Skip to main content
ABC News
Could A Stock Market Downturn Tank Trump’s Approval Rating?

The stock market fell on Monday. A lot. After a roller-coaster day, the Dow Jones industrial average ended down 1,175 points, making for the largest single-day point decline in its history.

In percentage terms, the decline isn’t nearly as bad — the 4.6 percent drop is well short of the record 22.6 percent sell-off that occured on Oct. 19, 1987, otherwise known as “Black Monday.” The Dow would have had to lose almost 5,700 points on Monday to equal Black Monday on a percentage basis.

But it’s likely that there will be further volatility in the days ahead. The Dow had a very good first few weeks of January, gaining almost 2,000 points and peaking at an all-time closing high of 26,616.71 on Jan. 26. The index has given up all of its 2018 gains and then some since then, however. Meanwhile, the Chicago Board Options Exchange’s VIX, a measure of implied volatility in stock prices over the next 30 days, spiked dramatically on Monday to finish its session at 37.3 points, a level rarely seen since the end of the financial crisis.

This is usually the point at which I’d pause to deliver various pedantic admonitions about the stock market. For instance, I’d tell you that the Dow isn’t a very good stock market index as compared to broader and more robustly constructed ones such as the S&P 500, which declined by slightly less on Monday. I’d also tell you that the stock market isn’t always a very good proxy for the nation’s overall economic health. In fact, the latest decline in share prices has been triggered, in part, by fears that the Federal Reserve would raise interest rates because of strong wage and job growth. Good pocketbook news for workers can sometimes be bad news for investors, and vice versa.

I’m someone who writes about politics, however — and in political terms, the decline in the Dow is potentially important. President Trump has spent lots of time bragging about the stock market over the past year, from dozens and dozens of references in his Twitter feed to a riff about it in last week’s State of the Union address. Moreover, the stock market is a highly visible economic indicator. Sure, something like real disposable income per capita might be a better overall measure of economic well-being. But the Dow gets mentioned a lot more often on the evening news.

Overall, the stock market still looks pretty good for Trump, with the Dow having gained 23 percent during his presidency so far. But could a further reversal in those gains harm his approval ratings, which have been improving lately? Trump’s approval rating is now 40.4 percent in FiveThirtyEight’s index — not good, but the highest it has been since May 15.

The short answer is, sure, there could be some risk to Trump. One should not necessarily expect there to be an obvious one-to-one correlation, however. After Black Monday in 1987, then-President Ronald Reagan’s Gallup approval rating actually improved to 51 percent from 49 percent two months earlier. The point is not that Black Monday helped Reagan — it probably didn’t — but that whatever effect it had was swamped by other news events, plus whatever random statistical noise there was in the polling average. My past research also suggests that stock market growth has historically been a fairly weak predictor of a president’s odds of being re-elected.

More recent academic research has found that there probably is some relationship between presidential approval ratings and the stock market, however. Furthermore, there’s fairly robust evidence linking overall economic performance to presidential approval.

I’m marginally skeptical of these studies, but only because it’s difficult to separate out the influence of the stock market (or any other economic indicator) from the rest of the economy. The present sell-off aside, there usually is some correlation between the stock market and other measures of economic well-being. The massive stock-market decline in 2008 was indicative of much broader problems with the economy, for example. Studies that take a bunch of highly correlated variables (e.g., the stock market, unemployment, inflation, consumer confidence, etc.) and attempt to correlate them with another variable (e.g., presidential approval) can run into all sorts of methodological problems such as overfitting.

In other words, there are lots of reasons to think there’s a reasonably strong1 relationship between economic performance and presidential popularity. It’s very difficult to say which particular indicators the public cares most about, however. (The approach we take for FiveThirtyEight’s election forecasts, which are partly based on economic data,2 is essentially to average a bunch of economic indicators together — including the S&P 500 — but weigh them equally.)

Moreover, the relative importance of different economic indicators can change from presidency to presidency, based on which ones that politicians and the media are talking about and which have had more visible effects lately. Inflation was a much more salient political topic during the 1970s and 1980s than it is now, for example.

But therein lies the risk to Trump. He has frequently highlighted the stock market — something that President Obama rarely did, by contrast, even though the Dow rose by almost 140 percent during Obama’s two terms in office. If it continues to decline, it will be tough for Trump to dismiss its significance. Nor will he be able to cast doubt on the reliability of the numbers, since the price of a stock is not a government-created statistic and is simply a market price.3

Most presidents avoid talking about the stock market because of its unpredictability and volatility. In an era of algorithmic trading, there’s some relatively minor piece of economic news, and voila! — the Dow can swing by 2,000 points in a few weeks or several hundred points in a day. By contrast, although macroeconomic conditions are not highly predictable at intervals more than about six months in advance — it’s too soon to know quite what the economy will look like at the midterms — we’re at least talking about changes that play out at intervals that span months to years and not days to weeks. U.S. GDP, currently $17.3 trillion, is extraordinarily unlikely to jump to $20.8 trillion or $13.8 trillion next month (a 20 percent increase or decline, respectively). But if the Dow were 20 percent higher or lower a month from today, it wouldn’t be any enormous surprise. Trump should probably stick to bragging about jobs and wage growth, because those numbers have also been pretty good and the gains are more likely to hold.


  1. Although by no means deterministic.

  2. The “polls-plus” version is partly based on an economic index; the “polls-only” version is not.

  3. There’s more work that goes into creating a stock market index, but it’s not nearly as complicated as how other economic indices such as GDP are calculated.

Nate Silver is the founder and editor in chief of FiveThirtyEight.