Midterm Miracles

I'd read about this before, in the form of the maxim that the midterm election year was usually a downer. But after a day's worth of data mining, I realized that since 1950, this pattern has held through bull and bear markets: the market (as measured by the Dow Jones industrials) was always higher in the Presidential election than it was at the midterm elections two years previously.

The way I arrived at the survey period was straightforward: from November monthly close in the midterm year (2002, 1998, 1994, 1990, 1986, 1982, and so on) to the November monthly close in the Presidential election year (2004, 2000, 1996, 1992, 1988, 1984, 1980, and so on). The best return was--surprise!--from 1994 to 1996, with a 75% return on the Dow Jones Industrial Average (DJIA). The worst return was 8% from 1958 to 1960. The average return was 39%, and the median return was 25%. The most recent cycle, 2002-04, saw an 18% return in the DJIA.

If that doesn't make you want to mark your calendar for November 2006, then I don't know what does.

I went further back, as far as the data provided by Prophet Financial Systems could take me. How did this midterm cycle fare during the first half of the 20th century, through the Great Depression and the world wars? From the 1946-48 cycle to the 1902-04 cycle, the average and median returns were less impressive--19.7% and 13.5%, respectively. But the underlying theme for speculators of "Buy the midterm and sell the Presidential" remained a compelling one.

There has been some measure of controversy surrounding Presidential cycles. Much of the controversy, however, appears to do with what specifically is being measured. For example, writing a commentary late last year for the Hussman Funds' "Investment Research and Insight," William Hester suggested that the "average gain in year to of (the) presidential cycle" hid important declines that investors and speculators should be aware of. Hester's own research suggested two things: one, that "the average market performance during the period from January through September of each second year of the Presidential Cycle has been roughly flat since 1933," and two, "the fourth quarter of the second year is actually where many third-year rallies are born." This second point is underscored later in his article: "the 12-month period beginning in October of the second year of the presidential term has enjoyed average total returns of more than 28 percent."

FIGURE 1: DOW JONES INDUSTRIAL AVERAGE, WEEKLY. The midterm bottom in 1994, which featured a historic takeover of the House of Representatives by the Republican Party, was the beginning of one of the most powerful bull market swing in market history.

Hester's conclusions appear to support my contention that measured from November of the midterm year to the November of the Presidential election year, more often than not there is a profitable speculation to be made. Hester's research suggests that the bulk of the benefit from this November-to-November time frame comes from "relief rallies ... in the fourth quarter of year two" plus true outperformance in year 3 and above-average performance in year 4. Hester is pointing to a low late in the midterm year (from which the fourth-quarter relief rallies come), a particularly strong year after the midterm (consider the markets of 2003, 1999, 1995, 1991, 1987, and 1983, for example), and then a solid Presidential election year. Again, these are averages, so some years will underperform relative to previous ones. See Figure 1.

Droke quotes Richard Hoskins, who wrote War Cycles/Peace Cycles, observing that manipulation of the economy as part of the political process is what has made the midterm election cycle possible. "Every four years--market bottoms," rails Hoskins in Droke's accounting. "Congress appropriates money. The president spends it at the time most advantageous for him, just before election. This causes the stock market to rise, making voters happy on election day. With all the money spent, after the election the stock market crashes. A new four-year election cycle then starts all over again."

Unfortunately, Hoskins is a pretty imperfect messenger, aligned as he was with some of the more racist and anti-Semitic ideologies that developed in the United States during the 1950s (one of Hoskins' earlier books was a white supremacist "gem" titled Our Nordic Race). What readers might gain from his observations on economic manipulation by politicians they are likely to lose in his anti-Semetic diatribes against "elite Jewish banking interests."

Fortunately, there are other, surer sources who have opined on the subject, including Marshall Nickles, professor of economics at Pepperdine University, whose brief essay for the Graziadio Business Report provides interesting data in support of the idea that significant buying opportunities tend to arise late in the midterm election year. Nickles notes early in his essay that:

[A]dministrations have often yielded to the temptation to exercise fiscal policy in a manner designed to pump up the economy just prior to a presidential election and thus garner voter approval for the incumbent party.

He adds, "These pre-election actions and campaign promises often have created some euphoria among voters and investors alike."

Yet all this euphoria does not come without a cost, and as far as Nickles' analysis leads him, the bill for that cost has tended to arrive approximately two years before each Presidential election. He writes:

[S]trong doses of fiscal and/or monetary policy stimuli unfortunately often result in creating inflation, which then must be addressed, thereby perpetuating the business and stock market cycles. Given the foregoing scenario, it is not at all surprising to find that the stock market often has made major bottoms about two years before presidential elections ...

Nickles' data, which spans the period from the 1942-44 cycle to the 2001-04 cycle, actually pinpoints an average of 1.87 years into the Presidential term when the market (here, the Standard & Poor's 500) bottomed. That works out to mid-October of the second year of the Presidential term (the inaugural year being year 1). Working for an investment strategy, Nickles compares the fates of two imaginary investors: one who bought the market in October of the second year and held through the end of the election year, and another investor who bought the market in January of the inaugural year and held through September of the second year of the term. Nickles tracked these imaginary investors from 1952 through 2000. The difference in performance was stark. As Nickles observed:

Investor 1 saw the original investment of $1,000 grow to $72,701. This is a percentage gain of 7,170 percent. Investor 2 was not so fortunate ... Investor 2 saw the original $1,000 shrink to only $643, or loss of 36%, in nearly five decades (not inflation adjusted).

FIGURE 2: DOW JONES INDUSTRIAL AVERAGE, WEEKLY. In the midst of a sniper panic in Washington, DC, and preparations for the invasion of Iraq, the markets carved out a bottom that served as the launching pad for a cyclical bull market that lasted for more than three years.

The Hirsches noted the crises that were coincident with midterm election year bottoms in 1962 (the Cuban missile crisis), 1974 (the Nixon resignation), and 1982 (the Reagan recession and fears of "international monetary collapse). And they slyly note that, in Chinese, the word for "crisis" is formed with two characters: "the first, the symbol for danger; the second, opportunity." More recently, we can add the midterm election bottoms in 1994 (Mexican debt crisis and Iraqi disarmament crisis), 1998 (the Russian debt crisis that brought down Long Term Capital Management), and 2002 (the specter of an American invasion of Iraq). See Figure 2.

One of my favorite long-term investing observations remains that of Marc Faber, who observed that a savvy investor could have made a single decision at the beginning of each of the past few decades that would have been more than enough to guarantee a fortune: buying gold in 1970, buying Japanese stocks in 1980, and buying American stocks in 1990. Perhaps "buying real estate" will turn out to be the smartest investment decision of the 2000-10 period--or at least the first half of it. And maybe holding cash will be the shrewd move afterward.

In any event, it is clear that there are a number of ways that investors can take advantage of longer-term trends, cycles, and tendencies in order to position themselves with the winds at their back for extended periods of time. Bottoms, especially major ones, do occur--and often when the masses are distracted in part by some other social or geopolitical crisis. It is the job of the savvy investor to spot those instances and step up, capital in hand, to pounce on the bargains that others fearfully have left behind.