Lesson From the Past Courtesy of the S&P 500

Dec.20.10 | About: SPDR S&P (SPY)

A lot can happen in 90 days. This article compares the average 90-day increase or decrease in the S&P 500 for two periods of time: 1950-2010 and 2000-2010, grouping the data by percentiles. The past decade saw two devastating market debacles that have traumatized many investors. This article puts the lost decade into perspective, by comparing it to the longer period of which it is a part. The conclusion is debatable - the lost decade may have been an aberration, or it may have been a paradigm shift, leading to substantially more risk in equity markets.

Using S&P 500 closing figures since January 1950, the trailing 90-day return was calculated on a daily basis. The resulting percent changes were then sorted and grouped into 100 buckets, the contents of which were averaged. The exercise was repeated starting with January 2010, and the two sets of results compared graphically. Here is the chart:

Click to enlarge
(Click to enlarge)

The results for the longer period portray a market where the most likely result was a respectable profit. Profits cut in at the 37th percentile, and by the 50th they average 2.23% quarterly. That's well in excess of the 8% annually people used to talk about. There was some inflation in that.

The past decade is a different story. Profits cut in at the 46th percentile, and by the 50th they averaged 1.12% quarterly, about 4.5% annualized. On the downside, the curve is not smooth; and pain, defined as down 10% or more, cut in at the 12th percentile.

Stress Testing

I was drawn into this investigation in the process of developing a strategy stress tester. Leveraged strategies as a general rule perform well in moderate markets but have a tendency to blow up under stress. By examining historical data, the probability and severity of adverse outcomes can be defined and then used as a basis for stress testing various strategies.

Predictably, bullish leveraged strategies when tested against the 2000-2010 period performed poorly. Although some of them would outperform reliably if applied continuously over 25 years or more, for five year periods there was too much variation, much of it negative. The only way to justify using them was to assume that nothing under the 5th percentile was going to happen. Many financial companies ran their businesses along those lines during the past decade. Some of them are no longer with us, the others are sadder but wiser.

Using the longer experience period, results were more encouraging. Disregarding only the 1st or 2nd percentile, well designed strategies consistently outperformed for most time frames. The task of hedging or stabilizing the portfolio with cash becomes more manageable. The quarterly average return of 2.23% would compound to 9.5%: however, by adjusting the whole curve downward to bring that down to 5%, more realistic results were projected, and quite acceptable.

Investment Implications

The issue of whether the past decade represented an aberration or a paradigm shift is a major concern in considering how to conduct investment activities going forward. It is difficult to put aside ideological or political blinders and focus on the extent to which the conditions that created the instability still exist.

Monetary policy that is loose by comparison with past history appeared in 2001 and has become more pronounced following the crisis. Fiscal policy has developed along lines that refuse to equalize revenue and expenditures, and continues to do so. To generalize, these two factors in the past have created periods of prosperity. However, common sense suggests both of them have logical limitations that have been or soon will be exceeded.

Deregulation, and specifically the abandonment of prudential regulation, has been ongoing for over 20 years. Systemic risk in the past was handled on an ad hoc basis: successfully prior to errors of judgment that permitted Lehman's demise to threaten the entire economy and financial system. At this point, regulators have tools to deal with systemic risk on an ongoing basis and legal standing to re-impose prudential regulation, should they be motivated and supported along those lines.

Housing asset prices were inflated by a combination of lenient tax policy, loose monetary policy and removal of prudential regulation. This resulted in the creation of worthless assets, a process that was exacerbated by speculative activities designed to propagate and profit from the resulting losses.

My impression is that the authorities have both the tools and the determination to prevent another panic. As such, I believe that market conditions will be more similar to what prevailed over the 1950-2010 time frame, in that the panicky destruction of value will be avoided. A recurrence of the madness that led to the bogus housing assets is unlikely, based on the public's unwillingness to participate again to the same degree they did in the past.

In any event, I intend to conduct my investments along lines that will be profitable if the 90-day increase/decrease of the S&P 500 index is more similar to what prevailed during the period from 1950 to the present than to the past decade.

Disclosure: I am short SPY.

Additional disclosure: I'm net long many constituents of the S&P 500. Short SPY by means of puts as a hedge.