This article is a guest contribution by John P. Hussman, Ph.D., Hussman Funds.
A week ago, we accumulated enough evidence to conclude that the U.S. economy is most probably headed into a second leg of recession. It is unclear whether this will be identified as a second recession or a continuation of an existing downturn. In either case, I've repeatedly noted that the apparent strength in the U.S. economy over the past year has been driven almost exclusively by an almost inconceivably large burst of fiscal and monetary "stimulus" last year, whereas intrinsic economic activity has stagnated. Personal income remains at its lows once government transfer payments are excluded, which is in stark contrast to typical post-war recoveries. Weekly jobless claims are pushing again toward 500,000, whereas prior post-war recoveries have seen jobless claims quickly retreat below the 400,000 figure that roughly delineates job growth from continued job losses. The most straightforward explanation of the economic data is that we've observed a stimulus-led recovery that has not translated into private economic activity, and that the effects of the stimulus are now diminishing.
Our recession warning composite, on which part of my present concern is based, reflects essentially the same combination of factors that produced the warning I reported in the November 12, 2007 weekly comment Expecting A Recession , as well as the recession warning I reported in October 2000 . Based on the high correlation and roughly 13-week lead that the ECRI Weekly Leading Index provides, compared with the ISM Purchasing Managers Index, I noted last week that the use of the ECRI Index provides somewhat more timely signals. I want to emphasize, however, that our use of the ECRI Index is as one component of a broader set of indicators that we look to observe in concert to produce a recession warning.
Emphatically, I am not suggesting that the Economic Cycle Research Institute itself is warning of a second downturn. How the ECRI uses its data is its own concern. But I am equally emphatic that when the ECRI index is included among the other components of our recession warning composite, the resulting recession signals have always and only been observed during or immediately preceding post-war economic downturns. The abrupt dropoff this month among various Purchasing Managers Indices in U.S., China and Europe is consistent with the ECRI data, though given the typical lead-time of the ECRI, we may have to wait another month or so to see those PMI indices breach readings of 50-54. I'll note in passing that the ECRI Weekly Leading Index growth rate deteriorated to -7.7% last week, from -6.9% the previous week.
From a Bayesian standpoint, if you always observe a certain combination of information when X occurs, and never observe that same data when X is not present, then even if X is hidden under a hat, you would conclude that X is most likely there. If I see clowns walking around the grocery store buying peanuts, and there's a big top tent with two unicycles in front of it in the middle of what is usually an open field, I'm sorry, I'm going to conclude that the circus is in town.
Investment Implications
My most urgent concern over the past few months has been directed toward investors with spending plans (tuition, home purchase, baby, medical, retirement) that rely on the near-term availability of funds. If you are following a disciplined investment strategy, your portfolio is well-diversified, and you are tolerant enough of risk that the declines in 2000-2002 and 2008-2009 did not materially derail your investment discipline, then please, ignore my views and the views of everyone else and just follow your discipline.
