The American presidential election system is a complex one that has evolved significantly over the years. At its heart is the Electoral College, which I have written extensively about. Surrounding the Electoral College is a dizzying array of evolved systems that help mitigate the problems with the Electoral College.
A key part of this is the presidential primary process used by the two major parties to select their candidates. The modern presidential primary system is one of the most recent arrivals on the scene, first used in 1972. It’s a complex process with good, bad, and ugly parts to it, and the DNC has proposed some recent changes.
The proposed changes shift the early part of the calendar around significantly, moving Iowa and New Hampshire backwards and moving South Carolina and Georgia forwards. This would represent a significant change in how the early part of the primaries work.
Why do we have this system?
The short answer is that we have the primary process because political insiders kept making bad decisions. The solution to this has generally been to bring the general public further in to the decision-making process.
In 1824, the established process for political parties nominating candidates was the congressional nominating caucus. William Crawford had an impressive set of credentials. He had been in public service for over 20 years, and had held not one, not two, but four of the offices that have led other candidates to the presidency:
Secretary of the Treasury
Secretary of War
Minister to France
Senator (and president pro tempore of the Senate at that!)
He had also secured the endorsement of former Democratic-Republican presidents Thomas Jefferson and James Madison. He was supposed to be the fourth member of the “Virginia Dynasty” of presidents. With the support of numerous key political insiders, William Crawford won the nominating caucus with 93% support. This did not translate to him winning the support of actual voters in large numbers.
When the actual election came along, the man chosen by Democratic-Republican insiders earned a little more than 11% of the popular vote. Three other candidates won more votes. Up-and-coming political newbie Henry Clay, who had been a thorn in the side of the popular Monroe administration, earned 13%. John Quincy Adams, who had only recently switched his affiliation card from “Federalist” to “Democratic-Republican” following the complete demise of the Federalists as a national political force, earned 31% of the vote. Andrew Jackson, who was widely detested by political elites of the day, brought home 41% of the popular vote.
Congressional nominating caucuses were dead, and they’d brought down the first party system of the United States with them. Soon, a replacement system emerged: The Democrats had their first national convention to nominate a presidential candidate in 1832. The Whigs followed suit in 1836, and the national nominating convention entered the American presidential election system.
Eventually, a few state parties started holding presidential primaries. Most were just “beauty contests,” which took the temperature of that state’s voters without influencing which delegates the party sent to the national conventions.
Fast forward to 1968. In a highly contentious nominating process, Democratic insiders chose Hubert Humphrey, who hadn’t really run in the primaries. This managed to outrage both young progressive voters and conservative voters in southern states, and Hubert Humphrey ended up losing the 1968 election to Richard Nixon.
The Democratic Party did some soul-searching after this, and came up with a changed set of rules for how they would choose presidential candidates. State primaries (and state caucuses open to all registered voters belonging to the party) would play a much more prominent role. The Republicans made similar changes. The people would play a more direct role in choosing the president; this is, after all, a democratic republic.
Why not a national primary?
The more candidates in an election, the more chaotic the outcome. In a crowded field - and presidential primaries often have crowded fields - a very unpopular candidate can do quite well.
Starting small contains the chaos. Early contests eliminate candidates before the election is complete. By the time the earliest states are done, the contest will be down to three or four candidates at most; and even the most hotly-contested primary campaigns will usually be down to only two viable candidates before they’re halfway complete.
For the extraordinarily crowded 2020 Democratic field, I predicted ahead of time that after the first two early states, the extraordinarily crowded field of the 2020 Democratic primary would shrink to 3-6 candidates who had placed in the top three in New Hampshire & Iowa (this group proved to be Sanders, Buttigieg, Warren, and Klobuchar), plus up to 1-2 more who had placed outside of the top 3 but had reason to be optimistic about minority voters in Nevada & South Carolina (this group turned out to include Biden and Steyer).
After South Carolina and Nevada, there were, as I predicted, only 3-4 candidates in contention on Super Tuesday. (To be fair, I missed Bloomberg’s remarkable feat of getting taken seriously by many voters in spite of skipping the early states.) The early state system worked correctly for Democrats in 2020, in spite of the fact that the Iowa Democratic caucuses were badly run. The current system is very flawed, but it does have its merits.
Evaluating the proposed calendar
It’s important to have early states that represent the whole country in order to test viability. Thus, the four traditional early states are from four different regions: Iowa for the middle of the country, New Hampshire for the northeast, Nevada for the west, and South Carolina for the southeast.
Nevada and New Hampshire are ideal choices, because they’re small swing states and have swing voters who are typical of the country. Iowa is a very ordinary state for its region; it’s rarely a swing state, but its Democratic voters are similar to swing voters in adjacent states. South Carolina is a bit of a misfit, because it’s a medium-sized state, very far from being a battleground, has an exceptionally low share of swing voters, and is politically out of step with the rest of the country most of the time.
Michigan and Georgia are large states, with Georgia being adjacent to South Carolina. Both are, in their defense, current battleground states; but this new proposed slate of early states is less similar to the national electorate as a whole. The most visible property of the new proposed system is that it front-loads the contest with voters thought to be most likely in Biden’s camp, while punishing the two early states that Biden did poorly in.
In other words, the change in schedule looks very much like an attempt by Biden to fend off a potential primary challenge in 2024. This means that if a serious primary challenge happens, it will be accompanied by accusations that the DNC has cheated in favor of the candidate of its choosing. This did not go well for Democrats in 2016. If there’s one lesson to be learned from the elections of 1824, 1968, and 2016 together, it’s that voters don’t like to feel cheated of a choice.