The Federal Reserve, Washington, DC
There has been a fair amount of concern expressed by some prominent analysts that the Federal Reserve has made mistakes recently that have the potential to push the economy back into recession. If this were to happen, it would be very reminiscent of the mistakes made by the Fed and FDR's Treasury Department in the mid-1930s, which caused the terrible second crisis of the Great Depression. I last wrote about this issue in March, so it seems prudent to re-examine the question now that the Fed has apparently reversed course on its planned series of interest rate increases. The comparison of the present day economy to that of the second phase of the Great Depression dates back at least to late 2014, when economist and Nobel Laureate Robert Shiller warned about the parallels between the state of the global economy then, and in 1937.
A counter-argument against this analogy was made by Michael Arone at State Street Research (cited by Mike Bird of the Business Insider blog), who pointed out that our modern economic weakness does not yet compare to that of the 1930s, and we also have not really faced years of deep deflation either. These are good points to consider. However, in March of 2015 famed hedge fund manager Ray Dalio wrote about the problems that could occur if we enter a new recession with monetary policy basically caught at the zero bound, setting up a liquidity trap like that seen in the 1930s (cf. Mark Melin, 2015). Indeed, in the event of a slowdown, there are a whole range of problems to consider beyond the idea of the liquidity trap.
Dalio was also concerned about several additional things: 1) the heavy burden of debt service in the present economy, which limits spending; 2) the parallel use of quantitative easing (QE) in the 1930s and recently, with huge stock rallies but ultimately (in the 1930s at least) a second economic collapse; and 3) huge over-valuations in stocks that left investors unwilling to hold them in a downturn, resulting in a nearly 50% drop in the Dow Industrials (Chart 1) in the 1937 crash, and the potential for the same now. In 1937-1938, after a series of mistakes by the government, what followed was a massive drop in industrial production (by 33%; cf. Frank G. Steindl) and prices (by 11%; chart 2), and a doubling of unemployment to about 19% (Chart 3). GDP fell by around 3.5% (Chart 4) by 1938, and commodity prices collapsed again (Chart 5) during the renewed, or second phase of the Depression.
Chart 1: Dow Industrials in 1937-1938 Crash
Chart 2: Industrial Production and Prices Plummeted in 1937-1938
Source: Frank G. Steindl, OSU; eh.net encyclopedia
Chart 3: Resurgence in Unemployment 1937-1938
Chart 4: GDP Growth Was Negative Again in 1938
Source: US Bureau of Economic Analysis
Chart 5: Commodity Prices Collapsed in 1937-1938
As already mentioned, I have discussed the causes of this second crisis of the Great Depression elsewhere, but briefly, it was triggered by several events: 1) the failure of monetary policy to get traction at the zero bound, followed by the Fed's premature increase (Chart 6) in the discount rate; 2) the Fed's decision to double reserve requirements for banks, which caused them to cut back on loans (Chart 7) to the private sector; 3) FDR's decision to balance the budget in support of the return to the gold standard, including the use of higher taxes; 4) the first (small) withdrawals of money from paychecks for the new Social Security system; 5) rapidly rising wages (up 11% in 1937) as a result of the 1935 Wagner (Labor Relations) Act, which cut corporate profits; 6) the reversal of capital flows to commodities-producing countries; and 7) the FDR Treasury's decision to sterilize gold inflows, which resulted in a 10% cut in the money supply (Chart 8). Of these, only items 1, 6, and 7 have been deemed critical by recent researchers in causing the disaster, although all of these factors seem to have had at least some negative impact. Items 1-6 all have parallels in the present economy, based on my previous research, and also that reported by analyst Dorothea Lange in her column for The Fiscal Times (2016).
Chart 6: The Fed Prematurely Attempted to Normalize Rates in 1936-1937
Source: US Federal Reserve
Chart 7: Fed Actions Caused Loan Activity to Drop in 1937-1938
Source: Frank G. Steindl, OSU; eh.net encyclopedia
Chart 8: Treasury Gold Sterilization Cut Money Supply 1937-1938
Analysts at Bank of America Merrill Lynch, and analyst Mike Bird of the Business Insider blog, revived the discussion about potential Fed errors when they also wrote later in 2015 about the 1937 analogy and the risk of a calamity occurring again as a consequence of the Fed raising rates prematurely. But then in December 2015 the Fed actually executed the first of what was then expected to be a long series of rate increases. And just last week the ghosts of 1937 were again mentioned, this time by Sue Chang at MarketWatch.com, who wrote about research done by a team at Morgan Stanley under economist Chetan Ahya. Once again, the fear of some researchers is that policymakers are repeating the types of mistakes that prolonged the Great Depression. Ahya pointed out that similarities in economic conditions between 1937 and now include an apparent major peak in the debt cycle (Chart 9), cash hoarding, deleveraging, and deflationary pressures.
