weak compared to?
    Look at the long depression of the 1890s. In that depression the unemployment rate took four years to rise from 4% to 18% and another 12 years to return to the 4% level it began it. But in the great depression the unemployment rate took four years to rise from 3% to 25% and another 12 years to return to the original 3%. So by this standard or norm the record of the recovery in the great depression was essentially identical to that in the long depression. So by this norm the New Deal did not weaken the recovery.
    So compare it to post war recessions. One of the amazing things about recessions is how symmetrical they are. On average, the number of quarters that payroll employment falls is almost identical it takes fro payroll employment to rise from the bottom to the prior peak.
    This table has the number of quarter’s payroll employment fall and rebounds to the prior peak in post war recessions-recoveries.
    Peak year Qs DOWN Qs UP
    Depending on how you calculate it the recovery is some 90% to 95% of the recession.
    But in the great depression employment fell for four years from 1929 to 1933 but
    Only took three years to regain the prior peak 1933-36. This is a 75% ratio compared to the 90% ratio for post war recession. So by this standard the recovery of 1933-36 was actually stronger then recoveries in the post war cycles. So this implies the New Deal made the recovery stronger.
    So I’ve presented two comparisons that very strongly imply the New Deal did not weaken the depression recovery. So all I’m asking is for those making the claim to show me the standard or norm they are applying so I can evaluate their claim.

  13. Stuart Staniford

    Speaking to the issue of agricultural issues particularly. I just finished reading The Worst Hard Time which is a pretty mind-blowing account of the dust bowl. I hadn’t quite realized the depth of the hardship at the epicenter of the dustbowl – literally nothing at all would grow at the worst of it, and a large part of the entire western plains were being lifted and blown around in giant dust storms on a near daily basis.

    His causal narrative for this essentially runs as follows: during the first world war with Russian grain excluded from the market wheat prices became very high, and remained high throughout the twenties. Wheat farmers were making larger amounts of money, and this drove very rapid homesteading of the Western plains, which were all placed under the plough for the first time.
    Rains were particularly good during those years, so there was a major speculative boom in wheat farming (he describes speculative non-farmers ploughing a section of prairie, throwing in a crop and then going back to the city till it was time to come out and harvest it – shades of day-traders or house flippers). By the time supply finally caught up with demand, it was 1929 and demand for wheat started to erode due to the financial hardship.
    Since many farmers had leases on machinery and mortgages on farms, the only way for them to cover payments on their equipment with reduced wheat prices was to plant more acreage, which many did – ploughing up the last of the prairies. Of course, while individually advantageous, as a collective behaviors, that only made the oversupply of wheat worse, and the price continued to collapse.
    Mother Nature conspired at this point to make the problem much worse by coming out with a series of hot dry years, which were not anomalous compared to the long-term historical record, but which were the worst since agriculture had been attempted in the Western plains. The combination of collapsing prices, farm foreclosures, and drought caused lots of acreage to be essentially abandoned with no cover. It then proceeded to blow, and destroy the productivity of much of the rest of the land (covered in dust-dunes and storms).
    So much of the New Deal agricultural programs were an attempt to bring this situation under control. There was a serious rational fear that most of the Western shortgrass prairie was about to become a large desert on a permanent basis. There was a serious oversupply of wheat relative to (much reduced demand), and much farming was occurring in places that were very marginal for agriculture. And there was a major feature of the problem that required collective action – I couldn’t fix my farm as long as the farms around me where all basically blowing dust, because the dust would blow onto me.
    Hence New Deal programs to create conservation easements and try to regrass particularly vulnerable areas, plant shelterbelts, create collective soil conservation districts, educate farmers to more sustainable practices etc, as well as help destitute farmers with loans, subsidence, relocation assistance, etc. These measures were taken in at atmosphere of desperation at what had been wrought by the excesses of the twenties and the unlucky coincidence of a depression and a drought at the same time.

  14. JDH

    Spencer, the standard of comparison in the Cole-Ohanian paper is simulation of the model with cartelization of the kind encouraged by the New Deal and without cartelization. What they calculate is the amount by which output is lower under the first simulation. In terms of asking whether this explains the data, what Cole and Ohanian point to is the fact that even by 1939, after 6 years of recovery, real GDP was still 25% below trend.

  15. Bradford DeLong

    Re: “I openly confess to believing that government policies that were explicitly designed to limit manufacturing, agricultural, and mining output may indeed have had the effect of limiting manufacturing, agricultural, and mining output.”
    And I would agree. But the NIRA rapidly became a dead letter, and the AAA and SDCA were small potatoes on the scale of the national economy. While the stimulative effects of abandoning the gold standard, implementing deposit insurance, and shifting fiscal policy from heavily contractionary to neutral lasted.

