For half a century, Republicans have languished under a presidential-selection system designed by the left wing of the Democratic Party. Whatever the merits of this system for Democrats, its results haven’t been pretty for the GOP. Since Ronald Reagan left the White House for Rancho Del Cielo in 1989, Republicans’ median presidential performance has been to lose by 100 electoral votes.

Better general-election results require better candidates. Republicans should therefore adjust their presidential-selection process to empower the party’s grassroots and produce nominees who are popular with a wider swath of the party.

As Christopher F. Rufo writes, the politics of 1968 “have set the patterns and bounds of our national life for decades.” Such is the case for presidential selection. The current process—which empowers donors, consultants, and the press corps—is an outgrowth of the 1968 Democratic National Convention. At the time, the New Left was frustrated because Hubert Humphrey won the nomination without entering a single primary. The McGovern–Fraser commission, formed in response, led to Democrats converting to the current process. Republicans obligingly followed suit.

The post-1968 system has affected each party differently. Republicans have repeatedly, maddeningly, defaulted to their next-in-line candidate—the candidate they rejected the last time. Nominating last time’s loser hasn’t produced a general-election winner since the 1980s. Democrats, meantime, have been more apt to go with new blood and prevail.

The Republican contest has also become a battle of the politically unemployed. Over the past 30 years, among candidates receiving even 1 percent of the Republican primary vote in open races (those without a GOP incumbent), 61 percent didn’t hold political office during the year of the election. Only two sitting governors in the past 50 years, and no sitting cabinet secretaries, have hit that 1 percent mark. It’s hard to sustain a two-year campaign when one has a real job.

The current process also often yields a winner who lacks majority support. John McCain, Mitt Romney, and Donald Trump each won just 38 percent to 40 percent of the votes that were cast before their top opponent dropped out. Millions of Republicans haven’t gotten to cast meaningful votes at all: in 2012, 76 percent of GOP voters either didn’t vote for Romney or voted for him only after runner-up Rick Santorum had conceded the race.

Rather than being satisfied with marginalizing their own voters and routinely producing subpar candidates, Republicans should adopt a new system.

Jay Cost and I have proposed such a system, modeled after the process used to ratify the Constitution. (We have described it at some length, recently in the Claremont Review of Books and earlier in National Affairs). Under our plan, 3,000 delegates—roughly one per U.S. county—would be chosen by grassroots Republicans in town meetings or local elections conducted by local chapters of the Republican Party, much like how delegates were chosen when the fate of the Constitution was at stake. Rank-and-file Republicans in a midsize state like Missouri or Arizona would get to choose about 60 delegates, through 60 separate town meetings or local elections.

Those delegates from across the country, supplemented by up to 600 party leaders (including governors, congressional members, and Republican National Committee members), would meet at the inaugural Republican Nomination Convention. Over a three-day period beginning roughly five days after January 1, the delegates would take a series of votes, listing up to five selections for their party’s nominee. After the third vote, the ten highest-scoring potential candidates would be contacted to confirm their willingness to compete if chosen. After the fourth and final vote, the proposed candidates with the five highest scores would officially be invited to compete for the party’s nomination. (In years in which there was an incumbent Republican president, the support of three-quarters of the delegates would be required to throw the weight of the convention behind the incumbent.)

Under our system, those five candidates—and only those five—would be invited to participate in the first Republican presidential debate. The first debate would take place roughly five months later than it does now. Debates would occur weekly thereafter, thus increasing their importance while reducing the sway of ads—and hence of donors and consultants. Primary voting would subsequently decide the nominee as it does now, with the selection being made official at the made-for-TV summertime convention.

Such a system would represent a marked improvement. It would shift control to the party’s grassroots—allowing them to choose the delegates who winnow the field, rather than having that winnowing be done by donors, consultants, and the press corps. It would produce better candidates, as many potential candidates who now avoid the two-year slog would welcome the chance to compete in a shorter race more focused on ideas than on fundraising. It would produce consensus nominees, as the five finalists would almost certainly be more widely admired candidates that the five de facto finalists that generally emerge under the current system. It would give more Republican voters a meaningful say, as the five candidates likely wouldn’t be quick to quit the race, before many Republicans even got to vote. (Right now, generally about half of all GOP voters cast their ballots only after the nominee has effectively been decided.) And it would leave far more money available for November, rather than wasting it on roughly a year’s worth of intraparty bloodletting.

Importantly, the delegates to the nomination convention would be chosen—through town meetings or local elections—by grassroots Republican voters. Local party leaders would not simply choose the delegates themselves. This is consistent with how delegates to the Constitutional Convention were generally selected, and it would empower rank-and-file Republicans.

Today’s near-permanent campaign increases the importance of money, and hence of donors. Shortening the campaign to something closer to its length in 1960, when John F. Kennedy announced his candidacy on January 2 of that year, would weaken the influence of donors and strengthen the influence of the grassroots.

Republicans routinely, and rightly, talk about the importance of Tocquevillian civil associations, a particular concern in a country in which people increasingly feel isolated. Alexis de Tocqueville himself said, “In democratic countries the science of association is the mother science; the progress of all the others depends on the progress of that one.” Without such thriving voluntary associations, a bureaucratic “elite” gradually takes over the activities previously undertaken by free men and women. But Republicans’ first step in restoring such vital civil associations should be to make their own party once again function like one. Giving local Republican voters the power to choose delegates to the nomination convention would strengthen local party chapters and go a long way toward breathing new life into these building blocks of republicanism.

Republicans don’t need to stay on the course set in 1968 by their political rival. If the 2024 GOP presidential nominee were to champion this new process—perhaps through launching a commission, which is how the system was changed last time—it could potentially be in effect as soon as 2028.

By adopting a nomination system that is more efficient, deliberative, consensus-based, and—yes—republican, the Republican Party could once again become worthy of its name.

Photo by STEFANI REYNOLDS/Getty Images


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next