The Midas Paradox So Far

That is Scott Sumner’s magnum opus on the Great Depression. I am part way through it. The importance of the book cannot be overstated. Here, I want to mention two problems I am having.

1. The structure of the book. It is difficult to get through. He constantly interrupts his own interpretive narrative to discuss other economists’ narratives. I would have rather seen the discussions of other narratives cordoned off into appendices.

2. Sumner wants to rely on stock market movements as indicators of the effect of monetary policy on the economy. I am never a fan of trying to interpret the stock market, and this time period seems particularly problematic. Robert Shiller’s famous argument against the efficient markets hypothesis is the “excess volatility” that he found in stock prices. The period that Sumner is describing has volatility that goes far beyond even what caused Shiller to reject market efficiency. Reading Sumner’s narrative, there seem to be a lot of days where the market goes up or down by 4 percent, 5 percent, or more. In today’s terms, that would mean the Dow often gaining or losing 700 to 850 points in a day. I am inclined to dismiss 90 percent of these movements as irrational (I would do the same for short-term movements in the market now). I understand that Sumner is looking for indications that the economy is responding to monetary policy, as opposed to some other factor, but I find it more persuasive when he cites medium-term trends in commodity prices than when he cites short-term stock price behavior.

7 thoughts on “The Midas Paradox So Far

  1. I wonder if someone will explain in a detailed and convincing way why the book is so important. In the only comment I have ever posted on Sumner’s blog I asked, in effect, if he discussed this in the book or elsewhere. Although he often responds to comments, he did not respond to this.

    • He is pretty clear about it in the introduction to the book.

      And I’ve left a comment about this before, but I’ll rehash it here. But before I get to it, let’s step back to see the big picture.

      Any proposed answer to the question of “What really caused the Great Depression” is inevitably pregnant with policy implications, and so is unavoidably incredibly politically charged. That’s one of the reasons so many top economists are so keen on studying it. Big names like Keynes, Hayek, Friedman, and Bernanke, just to name a few.

      In order to answer the question you have to present, explicitly or implicitly, some kind of model of “How The Macroeconomy Really Works”. That model then inescapably implies some kind of governmental strategy for “What Must Be Done” to maximize production and stabilize and optimize the growth path, and most especially, how to prevent – or otherwise minimize the pain of – any downturn or disturbance.

      But there is a huge amount at stake in whatever that recommended strategy turns out to be, because it may lend support to, or otherwise be inconsistent with, one’s personal preferences for a political agenda. Just like in US politics, there tend to be two big factions, either insistent or skeptical of the benefit of government management of the marketplace and of major state interventions.

      I don’t think it’s very controversial to say that these huge political stakes have a very corrupting influence on which model of the Macroeconomy (and thus, the compatible explanation for The Great Depression), one favors. The only controversial part is which faction is being pure and innocent and scientific, and which is being vile and corrupt. It’s also possible to think almost everyone is corrupt or otherwise wrong, but then your team is tiny, and apparently social networking is critically important in the upper echelons of the field and in terms of having your ideas taken seriously.

      Ok. As you might now expect, Scott Sumner’s explanation of the Great Depression is consistent with his Market Monetarist model of “How The Macroeconomy Really Works” and thus also lends support to his proposal for NGDP-futures level targeting as the optimal government economy-stabilizing management regime.

      But you can’t propose such a model and implicit suggestion for a management regime without also passing the fundamental test of credibility and plausibility in the profession, which is to test it against the universal yard-stick of how well it accounts for The Great Depression.

      And that’s what The Midas Paradox does for the Market Monetarists and the NGDPLT proposal. Sumner wanted it to be “The Midas Curse”, just like the myth, and indeed that would have been a better and more apt title. Midas for gold, Curse for being harmed by an excess of the thing you wanted (in this case, deflation caused by hoarding). But it was hard enough to find a publisher for this book, and they wanted paradox and they got it, and anyway it’s close enough.

      Central to the Market Monetarist perspective on things is the Efficient Market Hypothesis (EMH) and the role of expectations in terms of the trends in total income (NGDP) and inflation, and the fact that the best guess as to these expectations is embedded in market prices.

      Concordantly, sudden, major, and durable shifts in those market prices in apparent response to new, surprising, widely publicized information about international events, market conditions, or new government policies, embed information about the expectations regarding the likely impact of those contingencies. Yes, there is a lot of ordinary market volatility, but with a large enough sample you should be able to extract a signal from the noise if there is one there.

      As a proxy for ‘new surprising information’, Sumner used economically-relevant articles reported in the New York Times, which, given the dominant way and speed in which information was disseminated to all market participants at the time, is a reasonable-enough source.

