My Main Independent Variable...Whoopee!

So my dissertation project relies on a survey that people have been using for years. But what do you guys make of study with the following features:

1) Leaves out crucial data in the published data set, making replication impossible
2) Results change when statistical method changes
3) Leads to implausible conclusions
4) Doesn't correlate with other, similar data
5) Uses retrospective self-reporting
6) Plugs in the mean to fill in missing data

Sounds great, doesn't it? I know I'm thrilled about it!

What New Ideas?

From Matt Bai:

What conservatives did quite well in the era after Goldwater was to apply those principles to the emerging challenges of the moment: deindustrialization, anxiety among the white middle class, failing schools and communities, uncertainty in the world. Their argument was that intrusive government had contributed to all of these problems by devaluing individual responsibility and throttling free enterprise, and that it had made the country less safe by declining to stand up for those same values abroad. There followed from this a series of new reforms that made up the modern conservative agenda: supply-side economics, school choice, workfare, missile defense, etc. Conservatives applied enduring principles to a modern argument, and to deny them that is to risk badly undervaluing the role of an argument in building sustainable majorities.

Excuse me, but I fail to see where the "new ideas" are in the conservative agenda. On domestic policy, the only ideas conservatives promote are deregulation, tax cuts, and privatization, all of which just happen to shift resources to corporations. Schools are bad? Break the teachers unions and privatize the schools - giving public money to private institutions. Health care a problem? Create health insurance markets that subsidize payments to health insurance companies. Urban poverty? Eliminate regulations and give tax cuts to businesses. No matter what the problem, their solution is always to repeal another part of the New Deal social contract that emerged in the postwar era. Nothing new to see here....

The conservative position on foreign policy is somewhat more muddled, but in the main it appears to be maintain and extend U.S. hegemony, subverting any governments that pose a potential threat, and establishing client states in those regions of strategic or economic value. This is called imperialism, and let me tell you, there is no less new idea in the world.

Oh, and the new conservative ideas with respect to the Constitution? Repeal 20th century jurisprudence and return us to the days of the Lochner Court. Concentrate power in the executive branch. Reduce civil liberties. Yeah, lot's new here.

In sum, while it may be the case that liberals have been trapped in a New Deal/Great Society mindset (and I don't actually agree that we have), the force of Bai's argument loses most of its power if we acknowledge that his supposed model of innovative approaches to government - the conservative movement - is in fact more inflexible and boring than liberalism.

Before I Forget

A special shout-out to Zola, who just had a little baby boy. Well, actually it was his wife who had the child, biology being what it is. May he always cherish the memory of a good night's sleep!

Evening Report

Scotch good. Work bad.

That is all.

Repairing the Presidential Nominating System

This is the last one of these, I promise!

As usual, there is no way to make everybody happy. Obviously New Hampshire and Iowa are going to be displeased at any changes, but oh well. Also, to solve what I see are more pressing problems, it is probably necessary to minimize the role that promoting "outsider candidates" has in fashioning a workable primary system. After all, only two such candidates have ever secured the nomination (McGovern and Carter), and that was thirty years ago. While I do think we should limit the ability of media-dubbed "front runners" to dominate the process, I think that the price of maximizing the potential for an underdog candidates comes at too high a price. To do so we would have to maintain the sequential nature of the contest, which is what has created so many other difficulties.

So here's my plan. Despite grave misgivings, I am going to reverse a previously held opinion: I think something like a national primary is probably the fairest system. It would eliminate the bias towards any one region of the country or demographic group, and would also make for a much shorter and more sensible campaign. The national primary could take place in May or June, with the Conventions in July and August. Ta-dah! A short campaign. No more campaigning in April of the previous year (thank god).

To make a national primary workable, we would also need to make some other changes, otherwise big-money and big-media candidates would own the process, and people like David Broder would have too much influence. First, I would consider abandoning primaries in favor of the caucuses, since the latter rely more on activism (I also like the idea of parties picking their own candidates. Call me crazy). More importantly, I think the Convention would need to revive the old 2/3 rule requiring that a candidate needed an effective consensus rather than a simple majority. This would make sure that no candidate could cruise to the nomination without any trouble. To reduce the influence of money over the process, a strict system of public financing should be established in which candidates who "opt out" (like everybody is doing now) will be ineligible to have their names placed in nomination at the Convention. Finally, there should be a (very) few candidate debates in prime time on every network - just like the general election debates. This would give less well-known candidates a chance for exposure.

