Last week a few lawmakers in California went public with plans to introduce legislation in Sacramento that seeks to prevent presidential candidates who fail to disclose tax returns for the five most recent years prior to the election from having their names appear on the state’s November ballot in 2020 and beyond. The effort is patterned on a similar proposal being pushed by some legislators in New York state. The proposal there, dubbed the Tax Returns Uniformly Made Public (or TRUMP) Act, requires each presidential candidate to disclose tax returns prior to 50 days before the November election, else his name will not appear on the ballot and the state’s electors will be prohibited by state law from casting their votes for him in the so-called electoral college.
While many voters (and certainly many journalists) seem to want access to candidates’ tax return information (to see possible conflicts of interest, levels and directions of charitable giving, relative aggressiveness in seeking to minimize tax burdens, and so forth) before presidential elections are held (and were disappointed that Mr. Trump departed from modern tradition in declining to produce his returns), state legislative proposals like the TRUMP Act raise a number of legal and policy issues; in the space below, I address some of them.
Do States Have Legal Authority to Mandate Tax Return Disclosure?
Certainly a big threshold question is whether states have legal authority to impose such a requirement on presidential candidates as a condition for the candidates’ names appear on the ballot or their being eligible to be voted for by the state’s electoral college contingent. The New York Times published an editorial earlier this month expressing its belief—and quoting leading constitutional scholar Larry Tribe of Harvard—that laws like the ones being discussed in New York and California should be upheld in court. To be sure, there are arguments in favor of state authority here (because states have broad leeway to regulate so-called “ballot access”), but there are also arguments on the other side, such that prediction is dicey. (And the quote from Professor Tribe did not indicate anything more than that a state “might be able” to impose the kinds of tax-return-disclosure requirements being considered.)
Putting aside the question of the independence of electors who are selected (an independence about which I have my doubts), one possible line of constitutional attack against TRUMP Act-like laws is based on a 1983 Supreme Court case, Anderson v. Celebrezze, in which the Court struck down an Ohio law that required any independent (i.e., non-major-party) candidate for president to file a statement of candidacy and nominating petition with state election officials by March 20 in order to be eligible for inclusion on the November ballot. The Court weighed the state’s interest in having an early filing deadline against the associational interests of independent candidates and their supporters, especially in the context of national election where a state’s actions can influence a national outcome. As the Court observed:
[I]n the context of a Presidential election, state-imposed restrictions implicate a uniquely important national interest. For the President and the Vice President of the United States are the only elected officials who represent all the voters in the Nation. Moreover, the impact of the votes cast in each State is affected by the votes cast for the various candidates in other States. Thus, in a Presidential election, a State’s enforcement of more stringent ballot access requirements, including filing deadlines, has an impact beyond its own borders.
It bears noting, however, that 17 years after Anderson v. Celebrezze, the Court decided Bush v. Gore, in which several justices seemed to recognize a much broader authority of states to administer the presidential selection process. Some of the justices seemed to indicate that states didn’t even need to hold elections to gather popular input before selecting representatives to the so-called electoral college, and that as long as states were not treating votes from different parts of the state or for different candidates in a disuniform way (and a tax-return-disclosure requirement would apply equally to all candidates and their supporters), states continue to have broad latitude.
The short of it is that the Supreme Court (were it to weigh in on the validity of TRUMP Act-type laws) could go in a variety of different directions.
A second line of attack might be premised on U.S. Term Limits, Inc. v. Thornton, a 1995 case in which the Court struck down a state law that prevented congressional candidates who had already served a certain number of terms in Congress from appearing on congressional election ballots in the state, holding that states had no authority to impose additional requirements beyond those mentioned in the Constitution for the office of U.S. Representative or Senator, and that preventing the names of long-serving federal legislators from appearing on the ballot amounted to an additional requirement for congressional officeholding. The same kind of argument might be made about the TRUMP Act.
But here too, things are not so simple; popular elections for members of the House and Senate are constitutionally required, but perhaps not so for the offices of the president and the vice president. For this reason, states may have much less leeway to regulate congressional elections than they do presidential selection procedures.
Should States Enact Such Laws?
Moving beyond legal validity, are laws like the TRUMP Act a good idea? That too, depends. If only blue states like New York and California adopt such laws, then only Democratic candidates will have meaningful pressure on them to disclose their tax returns. To be blunt, nothing in the presidential election in 2016 would have changed if the name of Donald Trump (or of electors pledged or inclined to support him) had not appeared on California’s or New York’s November ballots. And that fact was known from the beginning of the election cycle. So, as with so many presidential election reforms, for this one to have any beneficial real-world effect, it would have to be embraced by either a mix of blue and red States, or at least a number of swing states where neither party can feel assured of a victory.
And ballot access regulations such as the TRUMP Act could themselves complicate other well-intentioned and viable reform efforts, like the National Popular Vote (NPV) Compact plan about which I have written a great deal on this website and in academic commentary. For any effort by a group of states to collectively allocate their electors to the national popular vote winner (regardless of who got more votes within each of the signatory states) to work, we need some reliable way of discerning voter preferences in all 50 states (and DC), so that the nominal national popular vote winner is indeed the choice of the greatest number of American voters. But if voters in some states are effectively prevented from registering their preferences (because of something like the TRUMP Act), then the title of national popular vote winner is deprived of much of its democratic legitimacy. (For example, imagine that Mr. Trump’s name did not appear on the New York and California ballots in 2020, and that he was beaten in the national popular vote tally by a relatively narrow margin; surely no one could think he meaningfully “lost” the national popular vote, since he no doubt would have received several millions of votes in California and New York, even though he was likely to lose those states, had he been on the ballots there.)
I have written in a 2011 Georgetown Law Journal article that if enough states do adopt the National Popular Vote plan such states might begin allocating electors according to the plan, then Congress has the power to and should adopt a law providing for more national uniformity to discern a meaningful national vote tally. Proposals like the TRUMP Act convince me all the more of the importance of federal legislation in this realm if and when states making up 270 or more electors sign onto the NPV plan.