A Perfect Storm of Defects

A habitual liar with no relevant experience but a long history of shady business dealings becomes the supreme leader of the most powerful nation in the world, a nation with a proud history of democratic self-rule. How did it come to this? Was Donald Trump’s upset victory in the 2016 U.S. Presidential election a triumph of democracy? Trump ran a “populist” campaign promising to do what the people wanted. The people voted, following the long-established rules of one of the oldest democracies on the planet. Those who didn’t support Trump might not like the results, but that’s how democracy works, right?

Actually, no. For starters, Hilary Clinton got nearly 3 million more votes than any other candidate. Under the rules of election to any other single office in the United States — or in the world, for that matter — she would have been the winner. But there are more defects of U.S. democracy that contributed to Trump’s victory.

The Majority Does Not Rule – Nor Does the Plurality

The U.S. presidential election operates under unique rules: there is no national election, but 51 separate contests, each with its own rules. One defense of the system trotted out following Trump’s victory is the need for “concurrent majorities” among the separate states. This argument is ahistorical, in that the Constitution allows states to select electors by any means, including appointment by state legislatures, and envisioned that the electors would have free agency to pick the president. The “winner-take-all” system of electing a general ticket of electors (rather than by district) evolved due to party competition between states, but is not part of the constitutional structure. (See previous post for more details.)

Further, no state actually requires a majority of votes in order for an entire slate of electors to be awarded to a single candidate. Only a plurality is required. In Michigan, Pennsylvania, and Wisconsin, the minuscule share of the vote for the Green Party was more than Trump’s margin over Clinton. Had these few voters joined most other left-leaning voters in picking Clinton as by far the least bad choice, Trump would have lost. It’s worth looking at just how tiny those margins were in these three decisive states:

Source: Wikipedia, United States presidential election 2016 – Results by State

If even half of the 77,744 voters in these state who gave Trump the plurality had instead voted for Clinton, the national result would have been reversed. These are tiny numbers given that there were an estimated 230 million eligible voters. Stein’s policies were the most opposite of Trump’s, and yet her supporters could have prevented Trump’s victory. In fact, it is likely that the vast majority of her would-be supporters did vote strategically for Clinton. Finally, the table above also shows that in these three key states, as in the nation, a solid majority of voters opposed Trump. But the rules do not require that the outcome satisfies the wishes of the majority.

Voters and Non-Voters

Consider some other numbers: Trump received almost 63 million votes, about 3 million fewer than Clinton. Nearly 8 million voters picked neither Trump nor Clinton in the general election. More than 2 million  of those who voted somehow failed to make a choice for President (based on Professor Michael McDonald’s data).

About 3 million U.S. citizens were disqualified by state laws prohibiting voting by felons or ex-felons. Another 3 million residents of U.S. territories (2.7 million in Puerto Rico alone) are U.S. citizens but not eligible to vote in Federal elections. (The half million citizens of voting age who live in the District of Columbia are eligible to vote for President, thanks to the Twenty-Third Amendment, but cannot vote for members of Congress.)

The largest group of non-voting adults, nearly 94 million, are those who are eligible to vote but did not cast a ballot in the presidential race. This number is staggering given that fewer than 39,000 lucky voters in three key states decided the election: that is, the outcome would have changed if that number of voters had picked Clinton instead of Trump. In most other states–representing 63% of the eligible voters–the election was not competitive (defined as a statewide margin of 10 percentage points or more). Additional votes for a candidate heavily favored in a particular state have no effect on the national outcome, thanks to state laws that mandate that all presidential electors be awarded to the plurality winner.

There are a number of other factors, besides the method of awarding electors, that might have deterred voters. Most states require voters to register at least three weeks in advance of the election. This may not seem like much of a burden (at least since Congress required states to offer mail-in registration). However, every year young people who have never registered, or people who have moved since they last voted, find themselves effectively disenfranchised because they have not registered in time. Only in the minority of states with same-day registration (or in North Dakota, where there is no registration) can new (or recently moved) voters vote on election day without any prior planning.

