Paul Krugman wrote in last Friday’s New York Times:
Many people in the news media do claim, at least implicitly, to be experts at discerning character — and their judgments play a large, sometimes decisive role in our political life. The 2000 election would have ended in a chad-proof victory for Al Gore if many reporters hadn’t taken a dislike to Mr. Gore, while portraying Mr. Bush as an honest, likable guy. The 2004 election was largely decided by the image of Mr. Bush as a strong, effective leader. (Article now available by subscription only.)
Many people on both the left and right agree with Krugman’s causal hypothesis. They assume that journalists have some choice about how to portray the characters of politicians, and their choices affect voting decisions. It is because people buy this theory that they expend enormous energy looking for bias in the major news media and trying to influence mainstream coverage. A belief in the power of journalists’ implicit judgments raises the temperature; it encourages people to be highly critical readers who focus on the “spin” in news stories.
Of course there must be something to the theory. (And the 2000 election was so close that anything could have changed the outcome, including a rain storm or fewer earth tones in Al Gore’s wardrobe.) However, I believe the importance of journalists’ implicit character judgments is often overstated. Most surveys find that average voters are quite inattentive to the news, to start with. News coverage is always diverse, even when there seems to be an overall tilt like the one that Krugman detects in 2000. Moreover, the spirit of news coverage is not completely under the control of journalists. Al Gore, for example, had some potential influence on the way Al Gore was covered in 2000; this wasn’t simply a discretionary call by reporters.
Finally, we can almost always explain a presidential election as a result of economic indicators, leaving news coverage aside. Larry Bartels has argued that 2000, like most US national elections, was determined by the change in disposable per capita personal income (dpi) over the twelve months prior to the election. It follows that Al Gore would have won in 2000 if the Clinton administration had decided to cut taxes, thereby raising people’s dpi. Instead, they decided to pay down the national debt, thereby increasing the odds that Bush would win. If anything, it is surprising that Gore got more popular votes than Bush.
If I’m troubled by anything, it’s not anti-Gore (or pro-Bush) spin, but rather the way that our democratic system seems to reward borrowing and punish fiscal responsibility. By the way, Bush’s popularity until now may have a lot to do with the billions he has borrowed and spent.