Last week I attended a seminar by Nobel Prize Winner Bob Lucas, on the current international credit crisis. He gave a very big picture view based on a comparison to the 1930s crisis. His bottom line was that as opposed to the Fed response in 1931, this time around the Fed policy response is not only the right one, but most likely in the right doses and timing. The policy response has been generally blamed for the severity of the 1930s economic depression. This talk was before last week’s Fed cut/signals.

I do not intend to produce an accurate nor a complete reckoning of his talk, but to mention a few points relevant to a few provocative points of my own. These are all related to my first write up, where I expressed a big picture view on why I think this downturn has a lower-bound not of catastrophic proportions. With that provocative motivation, I will make the following points along the discussion below:

1- We will not live a downturn like that of the 1930s

2- The Fed response should be enough, and hopefully fiscal policy is restrained (using both could show the lack of a clear diagnosis)

3- There should be no grand change to the structure of the US economy, nor a major change in regulation. There is an optimal level of losses that should work better than new regulation

The US Economy in the Long Run

The US economy has shown a remarkable 120+ year performance in terms of growth and volatility, with the exception of the great depression of the 1930s. When you look at data from such a long-term perspective, you see that there has been no major changes on how the US economy grows, nor at the beginning of the period while the industrial revolution was unraveling, nor in the recent years from the apparent liquidity bubble or US credit expansion. More importantly, after the great depression, the economy went back to its steady medium-term pattern of growth. (obviously, such long term trends are analyzed under high-frequency filters, in this case the standard Hodrick-Prescott filter)

The Parallel to the 1930s

There is a very interesting parallel between the 1930s and now. The framework of analysis is from Friedman-Schwartz seminal “A Monetary History of the United States” (chapter 7). The blame for the sharp GDP contraction that followed 1931 (accumulated in total to almost 40% peak-to-trough) is assigned to the sharp and continued run on bank deposits. In simplistic big picture terms, the policy mistake was to allow bank reserves to drop in 1931. What caused the bank run is less relevant, as you will see. The parallel to today’s financial situation is the freeze of the commercial paper market (CP).  When you look at stocks of CP, financial institutions dwarf other economic sectors in their reliance of such source of funding (by a multiple of 3). The economy as a whole is supposed to allocate liquidity funds to the CP market. As banks funded to a large extent in the CP market (liability side of their balance sheet), then had in their assets ever more complex investments, the valuations of which started to collapse due to both fundamentals and liquidity/technical reasons. The freeze of the CP market is equivalent to a generalized bank run (think of the liability side of financial institutions’ balance sheets). As in the characterization of the 1930s, what cause the freeze in the CP market is less relevant (though it all started with depreciations in sub-prime valuations from a housing market downturn, which in itself cannot cause a crisis of this magnitude).

Under this monetary framework, the impact of the bank run on banking activity is to some extent monitored by the behavior of banks’ reserves, which during the runs of 1929 and 1930 were kept roughly constant. Then, the Fed let bank reserves contract in 1931 (after they had been fairly steady despite a massive bank run). Such a bank run meant a reduction of means of payment (spending) in the economy, which produced the GDP contraction, as the beneficiaries of the graces of bank assets were forced to do without that funding/liquidity. Generalized asset sales happened then, as we have seen recently. Nowadays, the Fed has multiplied bank reserves by more than 10 since September. Moreover, aggregate data as shown in a recent paper by the Minneapolis Fed, lending to most other sectors in the economy has not necessarily collapsed (from “Facts and Myths about the Financial Crisis of 2008”, by Chari-Christiano-Kehoe, at the Minneapolis fed website).

Following Lucas’s argument, the parallel breaks down when Bernanke’s Fed aggressively allows monetary aggregates to expand several times (especially bank reserves), different from what happened in 1931. This is without even considering this week’s cut and signals of further asset purchases from the Fed.

The crux of his argument is that only the monetary authority has a chance at substituting means of payment (thus spending capacity for the private sector) when a bank run does occur. Thus, his point is that basing the comparison on the most appropriate analytical framework available (i.e. Friedman-Schwartz), the parallels due not hold. In other words, there is no empirical foundation to predicting the same kind GDP contraction.

The problem with the current policy response is the double-whammy. Both fiscal and monetary policy appear to be going ‘pedal to the metal’. Using both seems to be an indication of the lack of a clear view of what is this crisis about. Monetary policy is called for if your view coincides with the one above, while fiscal policy is called for if your interpretation has more of a Keynesian flavor (i.e. liquidity trap). In other words, the policy response should either be a sharp monetary expansion (which should eventually be withdrawn to avoid inflation), or it should be a sharp fiscal expansion because of the liquidity trap. But the US is embarking in both.

In the medium-term (i.e. whenever we pay the price of the double-whammy-let’s-try-everything strategy) this will produce higher inflation, higher interest rates and slower growth. However, today’s markets seem to respond to whether it works in the short-term. The only reason why the overkill strategy might not is that the signal of lack of leadership shown in the lack of a single diagnosis dents confidence. But I doubt the world is as neoclassical as me. I continue to believe this downturn has a lower-bound much higher than most people think.

No Major Changes to the Economic Structure and Regulation

The US economy has had a 75+ year run without a major crisis of this kind (or 120+ years with only 2 serious crisis). I would call that a success, and enough reason to hesitate before embarking in any major change. Growth data (properly filtered) shows an amazing streak of high average growth with relatively low volatility. Is there any other experience better than the US economy 1935-2007?

As expressed above, if one of the key suspects for this credit crunch is the disproportionate amounts of money-market-type-of-funds invested in the CP market, which at the end was funding much riskier assets than they were supposed to.

No amounts of regulation will prevent investors from taking disproportionate amounts of risk at the margin during periods of sharp increases of liquidity. Actually, losses will. I think the excess allocation of liquidity funds to financial institutions CP paper was disproportionate (see Chari-Christiano-Kehoe 2008), but it is not clear to me how to prevent such a phenomenon with new non-crippling regulation such an event. Risk-taking will evolve, and if a long period of high growth and low rates happens again, excessive risk taking will happen again. If we over-regulate now, we might never test this…

To wrap it up, I think a very relevant policy trade-off is the bailout vs losses. The integrity of the financial system requires for the governments to step in and bail investors out when not doing so means serious systemic risks. However, losses are the best preventive measure to minimize the probability of similar excessive risk allocations in the future. The best recipe for better risk management is good data on how similar risks produced economic losses.