  16. David

    Robert Higgs ( makes an interestin case that the regime uncertainty created from the almost ad-hoc evolution of New Deal policies created a decline in investment that didn’t recover until the WWII. Above and beyond the shortcomings Brad De Long concedes to re NIRA, AAA, and SDC, this uncertainty could be another reason why the net effect of the New Deal policies was to prolong the Great Depression.

  17. Bob

    I agree with Brad on this. While some New Deal policies did hinder recovery from the Depression, it is silly to argue that the net economic effect of the New Deal was to hinder recovery from the Depression.

  18. Robert Schwartz

    For the past generation economists have focused on monetary and fiscal policy as the drivers of the GD. This is all well and good, but there is another area that needs more attention — international trade.
    One of the defining characteristics of the GD was the collapse of international trade. The collapse of the gold standard and the ensuing monetary chaos are a large part of the story, but other factors need to be considered, including the Smoot-Hawley tariff, and the wave of Fascism that swept Europe.
    One of the clues that these are important factors is the attention that was paid to them in the Post WWII restructuring. The creation of the international bodies of trade and finance at that time, such as IMF, BIS and GATT, was one of the great success stories (just as the UN was a failure).
    “One would be very hard-pressed to find a serious professional historian…–who believes that the New Deal prolonged the Depression.”
    First, find me one who isn’t a shill for the Democrat party.

  19. TruePress

    “The New Deal and the Great Depression”

    There has been, over the past week or so, a discussion among economics bloggers about whether Roosevelt’s New Deal prolonged the Great Depression. Prof. James Hamilton of Econbrowser posts on the topic with a nice overview and links to the oth…

  20. spencer

    James — I read the Cole paper.
    Given his assumptions his claim that the 1930s recession was weak is a good case. But there are some very heroic assumptions in the alternative scenario he constructs.
    But my case that payroll employment growth was stronger then in any post war recovery is just as valid.
    Moreover, my comparison to developments in the long depression are also just as valid.

  21. Howie Copywriter

    On Dec 12th, good old Ambrose Evans-Pritchard of the Daily Telegraph commented on the expectation of the Pound Sterling to plummet. This becomes relevant where the role of the fall of the pound triggering a further catastrophic fall of the dollar is making the rounds this week.
    Meanwhile, the oil price is falling. This may be a welcoming sight for motorists around the world, but this will go into the mix, or shall we say the descent into the maelstrom. Hedge funds could take a big beating. They have been betting on big macho man Cheney to hit Iran and start the rise of oil toward $200 a barrel. But, it has been delayed, though not entirely called off. Here’s a little bit of advice to my hedge fund friends. If the Democratic congress nixes the “surge” in Iraq, then oil will continue down. And alot of hedge fund people are going to be hung out to dry, like what happened to the Amaranth Hedge Fund people up in Greenwich Ct. Or the Amaranth people out in the Cayman Islands. You know, the usual arrangement, the onshore and the offshore branches of twin companies. Anyway a real crash is definitely in the mix for hedge fund people nation. More at real crash

  22. JDH

    Those are useful points to remember about the Dust Bowl, Stuart. Gauti Eggertsson also calls my attention to a new paper of his suggesting some other reasons why, in the extreme macro conditions of 1933, an aggregate increase in monopoly power might have been beneficial.

  23. Patrick R. Sullivan

    With the Eggertsson paper we seem to have gone from ‘no normal person’ says the New Deal prolonged the Depression, to all normal people–including Keynes talking to FDR–say so. But, they might be wrong.

  24. John Thacker

    But in the great depression employment fell for four years from 1929 to 1933 but only took three years to regain the prior peak 1933-36. This is a 75% ratio compared to the 90% ratio for post war recession. So by this standard the recovery of 1933-36 was actually stronger then recoveries in the post war cycles. So this implies the New Deal made the recovery stronger.
    Spencer, you can’t say “It’s over” when the economy returns to its peak in 1937, if the economy then collapses mightily in 1938-1939 and takes longer to return to the peak again, which it did. The “long depression” from about 1873-1890 also is commonly broken up into two different recessions in the US; GDP returned to its peak in the US in the middle of the 1880s, but not the unemployment rate.
    spencer is committing data cherrypicking. A reasonable person can argue over whether to combine recessions so that you have two or one in the 1873-1890 period, and to have 1937-1939 be an entirely separate second recession from the Great Depression or not. However, treating one one way and one the other is suspicious to me.

