THE verdict of
the Dow Jones industrial average is in, and it
says Al Gore is headed for the White House.
The Dow closed
10,971.14, up 4.3 percent since
the end of July after a roller
coaster period that saw it soar
during August and then sink in
September and early October,
only to put on a strong rally in the
final two weeks of the month.
The Dow's verdict
flies in the face
of the conventional wisdom of the
polls, which show George W.
Bush with a narrow lead. But the
Dow's record of presidential
election forecasting is such an
excellent one that it commands
respect. It has correctly forecast
22 of the 25 presidential elections
since the Dow industrials made
their debut in 1897.
pattern is that when
the Dow rises from the end of July
through the end of October — the
period when the attention of voters
is becoming most fixed on the
presidential election, the incumbent
party keeps the White House. But
when the Dow goes down, the
insiders are thrown out.
This year the
Dow has seemed to
be a leading indicator of the polls,
at least of the most widely
followed tracking poll, the
CNN-USA Today Gallup poll. In
early August, just after the stock
market began its August rally, Mr.
Bush enjoyed a 17- point lead.
That soon began shrinking, and by
Sept. 6, a few days after the market's summer high was reached, Mr.
Gore held a 3-point lead. That widened to a 10-point gap, Mr. Gore's
biggest in the poll this year, on Sept. 20. By then, however, the rally had
petered out, and the market had come down a bit.
The Dow fell
sharply through Oct. 18, and by then Mr. Bush was back
to a 10-point lead in the polls. That lead then evaporated by Oct. 23,
while the stock market was rallying. But within a few days, even as the
Dow continued to move up, Mr. Bush's lead reappeared, peaking at 13
points on Oct. 26. But it has since been shrinking as the Dow rallied
again, and the latest result gives Mr. Bush a three-point lead.
explanation for the Dow's success as an election indicator in
the past is not that voters react to the stock market, but that both the
market and the election are measures of how Americans feel about the
future. A rising market indicates optimism, and such a good feeling can
benefit the incumbent party.
The most recent
failure of the Dow as an election indicator came in 1968,
a year that in some ways is similar to this one. Then, as now, the
Democratic candidate was an incumbent vice president whose
Republican opponent's last name had been on the ballot — and lost —
eight years before. That election pitted Vice President Hubert Humphrey
against Richard M. Nixon, who had lost the election in 1960. Now Mr.
Gore is opposing the son of President George Bush, who was defeated
important, the country was in what was then the longest
economic expansion in American history, a period of growth that would
not end until late 1969. That expansion now ranks second, having been
eclipsed by the current one that began in 1991. If Mr. Bush wins this
year, perhaps it will be right to conclude that voters are less focused on
the economy after a very long period of prosperity dulls the memory of
The other times
that the Dow failed as a forecaster were in 1932 and
1956. In 1932, a recovery in the Dow was tiny compared with the fall
from the 1929 peaks and had no chance to restore Herbert Hoover's
reputation or popularity. In 1956, Dwight D. Eisenhower cruised to re-
election despite a dip in the Dow.
On all those
three failures of forecasting, the Dow did reverse itself in the
couple of weeks before the end of October, rising in 1956 and falling in
1932 and 1968 — a move that could be seen as qualifying the otherwise
incorrect election forecast.
This year, however,
the last couple of weeks in October were strong
ones for the Dow, and if Mr. Gore nonetheless loses, it would seem to
indicate a clear failure for the Dow as an election indicator. Nor does the
weakness of the Nasdaq provide a convincing contrary forecast. Even
after yesterday's 5.6 percent rally, the Nasdaq composite is down 10.5
percent since the end of July.
for Republican hopes, the Nasdaq's record as a market
forecaster is not very good. In the seven presidential elections since it
began in 1971, the Nasdaq has correctly forecast just four of them. It
rose in 1980 and 1992, when incumbents were nonetheless defeated,
and fell in 1988, when the incumbent party kept the White House.