The period following the broad acceptance of a recession is usually far better for investors than the period that precedes it.

William Hester, CFA
November 2008
All rights reserved and actively enforced.

The job of NBER's Business Cycle Dating Committee - to mark the economy's peaks and troughs - is getting easier. Robert Hall, the Stanford University economist who leads the committee, said in early October that it was not clear the economy had entered into a recession because "we've had this perplexing period of rising output and declining employment."

But this week the government reported that GDP had contracted at a 0.3 percent pace in the third quarter. And some of the data within the report was much weaker. Consumer spending dropped 3.1 percent on an annual basis, the first decline since 1991 and the biggest since the 1980 recession. This contraction and the ongoing decline in labor statistics and industrial production spurred economist Jeffrey Frankel, another member of the Business Cycle Dating Committee (BCDC), to write in his blog this week that "at this point there can be little doubt that we are truly in recession."

So the official announcement from the BCDC may soon follow. Will it play a role in how stocks perform? Or said another way: how does the market perform from the point that the majority of investors recognize that the economy is in recession?

The following graphs look at the performance of the S&P 500 following dates where recession is broadly recognized, based on the available evidence at the time. It looks at stock returns and the volatility of those returns over four periods: 3-month, 6-month, and 1 and 2-year periods. Of course, there's no precise historical record of when the majority of investors decided that the economy was in recession. But we can try to infer widespread acceptance by coming at the problem from a few different directions.

The clearest set of dates with which we can be sure investors widely believed that the economy was in recession was when the Business Cycle Dating Committee declared it one. In real time, the group has dated four recessions: 1980, 1982, 1991 and 2001. The graph below shows the percent change in the S&P 500 following each announcement.

Since 1980, market returns have been mixed following the BCDC recession declaration dates. Despite a declared recession, stocks were typically unchanged 3 months later. In three instances (1980, 1982, and 1991), stocks were 10-20% higher after a year, but in the 2001 instance, stocks were down about 20% a year later. Two years after the declaration, stocks tended to be moderately higher, but usually after a lot of sideways movement.

The NBER dated the beginning of the 1980 recession in June of that year. A rally that began in March of that year continued into early 1981, finally rolling over on the prospects of a second recession beginning during the summer of 1981. That second recession would end up being longer and deeper than the preceding slowdown, and stocks continued to move lower after the peak in the economy was identified, eventually falling more than 10 percent. By the summer of 1982 stocks were trading at a price to peak earnings multiple of about 7, from which they rallied strongly.

The third series from the group includes the 1991 recession, where the peak was identified in April of that year. Here again the market was already rising by the time of the BCDC's announcement, likely sensing that the recession would be brief. The S&P did end up climbing 20 percent over the next two years as the economy recovered. The average returns based on these dates were weighed down by the performance of the market following the dating of the 2001 recession, when the market fell another 33 percent before eventually bottoming in 2002.

A second way to estimate when a recession was widely accepted is through economists' forecasts. One of the oldest surveys is the Survey of Professional Forecasters, which is now administered by the Philadelphia Federal Reserve. In addition to asking economists for forecasts of GDP and inflation, they are also asked about the probability of a contraction in output during the first full quarter following the survey period. Dubbed the Anxious Index, it has spiked during each of the last six recessions. For our purposes, each time the survey registers a probability of a contraction greater than 45 percent, we'll mark the end of the survey quarter as the date where a recession was broadly recognized.

We're able to take this analysis back to the early 1970's. Here again, despite a well-recognized recession, short term returns were mostly flat, though with above-average volatility. The graph shows that two-year returns were mostly strong from the recession recognition date based on the Anxious Index. Returns averaged 20 percent over two-year periods.

The periods with the deepest declines - December 1973 and December 2001 - are instructive, because stock valuations were still fairly rich at the point that the recession was recognized. The Anxious Index in December of 1973 climbed above 50, and receded early in 1974 before spiking again. In the 1973 instance, the price-to-peak earnings ratio was 12.7, but this period was unusual in that earnings actually grew during the recession. Based on 1972 end-of-year earnings, the p/e multiples was 16 in December 1973. In contrast, the much stronger 1974 period started at a price-to-peak earnings multiple of 7 (about 10 times 1972's peak earnings level). This is an important difference in beginning valuation levels. Similarly, the 2001 recession was acknowledged by the end of 2001, but stocks kept falling. Once again, high valuations at the date of the announcement were part of the problem, as the price-to-peak earnings ratio was 22 at the time.

