The danger of reforming the Senate is that, like health-care reform before it, it comes to seem a partisan issue. It isn't. Members of both parties often take the fact that neither Democrats nor Republicans can govern effectively to mean they benefit from the filibuster half the time. In reality, the country loses the benefits of a working legislature all the time.
I strongly favor moderate-if-infrequent policy changes. It is not ideal - I find the compromised, moderate Senate health care bill highly objectionable, and of course the filibuster can be used for narrowly partisan purposes - but it is preferable to the alternative of ideologically polarized policy-making.
Ross Douthat is, and has something intelligent to say about it:
if you can't shrink the stimulus package much more substantially than the centrists have done, you shouldn't shrink it at all. There's a case to be made for a stimulus that's radically different than the one we have now; there's a case to be made for a stimulus that's like the one we have now, but a great deal smaller and more targeted; and there's a case to be made for a stimulus that's absolutely gargantuan. But thanks to the centrists, we're getting the cheapskate version of the gargantuan version: They've done absolutely nothing to widen the terms of debate about what should go into the bill, and they've shaved off just enough money to reduce its effectiveness if Paul Krugman is right - but not nearly enough to make it fiscally prudent if the stimulus skeptics are right.
My take on this is that the filibuster works fine for some things but not for others. There are two relevant distinctions: whether the bill is Frankensteinable or not, and whether it is necessary or not. I don't have sharp definitions of these concepts but here are some examples. The stimulus is a good example of a Frankensteinable bill: as Douthat says, any coherent bill -- whether left-wing or right-wing -- would have been preferable to the spineless and ineffectual thing we actually got. On the other hand, Supreme Court nominations are not Frankensteinable: a compromise appointee is an actual person, not a hodgepodge of attributes. Supreme Court appointments are also necessary, like (e.g.) the budget, which is Frankensteinable. On the other hand, a bill restricting cloning would be Frankensteinable -- a compromise could have perverse consequences that a straight-up right-wing bill would not -- but unnecessary. Finally, a bill creating a new national holiday would be neither Frankensteinable nor necessary.
The arguments against the filibuster are (1) that it frankensteins Frankensteinable bills, and (2) that it allows a unified opposition to block necessary legislation, thus paralyzing government. The obvious counter-argument is that without the filibuster there would be a great deal of sweeping and unnecessary legislation. In the narrow sense I prefer, necessity means both sides agree that the status quo is intolerable in the short-to-medium term but disagree on how to change it. (A healthcare bill is generally agreed to be necessary to deal with the Medicare crisis, but e.g. climate change isn't necessary by this def'n as a lot of Republicans believe that no action should be taken.)
I tend to believe in overdoing the checks and balances, so I'd like to keep the filibuster wherever possible. On the other hand, maybe what's needed is for the filibuster to come with a procedural price. Here's a possible set of rules:
- When party A proposes a (frankensteinable) bill, party B gets either to (a) introduce a competing bill and forfeit the right to filibuster or (b) keep the right to filibuster and forfeit the right to introduce a competing bill for, let's say, 12 years. Party B can still, of course, vote for Party A-sponsored bills.
- For all unfrankensteinable bills (e.g. appointments), the filibuster is usable on a tit-for-tat basis.
Obviously there are some definitions left to be filled in but this seems to be the right sort of approach.