The Constraints of Leadership

January 14, 2009

With a new President coming in next week, now seems like a good time to ask, “To what extent does the President get to choose his policies, and to what extent do the policies ‘choose him’, so to speak?”

The frame for my question is formed by this Econtalk podcast with political scientist Bruce Bueno de Mesquita. Mesquita’s argument runs something like this:

1. Politicians will only follow policies that are likely to keep them in power
2. The policies likely to keep a politician in power are a function of the desires of the people who put leaders in office, and these policy recommendations cannot be affected by the politician himself


3. Since a leader’s policies are not a function of his identity, the identity of a leader does not matter much.

In the United States, I’d argue that Bueno de Mesquita’s argument is much truer than most people think. If I wrote some of that “alternate history” fiction, it wouldn’t be very interesting, because in my fantasy United States where Gore gets elected President in 2000, everything is pretty much the same. September 11th still happens, we still go to war in Iraq, and the occupation still goes terribly.

Overestimation of the influence of the President is among the reasons people are going to be disappointed with Barack Obama. Bush, legend has it, opened the detention center in Guantanamo because pick your label: he is ruthless, stupid, lawless, unprincipled, a tyrant, etc. But if it is really just a personal choice of the President’s, why are we now hearing that closing the prison “may take a year“?

As I wrote before the election, Barack Obama’s Presidency will be a good libertarian experiment. Everyone agrees that Obama is a better man than Bush. We’ll see if that makes a difference when it comes to how he governs the country.

Still, it seems obvious that leaders do, at least in some places, make a difference. For example, how the heck did Atatürk so thoroughly change the course of Turkish history after WWI? Certainly not through popular support for his policies. Did Stalin really have to be as egregiously evil as he was? Couldn’t he have just been mildly evil and been just as effective? Why were the policies of Mao so different from those of Deng Xiaoping?

My brief answer would be that the amount of discretion a leader has seems to be a function of the proportion of the population that has a voice in putting/keeping him in power. The larger the proportion, the more constrained the leader. The smaller the proportion, the more able he is to do whatever he wants. That’s good news if you are hoping for change in North Korea or Iran, but bad news if you are hoping for change in the United States.


6 Responses to “The Constraints of Leadership”

  1. Rrrobert! Says:

    Dunno, man – you’re probably right that the president matters less than most people think (bureaucracies have their own momentum, and rightly so, so johnny-come-presidential can’t just change things at a whim). But I think you underestimate the extent of his (or his bureaucracies’) room for maneuvering within the broad confines of “the desires of the people who put leaders in office”.

    Remember that those desires are not monolithic – part of the job of the president is to negotiate the agenda for what gets done amongst the coalition that put him in office. And while the outlines of what the people want certainly constrain the universe of possible policies, choosing priorities from among the various options is a very individual endeavor, where different leaders can choose very different options and still get re-elected (re-election being their only constraint, they don’t have to optimize their popularity). This is more and more true the finer the resolution you look at – while almost any president might have bombed Afghanistan after 9/11, it’s unlikely that Al Gore would have weakened rules for shop-towel disposal in industrial facilities.

  2. Yeah, you’re right. The right answer to how much leaders matter is something like “Less than most people think, but more than I think.”

    One thing the (simplified) Bueno de Mesquita model doesn’t account for is the role of the leaders in building a coalition. The ability of Stalin, Hitler, and the East German leadership to stifle opposition and hinder cooperation among their enemies must have played a huge role in shaping their set of options as leaders.

    And yeah, there’s definitely wiggle room, especially if you can change the groups you’re trying to please. That said, I’d say you can switch sides more easily on domestic than foreign policy matters, because people want genuinely different things domestically, whereas public opinion about foreign policy is much more uniform (i.e. “keep us safe”).

    I guess I don’t really buy the argument that arresting and killing suspected terrorists isn’t making us safer. If we do it forcefully and indiscriminately enough, the effect of killing them ought to outweigh the effect of the negative feelings we’re creating. Lots of secondary effects, so it’s hard to say for sure, but it seems true to me (at the extreme, if we kill all non-Americans, we’d be safe from foreign threat). Anyway I’d say that for the President, thinking about the short term, the threat of another terrorist attack is a much bigger political risk than that Americans are seriously going to be upset that we’re not giving due process to foreigners.

  3. Dan Says:

    Obviously, stupid people overestimate the impact of the president. Still, there’s no way Al Gore would have gone to war in Iraq. There was tremendous resistance and it required a lot of political capital that Gore would have preferred to use on his pet issues (Kyoto?). Iraq was the pet issue for some key Bush advisers before September 11th (I’m not saying there was some conspiracy, they just had different priorities). Obviously events like September 11th and the current economic crisis come along and demand attention, but even there the response can be drastically different, and Iraq is a good example.

    There may be more overlap in priorities in foreign policy relative to domestic issues, but the executive has far more discretion, meaning you will probably get more distinct outcomes from different presidents. Also, people don’t care as much about foreign policy as domestic policy, giving politicians more room to maneuver.

    Bueno de Mesquita’s argument (or your synopsis of it) ignores the bully pulpit, the crushingly boring nature of policymaking, the blunt nature of public opinion, and the insulation of presidents. Presidents can change minds and they make a lot of low-profile decisions that people don’t care about, but that have profound consequences. Also, politicians work out the details, which may be the most important part. People may want to invade Afghanistan, but don’t care which provinces we start with. People may clamor for a stimulus, but don’t care how big it is or which sectors it targets. Furthermore, any term that a president is not expecting to be reelected throws this theory out the window, since there is no possibility of staying in power. No one is ever going to be impeached for policy in the US, and Bush had no problem governing with an absurdly low approval rating.

