In this post, I take to heart Eric Posner's admonition that the scope of executive power in wartime should be determined by the relative strengths and weaknesses of the presidency as an institution rather than by our evaluation of any one president. He is absolutely right that "the presidency is an institution that is occupied by a succession of persons, and the proper structure of this institution is independent of who happens to occupy it during a particular term." He is also right to emphasize that, although there have been many "mediocre" (or worse) presidents, this fact is balanced by the reality that there have also been many mediocre Supreme Court justices and members of Congress. Incompetence and mediocrity is rarely in short supply in any branch of government. I even agree with Posner's claim that the Bush administration has trampled on civil liberties far less than previous wartime administration's, such as Lincoln's, FDR's, and Lyndon Johnson's (a view also endorsed by prominent liberal civil liberties scholar Geoffrey Stone).

Nonetheless, Posner and coauthor Adrian Vermeule are wrong to draw from all this the conclusion that:

The case for giving emergency power to the president rather than Congress rests on the simple point that a multi-member body cannot act quickly, decisively, and secretly. Once we reject the assumption that the members of Congress are likely to be smarter than the president, I don't see how any other factor would play a role.

It is true that the executive can act more quickly, decisively, and secretly than Congress or the courts. It is not true that this is the only factor that matters, even in an emergency. The comparative executive advantages stressed by Posner and Vermeule are balanced by several comparative shortcomings. Relative to Congress and the courts, the executive is more likely to fall prey to irrational small-group decisionmaking, more likely to excessively restrict civil liberties, and more likely to fall prey to a short time horizon. Let's unpack these three flaws.

I. Irrational Small-Group Decisionmaking.

A great deal of social science literature shows that, other things equal - small, like-minded groups are more likely to fall prey to error than larger groups with more diverse perspectives. For a good summary of the evidence, see this book by Cass Sunstein. Relative to Congress, the executive is far more likely to fall into the hands of a small group of like-minded individuals. In most administrations, the key decisionmakers are the president himself and a small group of advisers most of whom tend to be adherents of the same party and ideology as he is. It is easy for such a group to fall prey to ideological "groupthink" or simply to persuade themselves that whatever is in their immediate political self-interest is also good for the country.

By contrast, Congress is a much larger and more ideologically diverse body than the executive. Even the members of the president's own party in Congress are likely to be a more diverse group than the top echelon of executive branch advisers. While there are undoubtedly some deluded ideologues in Congress, it is far more difficult for a small group of such people to seize control of the institution than it is for the same thing to happen with the presidency.

Even the Supreme Court, with only nine members, is less prone to this pathology than the executive. Because Supreme Court justices are appointed by different administrations and confirmed by different Senates over a long period of time, the composition of the Supreme Court at any given point in time is likely to be more diverse than that of the executive branch.

II. Incentives to Overrestrict Civil Liberties.

If the nation is hit by a terrorist attack or suffers a military defeat, the executive is far more likely to be blamed for not doing enough to prevent it than Congress or the Supreme Court. Most voters tend to assign the lion's share of responsibility for such setbacks to the president rather than to other branches of government. In principle, the executive is also likely to be blamed for excessive violations of civil liberties. However, in times of crisis, historical evidence strongly suggests that the average voter will care far about security against attack than about even quite flagrant civil liberties violations.

As a result, a politically rational president faced with possible tradeoffs between security liberty is likely to err in favor of former. It may well be politically rational for the president to sacrifice civil liberties for the sake of possible increases in security even in cases where the cost is very high and the benefits very small. This problem is not confined to the Bush administration. Indeed, as I noted in my last post, it was far worse under many previous wartime administrations. The most notorious example is FDR's order to intern the Japanese-Americans during World War II despite the near-total absence of evidence that they posed any real threat.

Congress and the courts are not immune to this pathology. But precisely because they are less likely to be blamed for any security setbacks, they have a comparative advantage in protecting civil liberties. I do not suggest that security should always be sacrificed for the sake of civil liberties. However, it is important that Congress and the courts serve as at least a partial check on presidential excesses in this field.

III. Short Time Horizons.

Presidents are subject to election every four years, and under the Twenty-Second Amemdment, cannot serve more than eight years. By contrast, Supreme Court justices serve for life (an average tenure of 26 years), and many congressional leaders also serve for many years. So to do many rank and file senators and congressmen. Even congressmen, who are up for reelection every two years, are rarely genuinely at risk of defeat this often, because most represent "safe" districts.

For this reason, presidents have much stronger incentives than either Congress or Supreme Court justices to sacrifice the longterm to short-run political expediency. This problem is particularly important during times of war or emergency, when opportunities to score short-term political points at the expense of the long run abound. The tendency to overrestrict civil liberties for the sake of minor or even nonexistent security gains is just one of many such temptations. As a partial (though by no means complete) solution to this problem, it is important that institutions with a relatively more longterm orientation serve as a check on those with a shorter time horizon.