Chart 9: Parallel Peaks in the Private Debt Cycle, 1933 and 2009
Source: Steve Keen; debtdeflation.com/blogs
I would note two other similarities between then and now: 1) an Administration in each case that is hostile to business, willing to pick corporate winners and losers, and fond of massive, even debilitating increases in regulations; and 2) the gap or potential gap between productivity and wages (Chart 10; cf. Gerard Jackson, 2009; https://seekingalpha.com), which in theory should have caused higher and longer-lasting unemployment, and actually did in the 1930s. I have mentioned the dilemma of recently falling productivity growth elsewhere, but the data in Chart 10 suggest that this is a subject worth more analysis because the gap in the 1930s destroyed jobs, and we see weakening job growth, high levels of permanent unemployment, high part-time employment, and mildly rising wages right now. For now, it is reasonable to blame the FDR Administration's penchant for massive wage increases (decoupled from productivity), and workers' demands for the same in spite of steeply falling prices, on the common error known as the money illusion (cf. Irving Fisher, 1928; The Money Illusion).
Chart 10: Productivity-Wage Gap in Great Depression
Source: Gerard Jackson; seekingalpha.com
Having explored the parallels between the second phase of the Great Depression and now, and in light of the Fed's decision last week to back away from its planned rate increase, it seems like a useful exercise to now re-examine whether the analogy of the "1937 Mistake" still applies, or the danger has now been averted by the Fed recognizing (belatedly) the risk inherent in the present economic circumstances. Could it be that by halting the planned series of rate increases, the Fed has managed to extricate itself from error and all will be well? As much as I would like that to be the case, I have my doubts. For one thing, there were a number of interconnected errors in the Great Depression's second dip, and we are repeating in some manner or form almost all of them. In fact, if you accept that the eurozone is the monetary equivalent of the Depression-era gold standard, as some famous economists now believe (presented in another paper last March), then all of the significant causes of the "1937 Mistake" are actually present now.
Of course, it should also be readily apparent to the reader that there are some substantial differences between the nature and severity of the Great Depression, and our present weak economy. So to that extent, the 1937 analogy does not yet align very perfectly between present and past. However, we should also perhaps remember Leo Tolstoy's famous principle: "Happy families are all alike; every unhappy family is unhappy in its own way." Thus, just because we do not yet suffer anything like as much as people did in the Great Depression, does not mean that we cannot contrive to make it happen in the future. If this sounds pessimistic, it is nevertheless true: in economics we create all of our own problems, including depressions. If you answer that a modern economy could not conceivably degrade so thoroughly as to collapse into a depression (defined as two years of negative GDP growth with high unemployment), then I give you Russia (1991-1996), Japan (1998-1999), Greece (2011-2013), and Brazil (2014-2016) as likely examples. We will only avoid making fatal errors with the economy if our policymakers have learned from the past, and by that I mean, if they've learned accurately the lessons of the past.
There are plenty of reasons to fear that policymakers are indeed repeating a pattern of errors we've seen before, and that the dogma of modern economics has in some ways blinded many to the proper lessons to be drawn from the Great Depression and other historical events. For example, economist Brad DeLong (2013) wrote an extensive overview of policy errors in the Depression and the Great Recession. While he stipulates that the Great Recession was probably much less severe than the Great Depression because of the many correct things that the Bernanke Fed did in 2008-2009, he is concerned about the huge and apparently open-ended gap between potential GDP and current GDP growth (Chart 11). The gap has narrowed a bit since 2013, but it implies a structural problem that can't be dealt with using monetary policy.
Chart 11: The GDP Gap Has Not Been Closed
Indeed, one of the major mistakes made by the Bernanke and Yellen Fed has been the abandonment of fiscal policy in favor of monetary policy acting alone, as has been discussed in detail by well-known economists like Allan H. Meltzer (2013), Josh Bivens (2014), and Michael Lebowitz. These authors point out that the existence of the liquidity trap should have led the Fed to recommend strong fiscal intervention years ago, but it did not because of the intellectual blinders it wears in the form of accepted monetary theory. The fact that it hasn't worked has not dawned on them for years on end, in spite of the monumental pile of excess bank reserves that they built up with QE, which is prima facie evidence of failure. More evidence of failure comes from the enormous pile of cash held by corporations. Perhaps with the recent rate decision reversal they have begun finally to see the problem.
