Rethinking Presidential Primaries

The historic 2008 presidential election is now almost a year past, but my thoughts last week at a panel on presidential primaries hosted in Washington by the Brookings Institution returned to that election and a presidential candidate who never was - New York City Mayor Michael Bloomberg.

In 2007, before the presidential primary season officially got underway, a close political adviser to Mr. Bloomberg told me that Mr. Bloomberg was most likely to enter the race as a third-party alternative if the major parties nominated Democratic Sen. Hillary Clinton and Republican Gov. Mike Huckabee. The thinking was that the former Arkansas first lady and the former Arkansas governor would be the most polarizing candidates in their respective parties, fueling disaffection among independents and soft partisans that would pave a middle, split-the-difference path for the fiscally conservative, socially liberal New York mayor.

What Mr. Bloomberg may not realize is that his presidential aspirations, if indeed conditional on a Clinton-Huckabee matchup, may have been doomed from the outset by the nominating rules each party used in 2008. Those rules may change significantly by the 2012 election, but before turning to the future, the Bloomberg nonstarter scenario requires a bit of explaining.

Elaine Kamarck was the featured speaker at the Brookings event. An expert on presidential nominations, particularly on the Democratic side where she has been an adviser to nomination commissions and twice a so-called "superdelegate" to the Democratic National Convention, Ms. Kamarck discussed her new book, "Primary Politics."

During her remarks, she argued that last year the Democrats' proportional allocation rule for delegates prevented Mrs. Clinton from racking up delegates in the big primary states she won, including New York, New Jersey, Ohio and Pennsylvania. Had all state contests been winner-take-all, Kamarck estimates the former New York senator would have won the primary by about 300 delegates. Mr. Huckabee, she said, suffered the reverse curse: After splitting the Iowa caucus and New Hampshire primary with John McCain, he watched the Arizona senator parlay narrow wins in big, winner-take-all states like Florida, New Jersey and Virginia into a huge delegate lead.

Recent Democratic nomination contests masked the potentially prolonging effects of the current rules, which tend to allow non-frontrunner candidates to hang on longer. In 1996, incumbent president Bill Clinton was essentially unopposed. In 2000, Al Gore's advantage as incumbent vice president left little room for the challenge by Bill Bradley. And despite the crowded field of 2004 the desire to pick somebody - anybody - to take on George W. Bush compelled Democratic voters to fall quickly into line behind John Kerry after his Iowa and New Hampshire wins.

But last year's titanic clash between Barack Obama and Mrs. Clinton revealed what can happen when two equally matched candidates battle deep into the process. Operatives for the Obama campaign later claimed that the extended primary battle forced them to create a field apparatus in states like Indiana and North Carolina that, they believe, eventually helped them defeat Mr. McCain in those states. Still, many Democrats were rightly worried that a too-long primary fight was self-destructive for the party.

As for the GOP, Mr. McCain won the 2008 nomination despite thin support from the religious-social conservative wing of the party. He effectively cleared the field by early March but still managed to lose 19 of 50 states to either Mr. Huckabee or former Massachusetts Gov. Mitt Romney.

In a pleasant, even ironic turn of events from his defeat there eight years earlier, Mr. McCain won the 2008 South Carolina primary. But he did so with a mere 33 percent of the vote - because Mr. Huckabee and Tennessee's Fred Thompson split about 60 percent of the vote. The rules worked to Mr. McCain's advantage.

Ms. Kamarck advocates a mix of proportional-allocation states early in the process followed later by winner-take-all contests. She believes that would winnow out longshot candidates in the beginning, then force consolidation around the best candidate without unnecessarily prolonging the nomination contest. This sequencing could have the added advantage of discouraging every state from trying to move to the front of the calendar.

The rules mattered in 2008 and will again in 2012. If Mayor Bloomberg still harbors White House ambitions, maybe he should offer his own recommendations.

Thomas F. Schaller teaches political science at UMBC. His column appears regularly in The Baltimore Sun. His e-mail is [e-mail].