What Can Presidential Speeches Do? A Dialogue

Q: So, Obama’s jobs speech was a barnburner.  Surely this will pull Obama’s approval ratings out of the doldrums, no?

A: Presidential speeches don’t really move the president’s job approval ratings.  See political scientist George Edwards’s book, On Deaf Ears, Table 2.2.

Q: Really?

A: Really.

Q: Okay, well, then surely the speech will make people support his jobs plan, right?

A: Presidential speeches don’t tend to persuade people on policy either.  Take the “Great Communicator,” Ronald Reagan.  In The Strategic President, George Edwards shows that Reagan could not move opinion on signature issues like aid to the contras.  And Reagan’s advocacy for increased defense spending was soon followed by a decrease in support for additional defense spending.  Public opinion on government spending often moves in the opposite direction as presidential preferences and government policy.

Q: So maybe one speech can’t make a difference, but surely a sustained campaign can.

A: Chances are, it won’t.  Even Reagan realized this.  In his own words:

Time and again, I would speak on television, to a joint session of Congress, or to other audiences about the problems in Central America, and I would hope that the outcome would be an outpouring of support from Americans… But the polls usually found that large numbers of Americans cared little or not at all about what happened in Central America…and, among those who did care, too few cared…to apply the kind of pressure I needed on Congress.

And you can also think about George W. Bush and the effort to reform Social Security.  Lots of the examples in Edwards’s book involve multiple speeches on the same topic.

Q: So forget public opinion then.  What’s really important is whether the speech helps get his policies passed.  So do speeches influence Congress?

A. The short answer: presidents don’t often succeed in persuading reluctant members of Congress to go along with their views.  Take Lyndon Johnson, supposedly a master manipulator of Congress.  Edwards shows that support in Congress for Johnson’s initiatives was not systematically higher than Kennedy’s or Carter’s.  For example, on crucial votes in the House, Johnson won the support of 68% of Democrats and 29% of Republicans.  Kennedy did better among Democrats (74%) and worse among Republicans (17%).  Carter did worse among Democrats (59%) but the same among Republicans (29%).

Q: So do presidential attempts at persuasion do anything?

A: What presidents can do, Edwards argues, is “facilitate” change in favorable environments.

Q: What does that mean?

A: Let’s work though an example from a different book, Who Leads Whom, by the political scientist Brandice Canes-Wrone.  But it’s going to get a little complicated.  Are you ready?

Q: Ready.

A: Canes-Wrone examined every nationally televised primetime presidential address between 1957 and 2000.  She excluded State of the Union Addresses to focus only on speeches that were at each president’s discretion.  She noted every appeal related to domestic policy and then scoured polling archives to see what the public thought about the president’s ideas.  She found 99 cases where there was a relevant polling question about that policy.

Q: That sounds time-consuming.

A: No doubt it was.  Here’s what she found: On average, presidents made appeals about policies that were already popular.  On average, 56 percent of the public supported the president’s policy.  Of course, presidents do at times advocate for policies that are less popular, but that’s more the exception than the rule.  So you can see how, as Edwards argues, presidents want to capitalize on a favorable environment.

Q: Great.  Does this apply to Obama’s jobs speech?

A: The polling data isn’t ironclad.  But there’s some evidence that more people support what Obama is proposing to do than oppose it.

Q: So back to this study.  Do these appeals actually get policies passed?

A: Canes-Wrone tries to answer that.  She looks at budgetary data, where it’s easy to determine what the president wanted—more spending, less spending, etc.—because presidents have to submit a budget.  Then she looked at what they got from Congress.  And she also knows from reading all the presidential speeches whether the president made a public appeal about a particular budget area.  So now she can figure out whether a presidential budget appeal affected how closely Congress’s budget bills corresponded to the president’s budget proposal.  There are 1,225 presidential proposals.  Presidents made public appeals on 79 of them.

Q: But wait.  Couldn’t it work the other way around?  Couldn’t the president decide to make a public appeal about something that he knew Congress already agreed with?