The third worst drawdown occurred shortly after economists recognized a slowdown in March 1970. Stocks fell a bit more than 20 percent before recovering. That recession capitulation period also began with a peak earnings ratio of 14.8 when the recession was recognized. In contrast, the market experienced strong gains following the Anxious Index signal in December 1974 and December 1990, where peak-earnings multiples were below average.

A final way to measure recession capitulation comes from a measure of investor nervousness - the level of intraday volatility. Market analyst Ned Davis measures this using 44-day moving average of the percent difference between the market's daily high and low price. Unlike the VIX Index, we can track this measure of fear in the market back to the early 1960's. The graph below shows the occurrences where the moving average of the intraday volatility measure rose above 2.5% for the first time during a period which was eventually determined to be a recession.

Market returns were typically stronger using this signal because they mostly followed steep sell-offs in the market. But here again, returns were very mixed over the short term. Over longer-term periods, returns were generally strong and volatility moderated. Even including the 1973 loss, average 12-month returns were 14 percent. Two-year returns averaged 23 percent. The intra-day volatility signal came later in the recessions of 1970 and 1980 than when the Anxious Index is used to measure broad recognition. In both cases full-period returns were higher with less volatility.

Overall, market returns following the broad acceptance of recession using the three methods above have been mixed, with a bias toward gains except when valuations were still high when the recession was recognized. Despite a recognized, ongoing recession, short-term returns were fairly flat, though with above average volatility. Returns over a longer-term period tended to be stronger. Over our three indicators, average 12-month returns were 9 percent. Over 2-year periods returns averaged 18 percent.

Though the Business Cycle Dating Committee has yet to weigh in on an official recession, the two other indicators above suggest that a recession was broadly recognized either late in September of in early October, adding to the already severe losses provoked by the recent credit crisis. The accompanying damage is consistent with what John Hussman observed a few months ago in The Risk of Conceding Recession, which noted "Once an ongoing (and in my view, probably deepening) recession becomes broadly recognized, we may observe abrupt losses as the likelihood of more sustained earnings disappointments and much broader default risk becomes reflected in one fell swoop."

Although there were a number of times when markets moved lower after a recession was recognized (sometimes substantially so if valuations were also high), the period following the broad acceptance of a recession is usually far better for investors than the period that precedes it. The graph below shows the S&P's return from the beginning of each recession-induced bear market up to the point of acceptance, using the Anxious Index as the indicator.

The typical recession-induced bear market goes through three stages. The first stage is where stock prices grind lower as the economic data begins to suggest a developing recession. The depth of the declines and the volatility increase as more investors begin to price in the probability of a recession. This recognition comes slowly. On average, the first 100 trading days of recession-induced bear markets contain only a quarter of the bear market losses and have lower volatility compared with the full downturn.

The second stage is where the majority of investors recognize that the economy is in recession. This period usually coincides with deep and abrupt declines, with very high volatility.

The final stage is a work-out period, where stock investors struggle to determine the depth and duration of the recession. It's a period that's been marked by both strong rallies and continued declines, partly determined by beginning valuation. Rallies often retrace. And the early stages of this work-out period almost always contain above-average volatility. But once a recession has been recognized, and valuations become reasonable, the prospects for longer term returns typically improve.

---

The foregoing comments represent the general investment analysis and economic views of the Advisor, and are provided solely for the purpose of information, instruction and discourse.

Prospectuses for the Hussman Strategic Growth Fund, the Hussman Strategic Total Return Fund, the Hussman Strategic International Fund, and the Hussman Strategic Dividend Value Fund, as well as Fund reports and other information, are available by clicking "The Funds" menu button from any page of this website.

Estimates of prospective return and risk for equities, bonds, and other financial markets are forward-looking statements based the analysis and reasonable beliefs of Hussman Strategic Advisors. They are not a guarantee of future performance, and are not indicative of the prospective returns of any of the Hussman Funds. Actual returns may differ substantially from the estimates provided. Estimates of prospective long-term returns for the S&P 500 reflect our standard valuation methodology, focusing on the relationship between current market prices and earnings, dividends and other fundamentals, adjusted for variability over the economic cycle (see for example Investment, Speculation, Valuation, and Tinker Bell, The Likely Range of Market Returns in the Coming Decade and Valuing the S&P 500 Using Forward Operating Earnings ).