This is not a perfect system, by any means. The role of the media might even be enhanced, and it could benefit incumbents if the out-party had a messy convention. But on balance I think a national caucus with the tweaks I mentioned would do a great deal to make the system fairer, shorter, and more interesting. It would primarily benefit those candidates capable of generating enthusiasm and support across a broad stretch of the nation, and well as those generally liked by the entire party. If we combined a Caucus system (rather than primaries) with a more effective public financing regime, it would make place a premium on a candidate's ability to build an organization capable of competing across the country - which would pay considerable dividends in a general election.

The End.

Screwing Things Up On Purpose

Look, one shouldn't think that a rising crime rate is an accident. Yes of course part of it is due to a weakening economy, but a another cause is the deliberate effort by Republicans to weaken successful anti-crime programs. Why would they want to something so nutty? Because if the crime rate goes up, they can try to scare white suburbanites about the "scary brown people" again. They've been very grumpy about losing the "crime issue", you see......

On Vacation

Brazen Hussy and I are taking a long weekend to go camping in State Next Door. See you guys later!

Does Everybody But Me Know About This?

I didn't hear about the "Jena 6" case until the Barack Obama-Jesse Jackson flap. It's a pretty outrageous case, and what's almost as outrageous is that I didn't know about. I can't decide whether it's the fault of the media for not telling me or my own fault for not noticing.

From the little I've found, a fight broke out after a bunch of racial incidents in a small southern town. Six black boys are being tried for attempted murder for getting into a fistfight with a white kid. What the hell century is this??

Richard Posner Should Be Impeached

From the Washington Post:

Judge Posner's majority opinion declared the benefits of voting "elusive," given the small chance that one could cast a decisive ballot in an election, and he analogized voter fraud to the crime of "littering."

This is typical sophistic rigamarole. The value of voting rights doesn't rest with the probability that one's vote will alter an electoral outcome. To say that a right can be violated if there's no likelihood it will be exercised would be like saying that we could pass laws saying a poor person can't run for office because there's only a small chance that they will win. Would Posner uphold such a law despite its obvious unconstitutionality? Furthermore, it matters a great deal in the aggregate if a large number of people are being systematically discriminated against, which not only deprives a group of liberty but undermines the democracy itself.

Not that I expect Posner to care much about democracy...

Evaluating Proposals for Reform (updated)

(This damned post has been sitting in my "draft" folder for days.....)

This is my next piece on the presidential primary process.

Okey dokey. Here are the basic reform plans:

1) The rotating regional primary plan.
This comes in a couple of different varieties. One divides the country into the four regions (South, Northeast, West, Midwest) and alternates which primary goes first. A more sophisticated version divides the country into subregions - probably New England, the Middle Atlantic, the Southeast, the Middle/New South, Great Lakes, Farmbelt, Southwest, Rockies, and Pacific coast. But you get the idea. This plan would restore some order to the present system, consolidating and rationalizing it somewhat, without making any substantial changes.
My problem with this plan is that it does nothing to change the current system, really. It does end one kind of bias (that of IA and NH), but replaces it with other sorts of bias. "Big" candidates would be advantaged by the larger playing field of the regional primary. But the big beef I have with it is the periodic bias towards candidates from certain regions of the country. For example, if the South had the first primary in 2008, we'd almost certainly have a southern nominee. If the Northeast, a Northeasterner. In the same sense, a certain region's voters would get to choose the nominee, diminishing the influence of the rest. How fair is that?
It turns out that the bill under consideration in the Senate actually leaves IA and NH before any of the regional primaries. Which means this plan would resolve precisely nothing. Boo!

2) The small-state plan
The states are clustered by population size (and dispersed geographically into different regions), voting week by week until the big states vote at the end. This is a lot like the system worked in the 1970's, when candidates gradually built up momentum before the big showdowns in states like California and Ohio.
Probably the best of this set of proposals, it would effectively exclude the big states from having a meaningful say (which is why they started front-loading in the first place), and since small states tend to be rural and white, would still under-represent minorities and people living in urban areas. But it would help outsider candidates have a chance.

3) The competitive state plan
Under this approach, the states would be ordered based on their closeness in the last election (or the last few elections). The idea is that if the purpose of nomination is to select the candidate with the best chance of winning the general election, then the states that are most important in the general should also be most important in the primaries. I have a number of objections to this strategy. First, the primary voters in a swing state might be very different from the general election voters, so it might not work. Second, it would (funnily enough) still privilege IA and NH - two of the closest states in the nation. Third, I think Florida and Ohio have enough influence right now, thank you very much.

4) The national primary
Every state votes on the same day. This plan would mute the influence of any one region by eliminating the sequential character of the primary process, which is what creates it. Which is good. It would also magnify the influence of the press & wealthy contributors and effectively eliminate "outsider" candidates from the process (they couldn't compete on that scale). Which is bad. It also might lead to a brokered convention, which would be fun to watch but would also hurt the party in the general election - brokered conventions are not pretty.