    So if Obama governs better than Bush, libertarians will admit they’re wrong and end this anti-partisan non-sense? I’m going to hold you to that. Really, the Bush administration was the perfect experiment for libertarians. Was there a notable difference in policy after we elected a narrow-minded ideological extremist?

    You can’t kill all the terrorists. It’s like a hydra, killing them generates resentment that inspires more terrorists. Even following your argument, I’m not sure a terrorist attack is that bad for a president, it certainly wasn’t for Bush. Guantanamo, Abu Ghraib, wire tapping, etc, however, had serious political consequences. Happily, I think there’s a middle ground between waterboarding the entire Middle East and giving all the terrorists government scholarships to flight school.

    People want change, but around the margins. No one is expecting a socialist revolution from Obama, just a more pragmatic and responsible way of doing policy, which isn’t out of the realm of possibility. Your last paragraph sounds like an endorsement of democracy. Careful, or they’re going to revoke your anarchist card.

  4. On the question of whether Gore would have invaded Iraq, I think you overestimate public opposition to the war. Opinion polls suggest that it had a pretty good degree of support, and would have had even more if we could have convinced the UN to go along with us. Gore would probably have done better on that than Bush, too, since the Europeans loved Clinton, and they thought Bush was an arrogant jerk. As long we’re allowing leaders to make a difference, there’s a big difference that I think Gore could have made. Maybe he could have gotten people riled up about the environment in 2002, but I doubt it (Sure, An Inconvenient Truth made a big splash, but Gore had been talking about the environment forever. The Simpsons made fun of him when he was V.P. for it.

    On the model, yes it leaves plenty of stuff out but I do want to emphasize that “public opinion” is not the same as the “winning coalition.” The average person focuses on outcomes and not about specific policies, but there is always somebody who strongly favors one policy over another (e.g. many people want “less pollution,” but one company really wants mandated catalytic converters in every car). Also, the term “being in power” is about more than just getting elected, you need a whole apparatus to support you, even if you are term-limited. I think we just disagree about whether the “bully pulpit” is important. I don’t think it is.

    By all means, if my political views are unchanged in four or eight years, please hit me over the head with some blunt object. My views have changed a lot in the last four years and if they don’t change again, I’ll be worried. Last September, I thought anarchism was absurd.

    That said, it’s hard to know exactly what to take as “evidence” that Obama is governing better than Bush. He can’t govern the years 2000-2008 again, so we’ve got the ol’ Mankiw/Silver identification problem. It’s hard to know what evidence to look at as proof that he’s governing better. Concretely, if he manages to spend $800 bn without an absolutely massive increase in lobbying and pork, I will renounce public choice theory.

    It’s hard to establish objectively meaningful standards by which to tell how well a person is governing. I guess my experiment isn’t so good after all.

    On my “endorsement” of democracy vis-a-vis dictatorship, to paraphrase an MR comment, a shotgun blast to the foot is preferable to a shotgun blast to the face. Saying that doesn’t mean I endorse shotgun blasts.

  5. Dan Says:

    Iraq War

    From the CBS poll cited in your Wikipedia article:

    Public opinion was 2 to 1 in favor of a diplomatic solution rather than military action. And that’s weeks before the invasion, after a year of the White House twisting evidence and drumming up support. Other Wikipedia polls show approval spiked after the State of the Union and Powell’s UN Speech, which speaks to the bully pulpit argument.

    Public support would have been higher with UN support, but would Gore have sought? And Europeans hate Bush, but they also hate US intervention and sovereignty violations. Clinton gave up on Rwanda, Kosovo, etc when he ran into a UN brick wall, and Gore would likely have done the same, if he ever got that far.

    The point is that 1) the public was open to an invasion, but not clamoring for it and 2) the president was able to push public opinion towards an invasion. Bush didn’t respond to a public outcry, he created it.


    Gore had been talking about the environment for decades, but he was never president before. Coupled with growing consensus on global warming in the early to mid 2000s, he could have done a lot. He probably would have pushed Kyoto or a new cap and trade treaty. At the very least, he wouldn’t have gutted the EPA.


    The examples of Obama not being able to get out of Guantanamo or Iraq quickly are misleading because they aren’t his policies, they are both cleaning up after Bush’s mistakes. And they are largely logistical challenges, not a lack of political will. He’s going to do both, I’m confident, but his failure to immediately put those genies back in the bottle isn’t a good example of policy constraints.

    I supported Nader in 2000 because I thought all politicians were the same, but Bush has set me straight.


    I see your point about my focusing too narrowly on public opinion rather than lobbying, parochial interests, campaign donations, narrow constituencies, etc. Still, different leaders are beholden to different interests partly by choice. Oil companies have more influence under Bush, labor unions will have more influence under Obama. They are both limited, but in ways that make a big difference in policy. And I think voters consider that. Also, Congress will generally fall in line behind a powerful president, especially from the same party.


    Approval ratings are a blunt but at least serviceable measure of governance. Those crowds! They’re so wise!


    I’m not talking about you, really, as much as all libertarians and hippies who think all politicians are the same.

    Shotgun Blasts

    Careful, or the NRA is going to shut this blog down.

  6. […] which is basically the same thing. But I think we did a decent point/counterpoint on this back in January. Posted by William Bruntrager Filed in Uncategorized Leave a Comment […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: