The US Armageddon Scenario and 1937

The title of this post is a bit alarmist. But what I intend to write is more about reasonable worst case economic scenarios that the Fed is looking to avoid. While a 1937-style downdraft is something to consider down the line, in the medium-term the question revolves more around the Fed tightening into weakness that is exacerbated by turmoil abroad. I am not unduly concerned about another 2008-style collapse. However, I do want to point out where I think the limited downside is for 2015.

Framing

I am in Mexico and have a bit of writing and thinking time. So I decided to put this together based on the Ray Dalio scenario from the 11 Mar Bridgewater Daily Observations that many people are talking about. Dalio lays out three basic beliefs driving his concern regarding a 1937-style meltdown:

1. The same things occur again and again due to universal cause-effect relationships
2. A long-term debt cycle, while not well-understood, is important to economic and market outcomes
3. Being at the end of such a cycle, monetary policy is less effective.

The upshot of this analysis is that if the Fed does not err on the side of caution, a very serious relapse in market and economic contraction could occur. Dalio gives the 1937 example as the precedent to reflect upon. I would also suggest 1997 Japan as another similar precedent that might be more relevant for today, as well as 1994 for the US. 1937 is the worst outcome of the lot, with 1994 the best outcome.

The scenario plays out like this, according to Dalio

1. Debt limits reach a bubble top, causing economy and markets to peak (1929 and 2007)
2. Interest rates hit zero amid depression (1931 and 2008)
3. Money printing starts and a “beautiful deleveraging” occurs (1933 and 2009)
4. Shares and risk assets rally (1933-1936 and 2009-2014)
5. The central bank tightens, with a self-reinforcing downturn (1937 and 2015?)

I want to use this framework as a starting point for thinking about reasonable worst case scenarios going forward. I have three planks that differ from Dalio that I want to discuss upfront. First, I believe fiscal policy is a key consideration here that is not addressed and the 1997 Japanese scenario shows us this. In 1937, there was also a fiscal tightening in addition to the monetary tightening. Second, the common belief amongst monetarist and New Keynesian-type policy makers running the Fed is that Japan’s problem was how late it moved to a zero rate policy and quantitative easing. The aggressiveness of Fed policy ensures a more benign outcome for the United States. I don’t believe this. Rather, I believe that the debt supercycle requires still more deleveraging, something that aggressive policy has inhibited.Third, the Japanese scenario was more demographically challenged than 1937 and I believe demographics are something that will affect outcomes today as well.

Dalio’s Basic Beliefs

The three basic beliefs driving the scenario Dalio presents make sense. Regarding private debts, the cause-effect relationship to consider here is that as debt and leverage build up, more and more debtors within a given economy have less room for maneuver if the economy turns down. This creates financial fragility which leads to crisis and is a major reason we have seen so many crises in the past generation. If you go back to the 1980s, for example, and look at the macro picture in the US, one could make a reasonable case for the 1990-91 recession and the jobless recovery as resulting from the financial fragility of the S&L crisis and the leverage built up during the Predator’s Ball era.

Judging when the long debt supercycle is an at an end is the key aspect here. And here, what I am saying is that aggregate debt levels build up across cycles until they become unsustainable and then a secular deleveraging begins. In 2002, one might have thought that we at the end of the supercycle as yields dropped to 1%. But, in hindsight we were not simply because monetary policy has to reach its limits before the supercycle ends. What I am saying is that monetary policy has to be proven to have become ineffective to stop the addition of leverage or policy makers will intervene and economic agents will maintain leverage to boost returns.

We are now at zero or even negative everywhere. This is the effective limit of monetary policy. Credit cannot be scaled up further from here unless the real economy is cooperative. And when the economy turns down subsequently, there will be no monetary policy space to prevent a hard deleveraging. This is what we saw in 1937 and in Japan in 1997 and why 2015 differs from 1994, when interest rates were much higher.

The Supercycle End Scenario in 1997 Japan and in 1937 US

So the scenario that Dalio lays out in 5 points works equally well for 1997 Japan as it does for the US in 1937 or 2015. In each case, private debt skyrocketed and an economic crisis ensued when monetary policy was overwhelmed by financial fragility due to the number of debtors with overextended balance sheets. The differences I see here all suggest that the situation today has been more benign than in the two previous scenarios.

First, in the 1929-33 period, the central bank was unable to take dislocated asset markets onto its balance sheet for the simple fact that the failing assets were not in marketable securities. And so the deleveraging was much greater as a result. Second, the bank recapitalization in the United States was much more robust than anything we saw in Japan. Zombie banks that staggered away from the 1989 crash in Japan were vulnerable by 1997 because of dodgy balance sheets and poor capital adequacy. In this sense, the United States financial system appears better prepared for a recession than either 1937 America or 1997 Japan.

In the US, fiscal policy was a major input variable that created the contraction. Federal government outlays fell 10% in 1937 and another 10% in 1938. That is a massive fiscal contraction, which I believe was more important than the monetary tightening. Therefore, I am certainly less concerned about a 1937 relapse into economic depression as a reasonable worst case scenario due to monetary tightening. Let’s remember that the U-3 level of unemployment in the United States today is 5.5%. We would need to see a large 2008-style financial crisis, a terrific deleveraging and an economic bust to create a 1937-style economic relapse. I don’t see this as a reasonable worst case scenario at this time.

But what about a 1997 Japanese scenario? I see this as a better match for a number of reasons. But even here, we saw a number of policy mistakes come together at the same time. In Japan, the consumption tax was increased in April 1997, sucking money out of the private sector before a full deleveraging was complete. So, yet again, the fiscal tightening was a major factor in the depressionary relapse. A banking crisis began very rapidly then in November of that year, continuing until the spring of 1999. Japanese banks were undercapitalized and had to pay a “Japan premium” when borrowing in eurodollars in the London offshore markets. Credit had to contract in this scenario. The discount rate in Japan was 0.50% at this time, giving the Japanese another 50 basis points before hitting the zero lower bound. Zero rates were not introduced until the spring of 1999. The BOJ didn’t embrace quantitative easing until 2001 and this was only using government bonds, not private sector distressed assets as the Fed used in QE1.

When rates fell to zero in Japan, in the spring of 1999, IT shares boomed, spilling over into the wider Japanese stock market and asset-led recovery took hold at the time the manias in the US and Europe in TMT was ongoing. However, in 2000, just as the bubble in TMT was popping, the BOJ began to tighten, raising the call rate from 0 to 0.25% in August. Recession followed. This is when deflation set in in Japan.

The Supercycle End Today

We are already through the first four points in Dalio’s relapse scenario and the question becomes what happens in step five. The reasonable worst case scenario I see goes like this:

1. The United States Federal Reserve raises interest rates due to the U-3 unemployment level falling to its 5.0 – 5.2% threshold/trigger range because inflation has not declined enough to prevent normalization.
2. The US dollar rises in value due to the market’s reassessment of the prospects for further tightening in the US.
3. The strong US dollar causes a downturn in oil prices that creates another wave of problems in the shale sector, lowering capital expenditure and creating job losses, knocking tenths of a percent off of GDP growth. Contagion spills over into the high yield market.
4. At the same time, the strong dollar becomes toxic for emerging market debtors with significant US dollar liabilities in countries like Brazil and Malaysia. This causes an emerging markets crisis.
5. In China, where the growth in credit has not been as successful in maintaining GDP growth, the economy continues to slow despite monetary and fiscal stimulus. As a result, economies dependent on China for exports like Australia and Brazil are caught up in more problems.
6. Brazil becomes a focal point of the emerging markets crisis due to its dependency on a slowing China, on oil for tax revenue, and on the US dollar market for debt. But the contagion in EM is everywhere.

So, leaving out the European sovereign debt crisis here, how severe of a downturn in the United States can an emerging markets crisis and a shale oil crisis create? I don’t see this outcome as something that hits the United States’ financial sector in a fundamental way. Rather, it is something that has a derivative impact. We are only going to see a real severe downturn due to both a monetary and fiscal tightening. The monetary piece is not enough to derail the US economy at this point. I still see the underlying pace of consumer growth in the US as 3%ish and that means any external shocks will take it down to 1 or 2%ish growth at the worst. For trend growth to cause recession, we will need to see a major shock to US domestic wage and income levels, something which we are not seeing and I don’t believe this scenario presents.

Bottom line: The US could experience a shallow recession due to a confluence of events that catch the economy at a vulnerable quarter or two when inventory is being purged or capital investment growth is weak. But for a deep recession, we will need to see a shock to domestic incomes and nothing I see due to monetary tightening gets us there at this time. The base case continues to be 3%ish consumer demand growth, attenuated or supplemented by other economic factors.