HOW NOT TO FIX THE PRIMARIES

I'm not a fan of America's presidential nominating process, but David Leonhart's assessment of the process is flawed.

It has come to resemble a reality television show, in which a pseudo-scientific process (polls plus donor numbers) winnows the field. The winner is then chosen by a distorted series of primaries and caucuses: The same few states always get outsize influence, and a crude, unranked voting system can produce a nominee whom most people don’t want.

No wonder the current president is a reality-television star, not to mention the most unfit occupant of the office in our country’s history. “The victory of Donald Trump in 2016 is best understood as a failure of the process,” the political scientist Jonathan Bernstein has written, “and a failure of the Republican Party to prevent an outsider from taking its presidential nomination — the most important thing that U.S. political parties have.”

But Trump didn't win the nomination because early states have outsize influence -- he went on to win states in the North, South, East, and West. He won rural states and urban states. And whatever you may think of the system that produced him, he turned out to be a nominee most Republicans, at least, were very glad to have -- he's extraordinarily popular within his party, he won the presidency for the party after two straight losses, and he might win again. It pains me to say it, but this is a democracy, so I'd contend that the system worked. The real crisis is whatever happened to GOP voters over the past couple of decades to make the notion of a Trump presidency so appealing to them.

In other democracies, political parties have more sway in selecting the nominee, and voters then choose among the major nominees....

When voters are given the dominant role in choosing a nominee — as with primaries here — only an unrepresentative subset tends to participate, which skews the process. Party leaders, on the other hand, have a big incentive to choose a broadly liked candidate.

But in the recent British elections, the candidates for prime minister were chosen by the parties -- and the result was the same as in the U.S. in 2016: a broadly disliked candidate from the right defeated a broadly disliked candidate from the left.

The first change should be to the debates. The candidates’ electoral history and qualifications currently count for nothing. The 2020 Democratic field, for example, has included four two-term governors, all of whom have been excluded from debates despite a track record of winning votes and governing successfully. In their place have been candidates, like [Andrew] Yang, who managed to crack 4 percent in a few polls.

It makes more sense for only the true polling leaders to be guaranteed debate slots. Beyond them, the party could set aside at least one spot for a governor and perhaps one for a senator from a large state or swing state.

So there should have been quotas for participation in the debates, all because two uncharismatic moderates (Steve Bullock and John Hickenlooper), an uncharismatic liberal (Jay Inslee), and a late entrant no one seems to want in the race apart from a few plutocrats (Deval Patrick) didn't get enough screen time? (All except Patrick participated in at least one debate.)

It's bad that the last round of debates featured no candidate who was black or Hispanic, but at least the polling and fundraising criteria suggested a popular mandate. How would it have looked if a higher-polling non-white candidate failed to make a debate while a lower-polling white male got in thanks to a quota?

I think this is a good recommendation:

It’s also past time to end the special treatment that Iowa and New Hampshire receive, by always voting first.... The primary calendar should instead rotate every four years, with the first states always including a mix of states: big and small, young and old, urban and rural, coastal and heartland.

And maybe the party should customize the primary schedule based on what it believes it needs to do to win the next election. I'd have no problem with bumping Michigan, Pennsylvania, and Wisconsin to the front of the schedule this year -- Democrats need to win them in 2020, and the first two, at least, are a much better mix of rural and urban, white and non-white, than Iowa and New Hampshire.

But I'd rather see a much shorter primary season -- a year and a half is preposterous, and absurdly expensive. I'd appreciate candidates who can make a sacrifice to help the cause -- maybe either Bernie Sanders or Elizabeth Warren should drop out to give the remaining candidate a shot at beating Joe Biden or Pete Buttigieg; maybe Jeb Bush, Marco Rubio, and Chris Christie should have chosen one of their number to fight on against Trump in 2016, while the others withdrew for the greater good.

But ultimately I don't see an ideal primary process. Even giving party leaders a somewhat greater influence would smack of elitism and the "deep state." Would Leonhardt really want Democrats to go into 2020 -- running against Trump and, probably, a Jill Stein-like lefty candidate -- with that burden?