5) Scrap primaries entirely.
Nobody has seriously considered this approach, but it's possible to say that primaries won't pick delegates to convention, turning them into "beauty contests." It would be a return to the days before the McGovern reforms. Now on the one hand this would make for interesting nomination contests. On the other hand, I wonder how we would pick delegates. By caucus? I'm afraid that these would, like Iowa, become just like primaries, meaning you would have solved nothing. The "outsider" candidates would probably be excluded, unless they won a bunch of the beauty contests. I also don't think that anybody would really go for this plan, since it looks anti-democratic.

Of course there are many permutations and possible combinations of each of these plans, but as you can see none of them solves every problem. One of them (the regional primary plan) solves almost nothing, so it should probably be excluded from consideration. Unfortunately it's the one that's gotten the most consideration. Predictable.

Next time - my plan for reform.

Finally. Progress

I am still a bit burned out, but I worked through the whole weekend and finished a chapter for my dissertation. This is the first one I've managed to complete, so it's pretty exciting. It would be even more exciting if I didn't have four more of them to write.

Shoddy Reporting. Again!

The AP has reported that artic sea could completely melt during the summer by the year 2070. An article I thought was strange, because I remembered hearing that the year was 2030 - four decades earlier. So I tracked down this piece which says that 2070 is the old, optimistic estimate. If present rates continue, artic ice will melt in 2030. Which means the AP is feeding us a load of manure again.

Color Me Unimpressed

Alan Greenspan criticizing George Bush for unsound economic policies is liking Stalin reprimanding Mao for being a bloodthirsty tyrant.

(Rate of typos per sentence= 1:1)

Bush's Speech

Didn't see it. Don't care.

We all know Bush is just trying to run out the clock so he can hand off the Iraq War to the next President and then blame them for the mess. The Democrats won't cut funding because they're afraid they'll be blamed for the aftermath, and because Bush might just leave the troops there to get shot to pieces. And my advice is unchanged as well: since the Democrats in the Congress are unwilling to attack the Iraq War directly, they should do so indirectly but zeroing out funding of the White House. Let Bush get through the winter without heat and see how he likes it.

P.S. I have been so busy with work and my dissertation that I haven't yet been able to help the local City Council candidate campaign. I feel guilty. :(

Another Reason to Not Like Dynasties

I am not for or against Hillary Clinton, but situations like this exhibit why I believe dynasty politics is a problem.

People are evaluating Hillary Clinton based on her husband's performance in the White House. Some people are supporting Hillary because of her husband, and some are opposing her because of her husband. But on election day Bill Clinton's name won't be on the ballot - Hillary's will. We should be evaluating her candidacy based on her unique mix of positions and personal qualities, not those of her husband.

In the similar way, many voters projected their memories of Bush Sr. on Bush Jr., helping the clearly unfit W become a viable candidate for the White House. Voters may like dynasties because they make voting decisions easy, but in democracies we aren't supposed to be interested in what's easy, but what's right.

What Do We Want From A Nomination System?

The arguments over what is wrong with the way we nominate candidates for Presidents tend to reveal the unspoken beliefs about how we should nominate them. There tends to be different groups who have very different notions about what features a nominations system should have. A quick list:

1) The parties should nominate the candidate who has the best chance to win the election

2) The primary contest should minimize internal party divisions to maximize the chance of victory

3) The voters should be able to pick the nominees.

4) All candidates should have an equal chance at the nomination

5) The campaign should be interesting.

6) The campaign shouldn't be too long

7) Oh, and the nomination system should pick qualified candidates.

Let me discuss the consequences of each of the above principles one at a time.

1) How does one know that a candidate has the best chance? Oftentimes this is not clear until the candidates is in the midst of the general election. Just because someone is polling well know doesn't mean that they are the best candidate - they might just be better known.

2) This principle will push for a quick, front-loaded process that mitigates against 3-5. It is basically what we have now - short and sweet nomination contests, but that favor certain regions of the countries, well-known candidates, and makes the process boring.

3) This is of course a very slippery notion - which voters? Party activists or the general mass of voters? Should all voters everywhere have an equal say? And should every group of voters have an opportunity to make their preferences known? The latter two would push us towards something like a national primary.

4) This principle generates another big set of problems. If we have a sequential contest, then candidates from certain regions are favored (why do you think New Englanders do so well? New Hampshire, that's why). If we have a national primary, then less well-known candidates are favored.

