Bias, What Bias?

An interesting study from a couple of fellows from the AEI:

Economists have been puzzled this year by the persistence with which perceptions about the economy have lagged the economic data. For the most recent 12-month period for which we have data, for example, the economy grew almost exactly as fast as it did during the best 12-month period during President Clinton's two terms. But the economic mood of the country has been much different.
It isn't just the economy that influences people's perceptions. In research we just released, we find that media coverage is also an important determinant. We found that newspaper headlines reporting economic news on unemployment, gross domestic product (GDP), retail sales, and durable goods tended to be much more frequently negative when a Republican was in the White House. And this was true even after accounting for the economic numbers on which the stories were based and how those numbers were changing over time.
We also found that positive headlines explained whether people thought that the economy was getting better more than the economic variables themselves. Newspapers are indeed important.

...Actual economic data explains much about the headlines - but far from everything. We found that the incidence of positive coverage during Republican presidencies was fairly steady - but economic news under President Clinton received by far the most positive coverage. This partisan gap or bias (the difference in positive headlines between Republicans and Democrats for the same underlying economic news) consistently implied that Democrats got between 10 and 20 percentage points more positive headlines.
We also examined individual newspapers. Among the top 10 papers, we found strong evidence that the Associated Press, the Chicago Tribune, the New York Times, and the Washington Post were much more likely to have positive headlines for Democrats even with the same economic news. The New York Post showed no statistically significant difference. The Los Angeles Times did not tend to treat Republicans and Democrats significantly differently.

I'm shocked, shocked I tell you. Here's the full report.