The over-emphasis on monetary policy using experimental measures like QE has not only failed to stimulate the economy, it has caused stupendous mal-investment on things like stock buybacks and debt-funded mergers. Perhaps the most damaging mistake recently, though, has been the failure to hold to a monetary rule. The Fed is guided by some ephemeral version of the long-disproven Phillips Curve (which only operates in the short term); this does not exactly fill one with confidence. But then, to add insult to injury, the Fed has long operated on an ad hoc "data-dependent" basis, making it rely from meeting to meeting on extremely volatile and unreliable quarterly data. The excuse for this approach is that we are at the zero bound, but this has greatly increased uncertainty, appears arbitrary, and has achieved little. In my shop we call it "Cowboy Economics."
Interestingly, the Fed's best record for monetary stability comes from the only two periods where a well-defined rule was used (i.e., the gold standard from 1923 to 1928, and the Taylor Rule from 1985 to 2003, according to Meltzer. During the recent crisis (late 2008-2009) and its aftermath then, policy was entirely ad hoc (Chart 12), and by implication (compared to the Taylor Rule), too tight. Here it is important to note that fiscal policy was required once the Taylor Rule could no longer be followed below the zero bound. Instead we had wave after wave of QE, when only the first was required by the crisis. According to Dorothea Lange, this happened because policymakers focused too much on the inputs to monetary policy (QE and interest rates) and not enough on the outputs (inflation and nominal GDP growth). It would appear that instead of acting on a "data-dependent" basis, the Fed should have recognized that the game was up for monetary policy, and they should have pushed the Administration to act. This has still not been done, although both presidential candidates are talking about the need for infrastructure spending using the federal deficit as a funding source. Meltzer suggests Congress should require that a rule such as the Taylor Rule be used (except in liquidity squeezes) in future.
Chart 12: Ad Hoc Data Dependency vs. the Taylor Rule
Surprisingly, according to Meltzer, the Fed also has no explicit policy (and never has) on its role as the Lender of Last Resort -- its primary function under the Federal Reserve Act of 1913. In consequence, the modern Fed has completely misunderstood its responsibilities under Bagehot's Rule, just as the Depression-era Fed did. Bagehot's Rule states that as Lender of Last Resort, the central bank should lend freely against good collateral at a penalty rate of interest. In the Great Depression the Fed failed to lend freely to local and regional banks, so a panic eventually resulted (for this and other reasons). The modern Fed lent freely but not at penalty rates and not in exchange for good collateral. The result was the survival of zombie, Too-Big-To-Fail banks that to this day can still threaten the entire system.
In conclusion, the parallels between the "1937 Mistake" and the present are relatively too close for comfort, in spite of the (at first glance) weak analogy based on the depth of the respective recessions. The reasons for this are straightforward: the present weakness is structural and is capable of degenerating into another very severe recession or even a depression if another global economic collapse occurs while we are stuck at the zero bound.
The pages of many journals and blogs are filled with warnings about how unusual and scary our present circumstances are. So we start from a far milder situation than in 1937, but we are headed we know not where, and not only has monetary policy failed, but there is no fiscal policy to offset it. We are unlikely to drift in a benign state for very long without some sort of policy to boost productivity and NGDP. Thus, although the differences are clear between 1937 and now, the failure of policy is similar and could turn out very badly. The only ray of hope is that the Fed avoided the short term blunder of raising rates this time around, but they unfortunately remain "data dependent."
Trades of interest in this environment (if you are bearish after reading this) would include the purchase of long Treasuries (Wasatch-Hoisington US Treasury Fund [WHOSX], Vanguard Extended Duration Treasury ETF [EDV], Vanguard Long Term Government Bond ETF [VGLT], I-Shares 20+yr. Treasury Bond ETF [TLT], and I-Shares 7-10 yr. Treasury Bond ETF [IEF]), defensive sectors (Powershares S & P 500 Low Volatility ETF [SPLV], and I-Shares Edge Minimum Volatility ETF [USMV]), liquid alternatives (Otter Creek Prof. Mngd. Long/Short Fund [OTCRX], AQR Long/Short Equity Fund [QLENX], and AQR equity Market Neutral Fund [QMNIX]), and sophisticated hedge-like strategies like Nuveen S&P 500 Buy-Write CEF (NYSE:BXMX).
Disclosure: I am/we are long VGLT, TLT, OTCRX, QLENX, QMNIX, BXMX.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: This article is intended to provide information to interested parties. As I have no knowledge of individual investor circumstances, goals, and/or portfolio concentration or diversification, readers are expected to complete their own due diligence before purchasing any stocks or other securities mentioned or recommended.