5) An "interesting" campaign would have to be very long, and would divide both parties as they experience prolonged internal feuding. This would also generally favor incumbents, since usually their party is going to be more unified.

6) Alternatively, a short campaign means frontloading and establishment candidates, and will be "boring."

7) Who decides what a qualified candidate is? Well, we can certainly argue that the present system doesn't favor this criteria, since the chief assets a candidate can have is fundraising ability and a good media personality. On paper, Chris Dodd, Bill Richardson, and Joe Biden should be front-runners. Obama and Edwards should be out of it. Thompson, Giuliani, and Romney should be also-rans to Huckabee, McCain, and Brownback. Sort of the opposite of what we have now, isn't it? Frankly I'm not sure how one even gets this criteria considered.

So as you can see, there isn't going to be a perfect system, because there are so many competing principles at work. We're going to have to make a decision as to which principles are most important, and try to reduce the negative effects on the others. And we can also determine whether any of the many proposals out there meaningfully address these concerns, or does too much to privilege one over the others.

Next time: Evaluating Proposals for Reform.

What I'm Not Writing About

I'm not writing about the anniversary of 9/11, because it's my blog and I don't have to.

I'm not writing about the General Petreus testimony, because it's nonsense. I'd use the phrase "political theater," or "carnival or fallacies and sophistry" but that would constitute writing about it.

I'm not writing about my dissertation, because who wants to hear about that?

What I will write about is how much I love you guys. I asked for advice, and I got a lot of it - good advice too. Thanks for all the support, and don't think I won't take you up on your offers of more of it.

Unlocking Old Doors

Some time ago I gave up the notion of teaching. Not just because I was burnt out from being an adjunct, but because I couldn't imagine pursuing a tenure-track job. Doing so not only seemed ludicrous (2 tenured jobs at one college? puh-leeze) but just painful. I really don't want any part of the 3-5 publications/year, university-service grind. Intellectually I'd given up on any notion of teaching college, a decision which was more depressing than liberating. The PhD seemed like a silly extravagance, and for perhaps the first time in my life I had no idea what I wanted to do or be. To make matters worse, I was trying to accustom myself to seeking a tedious 9-5 job whose sole purpose would be to pay the bills. All and all a bleak prospect.

And then my wonderful spouse Dr. Brazen Hussy suggested that I look into becoming a full-time instructor. Apparently there are some universities that pay people to just teach. What a strange concept! I was a little shocked at the notion, so sure had I been that I'd never set foot in a classroom again. Now I'm wondering whether I'd just decided I didn't want to teach, whether I'd just persuaded myself, rather than feeling that I didn't want to teach. Now I'm remembering how energized I used to be in a classroom, how I did get something out of that experience. Maybe teaching and adjuncting are not the same thing - I suppose I just assumed that they were.

It's a tempting thought - that I can do something for a living that has some inherent rewards. But I wonder if I'm falling prey to the "grass is greener" syndrome, or if the life of a full-time instructor really is no different than an adjunct, or what. What do you guys think?

The Longest Day In the History of the Universe

Brazen's always accusing me of exaggerating, but I tell you, no day in the history of days has ever been as snail-like as this one. I swear I've been lapped by moss growing at least twice.

To make matters worse, tonight I have to go to an office party at a McMansion. Said party will likely involve upwards of 20 screaming children. Maybe I can convince Brazen to let me borrow her flask & fill it with scotch or something......

What's Wrong With Our Nomination System

As promised, here's a review of the major criticisms that have been made against our system for nominating Presidential candidates:

1) It's unrepresentative. New Hampshire and Iowa don't have any cities or minorities, so the candidates who are nominated pay little attention to those issues.

2) Frontloading biases things towards establishment candidates. The movement of big primaries closer and closer tends to give candidates with lots of money and endorsements an unfair advantage, excluding "outsider candidates."

3) The nominations are wrapped up too early. Why wants an 8-month general election campaign? Plus, we should spend more time talking about issues within the party and vetting each of the candidates more thoroughly. We shouldn't rush into picking a nominee.

4) The media has usurped too much more in the "winnowing" process. By establishing first and second tiers and setting expectations, the press corps can make it difficult for candidates to raise money or win votes by giving them a "loser" image. Plus the media is obsessed with stupid stories rather than policy.

5) The system has gotten so expensive that the importance of wealthy donors has been magnified.

6) Too few people in too many states have next to no influence over the process. The nominations are always wrapped up before most people get the chance to vote.

7) Party activists and the party leadership has too little influence over who the nominee is. Too much power has flowed to other groups (see above). Parties have the right to choose their own candidates.