  25. John Thacker

    I divide the U.S. Great Depression (1929-1939) into three episodes: (1) the initial downturn (1929-30), catastrophic free fall (1931-32), and slow recovery (1933-39).
    I would add a fourth episode, the recession from the end of 1937 to 1939. The war then follows.

  26. DickF

    I have been sadly amused at the new method of reasoning that has come unquestioned into our intellectual circles. The argument goes something like this. We have a system that is the best system, but because the government has the ability to abuse the best system we must choose a worse system. This is the argument that Bernanke and JDH use concerning the gold standard and the GD.
    During the 1700-1800s England followed a gold standard that made currency as good as gold. This period was one of the most prosperous in the history of the world. Now we enter the 20th Century and governments refuse to follow the monetary system that worked because they know they can confiscate capital from the people without direct taxation if they manipulate the monetary standard.
    But the truth of this argument is that it negates any system used in government. No system can be argued as valid because the government can always abuse the system. The result is the end to rational debate. We must accept what is as the only solution.
    Sadly the best monetary system is cannot be debated because the arguments of academia, while acknowledging the best system, reject it without debate. The new logic has ended debate.

  27. spencer

    Jim –I calculated a 3.5% long term trend growth rate of real gdp from 1900 to 2000. Using this trend, real gdp was 35% below trend in 1933 and fell to 20% below trend in 1936. The monetary induced recession in 1937-38 kept the gap from closing and it was still 20% below trend in 1939. A little different but not significant from the number you quoted. But the question is not how big the gdp gap was in 1939, rather it is was the reduction in the gap from 35% in 1933 to 20% in 1936 a weak or a strong recovery. Going from a -35% to a -20% gdp gap from 1933 to 1936 appears to be a stong recovery by historic norms.
    If the recession of 1937-38 was due to tight monetary policy rather than fiscal policy or the new deal — what I think but a debatable issue —
    I do not think I am cherry picking by seeing how the economy recovered from the deperession bottom compared to other economic cycles.
    The unusual thing about the depression was how severe the 1929-33 collapse was. So comparing how the economy came off the bottom in one cycle to how it came off in other cycles is not cherry-picking. It is a valid approach that is used all the time today in comparing one economic cycle to others.
    Why is my comparison of one set of historic data to another set of historic data cherry picking but using a completely hypothetical set of data
    not cherry picking?

  28. zinc

    1929 -1939 were the downside of a great speculative bubble that began in the early 20’s. To me, viewing the GD from 29 – 39 is like viewing the NASDAQ bubble from 2001 – 2003, rather than 1998 -2003.
    How does the above analysis jive with the recent Japanese experience ? The Japanese experience, like the great depression, seems to be typical of collapsing bubbles. The collapse of bubble induced over production and speculative asset price inflation seems to result in long periods of psychological depression. Like the psychological reprecussions experienced after 1929, Japan’s zero interest rate policy has taken an inordinate time to stimulate domestic spending, contributing to a worldwide speculative bubble via the carry trade.
    I think your analysis completely ignores more traditional measures to arrive at a rather offbeat conclusion.

  29. Zephyr

    The Japanese experience was a rerun of the failed restrictive policies that caused the Great Depression.
    In the late 1980s the real estate bubble contributed to putting the US and Japanese markets in a precarious position. In the US we had builder bankruptcies, S&L failures, insolvency of the S&L Insurance corp., and the creation of the Resolution Trust Corp. to dispose of the foreclosed bank assets in an orderly manner.
    To cool things off Japan repeated our restrictive policy mistakes that sparked the Great Depression. They had a similarly bad outcome — no surprise there. We were smarter than that and avoided the severe pain that would have come again in the 1990s if we had also followed that restrictive policy path again.
    There are other factors as well, but the difference in policy paths is a major factor in the outcomes of the 1990s.

  30. Howie Copywriter

    Bush is going crazy. He is shooting down any way out of the mess. Meanwhile, the US and the dollar are ready to collapse. The fight is to persuade the US Congress to adopt a real capital budget, and get out of the Thirty Years War in Iraq, Iran, etc.
    The idea of an expanded war is the option of “stabilization” of the financial bubble. Look at the US bombing of Somalia, this is part of it. Force panic stricken people to do anything for a so-called “war economy”.
    Meanwhile, the widespread fraud of not reporting the true state of the Housing Market, is seen in the NY Times, Jan. 7th. Also see Moody’s . The US Comerce Dept figures reported for new home sales both 1. significantly overstate the level of new home sales, and 2. understate the number of new homes listed for sale- inventory of homes for sale- due to Commerce’s not taking account of people cancelling their contracts to buy new homes.
    real crash

  31. DickF

    I hope you have rushed out and bought your copy of the Koran because if your attitude wins you will need it.

Comments are closed.