Hussman Funds: A Tale of Two Data Sets – August 31, 2009
Excerpt (click on the link above for the full version)
Straight to the chase – the U.S. stock market is overvalued, but not severely so. Prospective 10-year total returns for the S&P 500 are uninspiring, but they have been even more dismal for much of the past decade. Market action in the major indices has been clearly strong, as is breadth, but evidence of robust sponsorship is weak – particularly on the basis of trading volume. Several measures of risk aversion have abated, but have been replaced by exuberant investor sentiment, which has not been as bullish since the 2007 market peak. From the standpoint of typical post-war market cycles, there is little to get excited about, but there also does not seem to be a great deal to complain about either. That is, again, if we assume that this is a typical post-war market cycle.
As I frequently note, we try to align our investment positions based on the return to risk profile that we can expect, on average, given prevailing market conditions. Our dilemma here is this. There are two sets of data from which to draw those averages. One would include ordinary run-of-the-mill economic and market cycles using post-war data from the U.S. alone. The other is broader, and includes market behavior, both in the U.S. and in other countries, following major crashes. Extended economic dislocations were typically required before fundamentals durably improved.
Even giving the two possibilities equal weight is harsh, because as I’ve repeatedly noted, post-crash markets have included advances as large, and larger, than we’ve observed since March, but with devastating follow-through. In the current situation, our only choice is to consider the full data set, because we have no basis – whatsoever – to rule post-crash dynamics out in view of prevailing economic conditions. As a result, as things stand presently, we hold a substantial number of index call options (about 1-2% of Fund assets), but are otherwise hedged. The strike prices and expirations of our index calls currently result in a modest effective exposure to market fluctuations. Those call options are in place primarily to reduce the impact of our hedge in the event of further strength from here – to date, they have largely offset what would have been lagging performance due to our continued lack of holdings in financial stocks.
If we had a reasonable basis to believe that the recent economic downturn was an ordinary run-of-the-mill post-war recession having no lasting structural impact, and believed that the record profit margins observed in 2007 (about 50% above the historical norm) could be recovered and sustained, we would infer an average return/risk profile for the market that is still much less favorable than we have normally observed following bear market lows, but strong enough to warrant the removal of a good portion of our hedges outright, with a willingness to remove another portion of our hedges on market weakness.
On the other hand, using the same essential measures of valuation and market action, but including periods of major economic dislocation into the dataset, produces average return/risk inferences that are substantially less favorable. Indeed, the reason we were somewhat “burned” during the fourth quarter of 2008 is that we expected – too early, in hindsight – a powerful rebound from the extremely oversold conditions we observed, based on normal market behavior. The larger dataset also includes periods of similarly powerful rebounds – but the attempt to participate in them is less appealing due to their lack of predictability as well as their sometimes abrupt and costly endings.
We’ve seen several reasonable analyses of average market performance during, say, the 12-18 month period following the end of a recession, based on U.S. data. On average, post-recession market performance has been good, regardless of initial valuation. The difficulty with those analyses, however, is that they contain a premise that we are unable to take as given – specifically, that we presently have in hand an assurance that the recent downturn is behind us. So yes, if you examine the behavior of the stock market even following the end of the Great Depression, with the benefit of hindsight as to the date the Depression ended, you’ll find that stocks typically did well. But to assume the benefit of hindsight as a premise of one’s analysis strikes me as dangerous. There is no assurance, in my view, that the current economic stabilization we’ve observed is more than an artifact of enormous government spending, or that the underlying structure is invulnerable to a steep double-dip