If I can think of any others I'll add them later, but those are the main critiques. It's a pretty persuasive attack, although it can be segmented into a couple of rival arguments: that Iowa and New Hampshire have too much influence, that frontloading is bad, and that the primary system overall is a flawed model.

Next up: what we should want from a nomination system.

Dynasty Watch

And now for a double edition of "people running for office whose principle qualification is their last name."

1) Niki Tsongas, widow of former Senator Paul Tsongas, won a special election primary for a U.S. House seat in Massachusetts yesterday. (via Swing State Project)

2) Mike McWherter, son of the former Governor, is the likely Democratic nominee for the U.S. Senate should he decide to run. So much so that other people have said they'll drop out if he does. (via Senate Guru)

Yay democracy! :(

A Short History of the Nomination Process

Hope everybody had a nice holiday!

So I wanted to start my little series on the Presidential nomination crisis by describing the history of the process. We've gone through 4 basic systems for nominating Presidents.

1) The Congressional Caucus. The party's congressional delegation in Washington nominated its candidates for President. This system lasted from 1796 until 1824.

2) The Pure Convention System. After a number of experiments (like having state legislatures nominate candidates), the National Nominating Conventions were adopted by the major parties starting in 1832 with the Democrats. Martin Van Buren borrowed the idea from the Anti-Masonic Party. Convention delegates were chosen by the state parties, decentralizing political control away from Washington. Those state delegations were usually chosen by a party caucus or convention in each state. The old convention system tended to be dominated by "party bosses," but in many respects was a more open process than is given credit for. It lasted in its unadulterated form until around 1900. The Democrats added an extra wrinkle by requiring that 2/3 of the delegates were needed to select a nominee, which gave the South a veto. Today the convention system is used to select candidates in virtually every other western democracy.

3) The Mixed Convention/Primary System. Frustrated with the power of party machines, Progressive reformers created the presidential preference primary. Now it's important to note that the U.S. is the only country that uses primaries. My hypothesis for this is that because of the regional cleavages caused by the civil war, most people lived in what were effectively one-party regions. So they only way to create any political accountability was to create a formalized system of intra-party competition, giving voters the power to select nominees. At the Presidential level, a handful of about 15-20 states adopted presidential preference primaries. Most of these were "beauty contests" that brought no delegates. This system lasted from around 1900 until 1968. The 2/3 rule was disposed of in 1940 so that FDR could get nominated for a 3rd term. Under these reforms the voters had a say, but at the end of the day the party leadership (or bosses, if you prefer) still picked the nominees.

4) The Primary System. In 1968 a lot of Democrats were ticked off that Hubert Humphrey, who supported LBJ's policy in Vietnam, was selected as the nominee even though he'd never entered a single primary. So the party decided to revamp the nomination system yet again by requiring (among other things) the winners of primaries would have to get the delegates in that state. The result was a proliferation of primaries, and a strengthening of the role of the media in selecting nominees (although Kennedy was the first real "media candidate.") It was at this time that the party leadership lost effective control of the process and the modern system was created.

The system has remained in flux on the Democratic side, however. At first the primaries were winner-take-all, but then were modified so that they were allocated proportionately by congressional district (so if a candidate won 40% of the vote in the state but only 25% in a given district, he'd get 25% of the delegates in that district). The Democrats also added in what are called Super-Delegates (mainly elected officials) to restore some power to the party leadership, but that reform has never made any discernible difference.

The key element to understanding the primary system is that it is sequential: the elections happen across time as well as space. Candidates that lose early contests lose credibility in the eyes of the press, contributors, and voters, so the early contests are the most important. Ever since the nomination process was first used in 1972, the early primaries have been the most important. In 1972 McGovern won the Iowa Caucuses and came in a close second in New Hampshire; Carter won both New Hampshire and Iowa in 1976 and 1980; Mondale won Iowa and lost NH, setting up the Hart-Mondale duel in 1984; Dukakis won NH in 1988; Gore & Kerry both won IA and NH in 2000 & 2004. The sole exception is Bill Clinton, who did not compete in Iowa and came in 2nd in NH, but he won the crucial Georgia primary a little bit later.

Because the early states winnow the field down down to 2 candidates, every state has an incentive to be one of those early states. Iowa and New Hampshire have enjoyed a privileged position in this regard, which they have guarded ferociously. Over time more and more states have moved earlier and earlier. NH used to vote in March - not it votes in January. Most of the big states used to vote in April through June, and now they vote in February and March. This process of front-loading has been underway for some time now, and neither party has been very effective in trying to stop it. So far we've avoided moving into the year previous to the election, but if things keep going on this way it's bound to happen. If not this cycle, it'll happen next time.

Next time: What's wrong with the system we have now?