During Donald Trump’s first administration, his travel ban policy barring almost all migration from several Muslim-majority nations was challenged in court on the grounds that it was unconstitutional because motivated by anti-Muslim bias. Co-blogger Josh Blackman argues similar arguments should not be accepted by courts in the next Trump administration because polling data indicates Trump made significant gains among Hispanic and Muslim voters in the 2024 election. Judges should not take his advice. The key issue in cases where seemingly neutral policies are challenged for having unconstitutional discriminatory motives is the motivation of the people who adopted them, not the backgrounds of the people who voted for them.
In the travel ban case, the main evidence against Trump was not who voted for him in the 2016 election, but his own repeated statements indicating that his intention was to target Muslims for exclusion, plus the extreme weakness of the supposed “security” rationales for the ban. This should have led courts to strike down the travel ban based on longstanding precedent holding that evidence of unconstitutional discriminatory motive (such as discrimination based on race, ethnicity, or religion) should lead to invalidation unless the government can show it would have adopted the same policy even in the absence of such ilicit motivations. Ultimately, a closely divided 5-4 Supreme Court upheld the travel ban on the grounds that the executive deserves special deference on immigration policy that would not be extended in almost any other context. Elsewhere, I have argued this double standard is badly misguided. Be that as it may, no one on either side of that litigation argued the issue turns on survey data, election returns, or the ethnic and religious composition of the electorate that voted for Trump.
Voters are not the ones who adopt these policies. Moreover, the fact that a candidate has supporters from a given ethnic or religious group doesn’t mean he cannot or will not adopt policies biased against them. He might, for example, do that to satisfy other (to him, more crucial) constituencies. Democrats, for example, routinely get strong support from Asian-Americans (far higher percentages than Trump got from Hispanics or Muslims in 2024), but also support various racial preferences that discriminate against them. Such voting patterns also don’t necessarily show that members of the groups in question actually believe the candidate isn’t biased against them. Many Hispanics and Muslims likely voted for Trump on “lesser evil” grounds, or because of anger at the Democrats at the state of the economy.
Of course, Josh Blackman’s ultimate position may be that courts should ignore evidence of motive entirely. On that view, if a policy is facially neutral, it should be upheld, regardless of the possible motivations behind it.
In addition to going against many decades of precedent, that position would have terrible consequences. As I explained in a post written during the travel ban litigation:
If the Supreme Court rules that campaign statements cannot be considered, that would create a very dangerous precedent. Politicians could openly advocate discriminatory policies during the campaign, then rely on more careful and euphemistic phrasing after they take office. On the campaign trail, they can openly say they want to target blacks, Muslim, atheists, Evangelical Christians, or some other minority group. Afterward, they can adopt a policy targeting some seemingly neutral characteristic that closely correlates with membership in the group in question. And, after taking office, they can stick to carefully scripted official justifications for their actions that elide the true purposes.
I would add that one can use facially neutral criteria that correlate with group membership to target almost any racial, ethnic, or religious minority. Policymakers who seek to discriminate against blacks in hiring could reject applicants from majority-black neighborhoods or graduates of historically black colleges. As long as the policy in question doesn’t explicitly reference race, but merely enumerates neighborhoods or colleges, you’re in the clear! Want to exclude Orthodox Jews? Adopt a policy barring the hiring of people who refuse to work on Saturdays (the Jewish sabbath).
Clever policymakers can easily come up with similar facially neutral, but pretextual ways to target almost any minority group. Indeed, such strategies were repeatedly used by state and local governments to discriminate against blacks after the courts struck down open racially discriminatory policies.
It’s worth noting that excluding evidence of discriminatory motive from judicial scrutiny would shield discriminatory policies condemned by the right, as well as those opposed by the left. For example, universities and selective public high schools with left-leaning administrations sometimes try to use facially neutral admissions criteria to to keep down the percentage of Asian and white students [my wife, Alison Somin, was one of the lawyers representing the Asian-American plaintiffs challenging one such policy, in a case that almost reached the Supreme Court]. If evidence of motive is excluded, educational institutions would have a virtual blank check to use seemingly neutral criteria to get around the Supreme Court’s 2022 decision barring most racial preferences in college admissions.
Indeed, defenders of such preferences could adopt Josh Blackman’s election-returns argument. After all, the left-wing Democrats who enact these policies often win majorities of Asian voters, and they get large percentages of the white vote, too, even if a minority. Many of the relevant policymakers are even whites or Asians themselves.
In the travel ban case, the Supreme Court ultimately did not rule that campaign statements or other evidence of discriminatory motivation should be ignored. Instead, as noted above, it based its ruling on the supposed special deference due to the executive on immigration policy. That was a bad ruling. But a holding indicating that evidence of discriminatory motive is barred more generally would have been much worse.
In sum, election returns should not influence judicial evaluations of possibly discriminatory policies. If such cases arise in the second Trump administration, as they did in the first, the focus should be on motives of the officials who actually adopted the policy in question.
