Ezra Klein has a piece in the latest New Yorker on the ineffectiveness of Presidential speeches. Like most of what Ezra writes, it’s great stuff, and I encourage you to read it. (Go on, you can do it!)
As is typical of DC journalists, however, the piece also displays a strikingly poor appreciation of context. To be more specific, the piece fails to consider the significant influence that Presidential rhetoric can have—and has had—on the country’s political discourse.
Before I explain, a few quotes to give you a flavor of the article:
The annual State of the Union address offers the clearest example of the misconception [that Presidential speeches are effective]. The best speechwriters are put on the task. The biggest policy announcements are saved for it. The speech is carried on all the major networks, and Americans have traditionally considered watching it to be something of a civic duty. And yet Gallup, after reviewing polls dating back to 1978, concluded that “these speeches rarely affect a president’s public standing in a meaningful way, despite the amount of attention they receive.” Obama’s 2012 [state of the union] address fit the pattern. His approval rating was forty-six per cent on the day of the speech, and forty-seven per cent a week later…
If speeches don’t make a difference, what does? Another look at the Presidencies of Franklin Roosevelt and Ronald Reagan offers an answer. Roosevelt was one of only two Presidents in the twentieth century whose parties won seats in a midterm election. That was in 1934—a year in which the economy grew by ten per cent. But in the midterms of 1938, the year after the economy plunged into a double-dip recession, the Democrats lost seventy-two seats in the House. If Roosevelt had been running for reelection, he, too, would almost certainly have lost…
Presidents win victories because ordinary Americans feel that their lives are going well, and we call those Presidents great communicators, because their public persona is the part of them we know.
Ezra draws mainly on the work of George Edwards, a political scientist at Texas A&M, in making these conclusions. Edwards has made a name for himself by demonstrating that Presidential speeches have no positive impact on a President’s poll numbers. There’s more to the article, but that’s the gist of it.
So what’s the problem? Well, poll numbers are not politics. It goes without saying that the economy is the major factor in determining a President’s popularity. But gauging whether Presidential rhetoric is effective by tracking short-term shifts in approval numbers vastly oversimplifies a complex phenomenon.
I’ve looked at Edwards’ evidence, and it is spot-on: a President’s approval rating does not go up following major speeches. But I’ve also studied history and know enough to say that rhetoric does matter. Even though most of us who write for a living tend to pay too much attention to what people say, there is no question that Presidential speeches can alter political discourse…maybe not so much in the short-term but definitely over the long-haul. FDR’s rhetoric, for example, helped legitimize an expanded role for government in our lives. Moreover, even if it is unpopular with voters, political rhetoric can influence the media and subsequent understandings of a presidential legacy, as was the case with Reagan. (Video clips of him saying “tear down this wall” have done more to shape understandings of the end of the Cold War than hundreds of books.)
Of course, there is no way to clearly measure the precise influence of Presidential rhetoric. And that, to me, is the main lesson that one should take from the work of Edwards and others. Like economists, political scientists have a tendency to valorize numbers…and to dismiss as unimportant anything that can't be quantified. We’ve seen what a disaster exclusively quantitative approaches to economics have had. Let’s not make the same mistake when it comes to politics.