Today's investment environment, roiled by stocks dropping more than 20% from their highs of last October through the first two weeks of July, is as challenging a one as we've faced. Even as we recognize that market declines are inevitable, often creating compelling opportunities for long-term investors, we share concerns over the market's recent plunge, how far prices may drop and how long the pain could last. It's extremely difficult to project the extent of the damage that the unwinding of the housing and credit bubbles will ultimately inflict on the economy and capital markets.
Unlike other periods of market excess, investors have been slow to get their arms around the extent of current problems. But while the unique aspects of this environment remind us that all declines come with their own set of causes and conditions, we also see an aspect typical of all downturns: the sense of fear that magnifies perceptions of risk and causes stock prices to discount more than fundamentals would dictate.
As emotions run high during challenging markets, historical data can be valuable for deriving rational perspective and insight. With that in mind, we decided to look at performance history to analyze how past market downturns have played out and, more specifically, the ultimate impacts of different levels of decline.
Even after an extensive analysis of historical data, we did not expect to discover statistical rules of thumb or predictive factors that provide specific readings on this or future bear markets. We explored historical market downturns to look for patterns or trends that might help us understand why the market has a more extreme reaction in some situations than others and what we might expect during periods following sharp declines.
Enlisting the help of Ned Davis Research, we started by identifying any losses of 15% or more in the S&P Composite—and its successor, the S&P 500—since 1950. (We did not include this year's downturn because we are still in the midst of it.) Our starting point raises two questions: Why set the bar at 15%, and why confine scrutiny to the period since 1950 when data going back much further is available?
We chose the post-1950 period because we feel that data from the first half of the 20th century is less relevant to present market conditions, given changes in the economy and how it is managed. (For example, there are important market sectors, such as technology companies, that didn't exist prior to Word War II.)
As for why we initially chose to identify periods with losses of 15% or more, the simple answer is that this was the level of decline we had reached when we started the work a few months back. We also wanted to have a realistic framework for setting expectations.
While a 10% loss is certainly not fun, we did find numerous market corrections between 10% and 15% that quickly reversed themselves. Including these in our data set would risk minimizing the potential severity of market losses and the length of expected recovery times.
That brings us to our first interesting discovery. Whereas market declines of 10% often reverse themselves, there have not been many periods during which the market decline reached 15% and didn't then go on to hit "official" bear market status—a decline of 20% or more. Eight of the 12 periods we examined experienced declines beyond 20%.
After we conducted this research, the most recent down market extended from 15% to beyond 20%. So our first observation is that while the slide into bear market territory in late June and early July generated many headlines and much hand-wringing, it is not surprising in the context of history.
In one instance when the markets did rebound, the S&P 500 dropped 17.1% in the winter of 1980 and then regained its high by the summer. Three other periods stopped just shy of the 20% mark. We included these three periods along with the 17% drop in 1980 in our analysis.
A Dozen Declines
As mentioned, our study identified 12 periods when the market declined 15% or more. It also showed that over the last 58 years, the S&P 500 has spent far more time going up than down. Whereas the market declined during about 12 of the last 58 years, it has generally risen over the other 46 years. So it's no surprise that bear markets over the past six decades generally have not lasted as long as bull markets.
The average downturn of about a year was less than a third of the average duration (three and one-half years) of the intermittent rising markets. Beneath the surface, though, the 12 periods we examined differed markedly in a number of ways.
On average, a 15% loss in the S&P 500 unfolded over a period of six months. But in some instances, a 15% drop occurred more quickly (within one or two months). Other drops—among them the decline that began in March 2000—took longer.
There does not appear to be any correlation between the amount of time it took the market to sink 15% and the depth of the ensuing trough. However, with a few exceptions, when the market quickly dropped 15%, it also reached its ultimate depth quickly, regardless of the depth of that trough. If it dropped 15% at a more languorous pace, it also took more time to reach its eventual bottom. In terms of the current bear market, which ensued from the market's October 2007 high, the 15% drop in the S&P 500 only took about three months.
However, the market then took almost six more months to exceed the 20% bear market threshold. In this case, this slowness reflects the inability of investors to get their arms around the extent of the current housing and credit problems.
We also examined market downturns in a broader context, exploring any potential relationships between the severity of market downturns and the level of gains in the preceding period. In the 12 rising and subsequently falling periods we examined, only two periods—1982-1987 and 1998-2002—had above-average gains followed by above-average losses, showing once again that past performance can't be relied on to predict future investment results.
Given today's concerns about falling into a recession, we looked for any relationships between market downturns and recessions. Seven of the nine recessions since 1950 overlapped with market downturns.Confirming the market's forward-looking nature, in six of these periods, stock prices began to climb before the recession actually ended. However, a recession is by no means the requisite condition for a market downturn.
In five of the 12 market declines we examined, investors sustained losses of more than 15% without recession. These markets include that of late 1987, which plummeted 33.5% in four months.
However, the average loss in the seven bear markets that overlapped with recessions was 31%, compared with 24.5% during non-recession bear markets. The seven recession bear markets were longer, lasting more than 15 months on average compared with about half that duration for downturns amid healthier economies.
The question of whether we are currently in a recession may be moot as far as expectations are concerned. What actually matters is whether the market is already sufficiently discounting for economic weaknesses.
As long-term investors, we constantly preach that riding out market downturns requires patience. We were curious as to how long it takes for that patience to be rewarded in the recovery phase following a decline.
In the 12 bear markets we examined, on average, investors recovered their losses over a period of about 14 months. This number includes the recovery durations following two severe declines, one in the mid-1970s and the other in early 2000. The median recovery time among the 12 markets was eight months. In nine instances, recovery took only a year or less.
The Good News
Of course, our expectation is that equities don't just recover their losses, they also reward patient investors with significant gains over time. With that expectation in mind, we looked at annualized total return performance for the S&P 500 in the years immediately following downturns of 15%.
The good news is that among the 12 declining periods we examined, the bear market that began in March 2000 was the only downturn producing a negative three-year total return following the point at which a 15% loss was reached. Even in the 1973-1974 period, when the market fell 48% over nearly two years, investors managed to eke out a modest 4.6% annualized return in the three years after they had sustained a 15% loss.
The bad news is that in the five years following a 15% loss, the positive returns were below 10% (annualized) in about half of the periods we examined. However, the bigger picture reinforced our view that patience during downturns will ultimately reward market investors.
On average, the market returned 9.8% annually in the five years following the point at which it reached a 15% loss—slightly less than the historical long-term return on stocks, but still an ample reward for staying the course. Moreover, this figure is a far better return that what investors would earn if they were to become sufficiently risk averse and reduce their equity exposure soon after a significant market drop.
Past market data can't give us answers, of course, but it can give us context. It reminds us that, while all market downturns evoke fear, they are also a normal part of the market cycle, and they all have to come to an end. Also, market downturns typically set up investors for periods of better returns in the years that follow, provided investors have the fortitude and discipline to look forward rather than backward.
Stephen Savage is a partner with Litman/Gregory Asset Management. This article is based on a study originally authored by LGAM Senior Analyst Baie Netzer.