Obscure Economic Indicator: Presidential Approval Ratings

When they go down, stocks go up.

George Bush

If President Bush were a stock, his chart would look less like that of Texas-based oil giant Exxon-Mobil and more like that of Wal-Mart —a spike in 2001 followed by a slow and seemingly inexorable slide down. (Here’s an aggregation of all his poll numbers.)

Bush’s poor job-approval numbers and exhaustion of political capital may be bad news for Republican constituents hoping for investor-friendly legislation like extending the temporary tax cut on dividends. But according to one long-standing indicator, Bush’s woes might be profitable for investors.

Ned Davis Research, Inc., a well-respected research firm based in Venice, Fla., has charted presidential approval ratings as measured by the Gallup Poll against the weekly performance of the Dow Jones Industrial Average since August 1959. (With the firm’s permission, we’ve reproduced the chart here.)

And the data show some strange correlations. Stocks did better when presidents were doing poorly, and they did worse when presidents were more popular. In the weeks when the presidential approval rating was below 50 percent, stocks rose at an annualized rate of 9.2 percent. In weeks when the approval rating was between 50 and 65 percent, the Dow rose at an annualized rate of 5.4 percent. And in those periods when approval ratings were above 65 percent (about one-fifth of the time), the Dow rose at a 2.6 percent annual rate.

Note that the stock market has been generally positive in all those periods, reflecting the market’s long-term upward trend. Note, too, that people are generally favorably disposed toward their presidents. The approval rating has been above 50 percent 62.9 percent of the time.

What gives? As Ed Clissold, senior global analyst at Ned Davis Research, describes it, polls are sentiment indicators. In many instances—though not all, as we’ll see below—presidential approval ratings reflect the public’s view of how full their wallets are. “If economic growth is not particularly strong, or there’s a recession, the president is usually held accountable,” said Clissold. But it takes a while for bad economic news to filter into polling data. By the time poor approval ratings measure this economic pessimism, much of the bad news has already been priced into stocks. Shares have already been pushed artificially low and thus are more likely to bounce back up. A similar process happens in reverse. Highly positive presidential ratings often stem from widespread optimism about the economy. Emboldened investors tend to plunge more cash into the stock market when they think things are going well. And as we know from bitter experience (Dow 36,000, anyone?), optimism is always highest just when stocks are about to tank.

This contrarian dynamic certainly seems to have been borne out in the Bush years. His popularity spiked in 2001 and remained at very high levels as the stock market plunged in 2002. As Bush’s popularity declined, the market recovered, and the bourses have remained buoyant even as his popularity began to plumb new depths. Bush has never been more unpopular than he is today. And yet the S&P 500 today pierced a level it hadn’t seen since May 2001, while the Nasdaq sits at a five-year high.

But there are a few caveats to the presidential indicator. Historically speaking, when the public thinks presidents are doing badly, it has been good for stocks. But when the public thinks they’re doing really badly, it’s bad for stocks. Ned Davis Research crunched the numbers on stock returns when presidential approval ratings sink below 38 percent, which they have only 7.2 percent of the time since 1959 (Bush stands at 37 percent in the current Gallup Poll). During such periods of “extreme low ratings,” says Clissold, the Dow falls at a 2 percent annual rate.

Second, polls are measures of feeling—not trading triggers. They are indicators of relative measures of optimism and pessimism, not when to buy and sell stock. Clissold watches several other sentiment indicators, including the Conference Board’s Consumer Confidence Index. And that measure shows that confidence in the economy is quite high. In other words, the sentiment indicators are sending out mixed messages.

Finally, presidential approval ratings sometimes have comparatively little to do with the economy. The economy and the stock market were doing quite well during Lyndon B. Johnson’s second term, when the Vietnam quagmire undermined his popularity. “Right now, the presidential approval rating probably reflects the global situation as much as what is going on with the domestic economy,” says Clissold. In other words, it’s likely that the American people think Bush is doing a poor job not because the economy is doing badly, but because they simply think he’s doing a poor job—especially on Iraq.