      Sumner’s thesis is that there is indeed a good signal there, in which real news is highly consistent with near-immediate shifts in market prices, which embed that information about expectations, which are, in turn, well-correlated with future production. The most important market price was gold, because of the various different gold monetary standards at play at different places and time during that historical period. Expectations about future Macroeconomic trends and prices made it individually rational for individuals and governments to alternately hoard or divest themselves of gold, and, Sumner demonstrates, these swings were critically important drivers of Macroeconomic conditions.

      He goes on to argue that this does a much better job explaining the dramatic twists and turns of The Great Depression than the rival macroeconomic theories. (For a great narrative account of what these twists and turns, booms and busts, felt like to an ordinary citizen, I highly recommend Benjamin Roth’s “The Great Depression: A Diary”)

      And since this rival Macroeconomic theory does such a better job meeting the ‘gold standard’ of explaining The Great Depression, then the model deserves to be upgraded in status and credibility (with the other ones, dominant before the GFC, deserving severe downgrades). And consequentially, the proposed Market Monetarist regime of NDGPLT based on that model should be embraced by central bankers in developed countries in favor of the Taylor-Rule / inflation-targeting model, which arguable didn’t ‘work’ and still isn’t ‘working’ and so ought to be abandoned.

      Obviously there is a lot of controversy about those assessments, but those controversies cannot be easily resolved because of the weak epistemic status of Macroeconomic theories, in which controlled experiments to test theories are practically impossible. The best one can do is create an interpretive framework and observe it is seems to do better or worse than others in producing estimate that are consistent with major past crises as well as current events.

      Sumner argues that his interpretive framework is superior, and this book is the “consistency with the most significant past event” pillar of that argument.

  2. Thanks Arnold. You are right that the structure of the book is not very convenient for the typical reader. It’s not how I would have written it today. The book was written many years ago, and aimed at a “economics journal” audience. That’s why there is all the discussion of other theories.

    A few comments on stock prices. The standard deviation of daily movements in stock prices was about 2% in the 1930s, vs. about 0.8% in the postwar stock market. So yes, it was much more volatile. As far as the question of how meaningful the movements are, I strongly disagree with Shiller’s view. I believe significant movements almost always occur for a reason. Maybe the reasons are in some sense incorrect, at least ex post. Many large market movements during the 1930s are usually linked to significant news stories, with the exception of 1929 and late 1937. On those two occasions, I think the market was falling on signs of collapsing AD, but can’t prove it. In my view the stock market data is most useful when correlated with other data, like German bond yields, forex movements, etc., not when examined on a standalone basis. It’s also important to recognize that the “significance” of a stock market /policy correlation goes up one or two orders of magnitude when there is a real time connection during the trading day.

    We saw huge stock market declines in October 2008, does anyone seriously believe those were random, and not related to a growing perception of a very severe recession? Or as an another example, we often see huge swings right after a Fed announcement at 2:15pm. Does Shiller believe that’s a coincidence?

    Some of the market movements that occurred during the 1930s, right after policy announcements were among the largest in history. That’s a pretty amazing coincidence. This the two day rally after Glass-Steagall (1932 monetary stimulus bill, not the more famous 1933 bill) was the largest two day rally in history. That’s an extreme case, but there are lots more that are only slightly less extreme–the biggest one day fall in 1930 was the day after Hoover announced he’d sign Smoot-Hawley.

  3. When I commented on his blog that the very day of the stock bottom of the great recession was the announcement of the relaxation of fasb 107 Professor Sumner rather tersely replied that he thought the stock market is a random walk.

  4. Thank you for your responses.

    People may be underestimating the naivete of my question. I was not asking about the methodology used to arrive at the conclusion, but about the significance of the conclusion. Crudely: Why should a nonspecialist, with no particular interest in economic history but just a citizen’s interest in politics, care about it? What specific propositions, vital to non-Sumnerian macro theories, would be refuted (or something like that) if his analysis of the GD is true? I know that Keynesianism was developed mainly to explain and deal with the GD, but I don’t know how important the ability to do that is to the current intellectual justification for Keynesianism. Handle describes that ability as the universal yardstick for macro theories. Does this mean that Keynesians (and other non-Sumnerians) would agree that if this book is correct, their theories are fundamentally wrong or have lost most of their foundation? That might be true for all I know, but things are rarely so clear-cut.

    I will, as recommended, look at the book’s introduction.

    • I don’t think the ability to explain the Great Depression is the “yardstick” by which macro theories are judged. However … if a theory can’t, it won’t be taken seriously, and it’s policy recommendations won’t be taken seriously.

      There are a number of theories that seem to at least partly explain the GD. In this book, Sumner is trying to show that his theory belongs in that charmed circle.

Comments are closed.