A: Canes-Wrone takes account of that in her statistical model.  I’ll spare you those details.  Here’s the bottom line.  A public appeal by a president is associated with an enacted appropriations that is 11-16 percentage points closer to the president’s proposal, relative to appropriations that did not receive an appeal.

Q: So the speeches mattered!

A: Yes, they did.  But one qualification: I’ve been discussing domestic budgetary items.  When the president makes an appeal about foreign budgetary items, the effects are much more muted—about 2 percentage points.

Q: But since Obama’s speech was about domestic policy, I’m still pretty excited.

A: As a social scientist, I am required to point out that Canes-Wrone’s study focuses on a particular type of policy—the budgetary and appropriations process.  Obama is not proposing something as part of the regular budgetary process but is asking for additional tax cuts and spending.  In other words, there’s always a question of whether a set of findings will generalize to any particular case, and especially to a somewhat different case.

Q: Enough.

A: And I’m also required to point out that because at least some of Obama’s proposals are already somewhat popular among the public and already have some support among Republicans, it’s not going to be easy to figure out whether the speech mattered, above and beyond those factors.  It’s hard to prove causation with a single case.


A: Okay, I’ll stop.  But let me point out one thing.  A lot of commentary about presidential rhetoric centers on the exact words that the president did or didn’t or should have used.  None of these studies proves anything about the content of rhetoric, merely its existence.  We really don’t have any evidence that how presidents frame issues actually affects public opinion or policy outcomes.

Q: Fair enough.  So what’s the ultimate takeaway?

A: First, the president’s ideas have at least some support already.  If they didn’t, he probably wouldn’t have given the speech.  Second, as the political scientist Jonathan Bernstein notes, the speech thus serves not so much to persuade lots of recalcitrant voters or members of Congress, but more to signal the President’s intention to push for these policies and, equally if not more important, to bargain about these policies.  In other words, the speech, whatever its tone, was not laying proposals that are set in stone.  Expect the speech simply to spawn additional debate and negotiation.

If the end result is a bill on the president’s desk, we’ll have an excellent example of how presidential leadership really works.  It’s not about magic words or eloquent moments.  As Edwards said, it’s about facilitating change.

22 Responses to What Can Presidential Speeches Do? A Dialogue

  1. paul g. September 9, 2011 at 2:03 pm #

    Also Jacobs and Shapiro book has insights on this question

    • John Sides September 9, 2011 at 3:13 pm #

      They have a good discussion of how Clinton’s efforts to sell his health care plan via a public campaign did not succeed.

  2. foosion September 9, 2011 at 2:09 pm #

    One claim about presidential speeches is that they help set the agenda. If the president is focused on jobs, then that’s a major part of the conversation. For example, Obama’s speech might get us talking about jobs rather than a sole focus on deficit reduction.

    Any support for this in the literature?

    • John Sides September 9, 2011 at 3:15 pm #

      Edwards is dubious, noting that presidents rarely focus their attention on a single policy initiative, that they must compete with messages from many other sources, that the media may not report their campaign faithfully, and that voters simply may not be paying attention.

      With respect to this, I wrote some more about SOTU addresses here: http://blogs.wsj.com/ideas-market/2011/01/25/the-state-of-the-union/. In short, the agenda-setting effects are probably short-lived.

      • Scott Monje September 9, 2011 at 5:17 pm #

        But this was far more focused than State of the Union addresses, which often cover more topics than people can even remember.

  3. Josh R. September 9, 2011 at 4:32 pm #

    A few questions/points:

    (1) You might want to add something explicit about veto bargaining, and the transmission of information about the President’s preferences to Congress, and how that affects policy that gets passed (cf. Cameron’s book on veto bargaining) – it’s kinda hinted at in the final paragraph, but Cameron’s book generally deserves attention.

    (2) I haven’t read the Edwards pieces, and I don’t think Canes-Wrones does this, but do they take into account failure to treat? In other words, there is an implicit model in which the President gives a speech, treatment X, which the public receives and that any change in the public’s support, be it for policy or President, can be thus traced to that speech. But, obviously, not all members of the public listen to the speech, e.g. there are significant numbers of individauls who are not treated. If a researcher restricts their attention to those who do listen, then does the presence of a speech prove important? This will obviously depend on the nature of individaul’s who continue to watch Presidential speeches in an era with both high levels of polarization and also high levels of media choice–if only partisans are turning in, then we should probably REALLY not expect much movement since the Republicans will counter-argue and the Democrats will already agree to a large degree.

    (3) What about activists and interest groups? In the above, presdiential rhetoric is taken as having an effect/no effect on “people,” e.g. the mass public or the mean level of support for a policy or a President, or on Congress, e.g. through passing laws. What about alternative measures – funds donated to the party or the President, increased likelihood to contact Congress or perform other actions, etc. Is there sound evidence that Presidential rhetoric also fail to move those efforts?

  4. Manoel Galdino September 9, 2011 at 4:54 pm #

    It seemed like those dialoges we find at Cosma Shalizi’s blog. But I missed more sarcarsm and irony.
    Nice job anyway…

  5. matt September 9, 2011 at 5:04 pm #

    Agenda-setting is extremely important the relationship between speeches and legislative success. Remember that Canes-Wrone’s theory assumes that presidents can increase issue salience, i.e., set the public’s agenda. That is, presidents speak on popular issues, which increase salience, and this is what converts the president’s rhetorical power into greater legislative success.

    Alas, only a limited body of literature looks at presidential leadership of the public’s agenda. In addition to Cohen’s work, Bryan Jones shows some evidence, as do Eshbaugh-Soha and Peake in a recent book. Even setting the agenda or increasing an issue’s salience is hardly guaranteed, of course.

  6. Andy Rudalevige September 9, 2011 at 8:09 pm #

    John is absolutely right on the gist of the literature. I would love to hear of a study that controls for the amount of work done by a president or his staff in conjunction with a given speech. Let’s say a speech is just a signal — “message: I care” — about whatever issue it happens to be, from a given line-item to a bigger piece of legislation. If so, it would probably change the way the president’s staff, bureaucracy (see Andrew Whitford and Jeff Yates’s book) and bargaining partners prioritize that item vis-a-vis others. My question is whether it is the speech that actually causing an item to pass that would not otherwise have passed, or whether it is the fact that the president cares about that item. Because he cares, it shows up in a speech; but because he cares, he works harder in negotiations to attain that item. Which gets the job done? As usual in presidency studies, I fear we run aground on the ability to measure “hard work” in this arena in any systematic way.

  7. Davis X. Machina September 9, 2011 at 9:22 pm #

    The correct line is to dismiss Edwards’ work out of hand because of the name of the institution in which he works — the George Bush School of Government and Public Service. Or because he reaches the wrong conclusions. And bullyfiresidearmpulpitchattwtisting. And Obama sold us out.

    This at least has been my experience throughout the left half of the blogosphere.

  8. Jeff Peake September 9, 2011 at 10:18 pm #

    Of course, you realize that the Department of Political Science at Texas A&M University isn’t part of the Bush School, so Edwards does not work for the Bush School.

    I’d say that Obama is in a good situation to impact the agenda (or the jobs issue’s salience among the media and public), but we need to keep in mind the entire reason for the speech is that the issue is very high on the list of problems facing Americans (according to the Most Important Problem question asked by Gallup). The speech may serve to focus media or congressional attention on the issue.

    Unemployment has been the highest among specific economic issues in this poll all year, and even after the debt debate in July/August, unemployment remained the highest in the mid-August poll.

    See: http://www.gallup.com/poll/149063/Americans-Satisfaction-National-Conditions-Dips.aspx (halfway down) and for the data all year: http://www.gallup.com/poll/148592/Important-Problem-PDF.aspx

    • Davis X. Machina September 10, 2011 at 7:46 pm #

      Of course, you realize that the Department of Political Science at Texas A&M University isn’t part of the Bush School, so Edwards does not work for the Bush School.

      Not precisely the claim I made, now is it?

  9. Scott September 9, 2011 at 10:49 pm #

    “None of these studies proves anything about the content of rhetoric, merely its existence. We really don’t have any evidence that how presidents frame issues actually affects public opinion or policy outcomes.”

    Really? “It appears, then, that what proponents were saying was more important than simply being able to send the loudest signal” (Jerit 2008, 15). Granted, the study is about elite rhetoric, not specifically presidential rhetoric, but it focuses primarily on elected officials and party members.

    Overall, the general message seems right – we don’t have enough evidence on the question and people tend to overestimate the power of the president – but let’s not swing too far in the other direction.

    • John Sides September 9, 2011 at 11:15 pm #

      Jerit (2008: 17): “proponent engagement was more effective at increasing support for reform than framing.” What this means is that attempts to engage and respond to the opponent’s framing were more important than the proponent’s attempts to frame the issue in their own preferred ways. This runs counter to the argument I was trying to rebut–namely, the argument of Westen and others that presidents can move public opinion by framing arguments themselves. And I think it doesn’t quite invalidate the thrust of my conclusion.

      But I’m perfectly happy to change “any evidence” to “much evidence.” Or how about this? “The power habitually attributed to presidential framing is far, far, far in excess of the available evidence.”

      (NB to other readers: The article is about rhetoric and public opinion toward the Clinton health care plan. A gated version is here: http://www.springerlink.com/content/92242q2656430278/)

      • Scott September 10, 2011 at 12:18 pm #

        To reiterate, I agree with your overall point – people vastly overestimate the power of presidential rhetoric. My point is that it is going too far to say that we don’t have any evidence that the content of rhetoric matters for public opinion, as shown by the Jerit paper.

  10. LFC September 9, 2011 at 11:35 pm #

    So presidential leadership is not about eloquent moments.

    Except maybe on certain occasions.

    “The mystic chords of memory…will yet swell the chorus of the Union, when again touched, as surely they will be, by the better angels of our nature.”

    “.. with malice toward none….”

    “…yesterday, December 7, 1941, a day which will live in infamy….”

  11. Emily Shaw September 10, 2011 at 7:07 pm #

    Too dissmissive by half. Even Edwards admits that the president gets a brief bounce out of public addresses. Consider the timing, too: we’re in the early stages of the Republican primary campaign, a.k.a. the start date for Obama’s second job. Campaign rhetoric is far more mobilizing, vibrant and simple than governing rhetoric. If the jobs address effectively rallies a base heavily demoralized by ineffective leadership through the debt ceiling debacle, then mission accomplished.

    In other words, look for “Pass This Bill!” to be the “Yes We Can!” of 2012.

    • John Sides September 11, 2011 at 1:12 pm #

      Emily: A “brief bounce” in what? And re: the demoralized base, I’ll only note that as of last week, 81% of liberal Democrats approved of the job Obama was doing as president. That’s lower than it was, but it’s lower in every subgroup — which is a likely consequence of the economy.

      That said, if by “base” you mean “liberal pundits,” then, yes, I see some evidence that they were pleased by the speech. And I also agree that Obama’s campaign rhetoric is likely to be more aggressive than his governing rhetoric.

  12. what women want from men March 11, 2013 at 10:13 am #

    Amazing issues here. I’m very happy to look your post. Thank you a lot and I am taking a look ahead to touch you. Will you kindly drop me a mail?

  13. number plates birmingham March 11, 2013 at 3:11 pm #

    great publish, very informative. I ponder why the other specialists
    of this sector don’t understand this. You must continue your writing. I am sure, you have a great readers’ base

  14. Jonas April 17, 2013 at 2:06 pm #

    I have recently become obsessed with the setting of presidential speeches. Being European, it strikes me as incredibly odd that there are so often people in the background when he’s giving a speech. What’s the motive for having an audience in the back??

  15. Penny July 16, 2013 at 6:56 am #

    I’m really enjoying the design and layout of your blog. It’s a very easy on the eyes
    which makes it much more pleasant for me to come here and visit more often.
    Did you hire out a designer to create your theme? Superb work!