People protest against President-elect Donald Trump in Miami on Nov. 11. (Javier Galeano/Reuters)

Now that it is clear that Hillary Clinton will win the popular vote in the 2016 presidential election by well over a million votes, we once again face the issue of the peculiar means by which Americans select their presidents.

There is no question that the electoral college violates political equality because all votes are not equal. As a result, the electoral college sometimes denies the people their preferred choice for president, as in 2000 and 2016. But does the electoral college provide some benefits that might compensate for its shortcomings?

The short answer is no. Indeed, two of the most prominent arguments in favor of the electoral college cannot withstand scrutiny.

The electoral college does not prevent tyranny of the majority

The Framers were concerned about tyranny of the majority and therefore incorporated rules in the Constitution that require supermajorities in order to take action, such as the requirement for two-thirds of the senators present to ratify a treaty. Is the electoral college’s violation of majority rule just another example?

It is not. The Framers designed all but one of the Constitution’s extra-majority provisions to enable minorities to prevent an action. The electoral college is different. It allows a minority to take an action — that is, to select the president. It is the only device of its kind in the Constitution.

Thus, the electoral college does not prevent tyranny of the majority. Instead, it provides the potential for tyranny of the minority.

People sometimes think that, if not for the electoral college, a candidate could win by garnering an overwhelming number of votes in one region of the country, imposing that region’s choice on the rest of the country. A quick look at the census shows that this is impossible. The electoral college isn’t necessary to prevent this scenario.

Does the electoral college ensure that the winner receives majority support from different social groups, thus protecting minority interests? No. In 2016, Donald Trump won a smaller percentage than Hillary Clinton among women, African Americans, Hispanics, Asian Americans, voters ages 18 to 44, members of labor unions, those with an income under $50,000, college graduates and those with postgraduate educations, Jews, liberals and moderates, urbanites, and those living in the East and the West.

It strains credulity to claim that Trump’s vote represents concurrent majorities across the major social strata in the United States. What actually happened in 2016 was that the electoral college imposed a candidate supported by white male Protestants — the dominant social group in the country — over the objections not only of a plurality of all voters but also of most prominent “minority” interests in the country.

The electoral college isn’t necessary to maintain the two-party system

Some critics say that allowing voters to directly elect the president would splinter the two-party system. It would encourage many candidates to contest the general election, thus producing a winner with only a small share of the vote.

This is also wrong. In a system of direct election, potential candidates risk their political futures by running against the official party nominees. And there is no compensation. You win nothing by coming in third. So there is little incentive to run.

By contrast, the electoral college encourages third parties, especially those with regional bases, because by winning a few states they may deny either major-party candidate a majority of the electoral vote. You can come in third and still win a prize. Such a result was certainly the goal of Strom Thurmond in 1948 and George Wallace in 1968. Imagine giving these racist candidates leverage to negotiate with the leading candidates before the electoral votes were officially cast.

Moreover, even without winning any states, Ralph Nader inadvertently distorted the vote and determined the outcome of the 2000 election. If Nader had not been on the ballot, Al Gore would have won both Florida and New Hampshire. Either victory would have put him in the White House.

Some defenders of the electoral college seem to believe that direct election would require a runoff between the two leading candidates. But it’s circular reasoning to argue that direct election will produce a plethora of candidates, which in turn will force a runoff, which in turn will encourage candidates to run. In fact, direct election would not produce more general election candidates than does the electoral college, and there would be no need for a runoff.

Moreover, victorious presidential candidates under the electoral college — including, most recently, John F. Kennedy (1960), Richard Nixon (1968), Bill Clinton (1992 and 1996), George W. Bush (2000) and Donald Trump (2016) — have received less than a majority of the national popular vote about 40 percent of the time since 1824. We have not needed a runoff.

More broadly, the electoral college is not the basis of the two-party system. Single-member districts and plurality election are, and the nation would be one electoral district under direct election. Thus direct election would not splinter the party system.

Core arguments for the electoral college are based on faulty premises. The electoral college violates political equity and popular choice, but doesn’t provide any compensatory benefits.

Indeed, the electoral college has perverse consequences, exactly the opposite of those claimed by its defenders. Not only does it not protect us from tyranny of the majority but it actually allows tyranny of the minority. Similarly, not only does it not strengthen the party system but it actually provides incentives for fragmenting it.

As Americans consider changing our election system, it is important that they have a clear understanding of the electoral college’s actual consequences.

George C. Edwards III is University Distinguished Professor of Political Science and Jordan Chair in Presidential Studies at Texas A&M University. He is also Distinguished Fellow at the University of Oxford.