Since 2010, twenty states implemented policies to discourage voting, such as reducing or eliminating early voting and requiring photo identification without issuing identification to citizens who lack it and without providing a backup mechanism. One of these states, Wisconsin, which implemented a photo ID law in 2016, was one of the three decisive states. Another of the three, Pennsylvania, would have had a similar law in place except that it was invalidated by a court.

Voter turnout in 2016, 59% of those eligible according to Prof. McDonald, was not low compared to other U.S. presidential elections since 1900. However, this level of participation is regularly exceeded by most other democracies, especially those among the developed world. Turnout as a percentage of voting age population, the statistic available for comparisons, was 55% in the 2016 U.S. presidential election. According to data from the Institute for Democracy and Electoral Assistance, for all presidential elections held globally since 1990, average turnout compared to voting age population exceeded 55% in 79 countries and was lower than 55% in only 30 countries, all of which were in Africa, or were among the lowest income countries of the world, with the exception of Ireland. The Irish presidency is largely ceremonial; average voter turnout in Irish parliamentary elections since 1990 is 66%, still on the low side by international standards but significantly higher than U.S. presidential elections.

Even a small change in turnout (especially in the right states, under the U.S. presidential election system) can have a big impact. The election of 2008 had the highest turnout of any since 1968: 61.6% of eligible voters (56.9% of the voting age population). If the 2016 election had matched this rate, there would have been 5.3 million more votes counted, certainly enough to change the outcome, given that non-voters may not have the same preferences as voters.

Lack of Ballot Choice–Primaries Are No Substitute

Some voters may have stayed home because their preferred candidate was not on the ballot. Bernie Sanders mobilized millions to become active in politics, and although many of his supporters voted for Clinton, and some for Stein, it is likely that millions more chose not to vote. One post-election poll of registered voters found that self-identified Democrats and independents were less likely to turnout than self-identified Republicans. Five Thirty-Eight’s analyst concluded, “had the non-voters cast a ballot in accordance with their party identification, Clinton’s advantage over Trump nationally would have expanded by about 2 to 3 percentage points. That almost certainly would have been enough to flip enough states for her to win the Electoral College.” Further, the poll found that “the biggest reason given by non-voters for staying home was that they didn’t like the candidates.”

If U.S. presidential elections had a majority requirement, it is likely that many more Sanders supporters would have expressed their true first choice at the ballot box, and then would have given a second choice vote for Clinton (either in a two-round system or a ranked choice system), rather than either staying home. Stein supporters would have been able to both express their true preference publicly and cast a useful vote — at least one that did not contribute to their least favorite candidate’s victory. At the same time, those who opposed Clinton should have been able to vote against Trump without fearing that doing so would contribute to Clinton’s victory.

But didn’t voters make choices among many candidates during the six-month long primary election season? Primary elections are not a substitute for a majority requirement in a general election. Presidential primary elections and caucuses are a means of selecting delegates to a national political party convention, with rules varying from state to state and party to party. Over time, state legislatures intervened in the formerly private political party process to create quasi-governmental elections. Primary elections are seen as part of the standard Presidential election process, even though at the presidential level they have only been in widespread use since 1972 and remain much more heavily controlled by party rules than other primary elections.

Many Sanders supporters were outraged that the Democratic Party leadership supported Clinton, both in terms of institutional support and superdelegate votes. Yet isn’t it fully within the rights of a political party to pick its favorite candidate, especially over an outsider who had never been a party member before seeking to become its nominee for the highest office in the country?

The rules are hardly fairer on the Republican side, where in many states delegates are not awarded in proportion to the vote (as is true under Democratic Party rules), but instead by winner take all, or proportionally but with a high threshold, or by state conventions, or by any number of unique combinations of rules. The lack of proportionality explains why Trump received 62% of the pledged delegates with only 45% of the vote.

Originally the system was designed to select delegates to a convention where they would actually choose the party’s nominees. Instead, delegates and conventions have become vestigial, and it is now considered abnormal that they would do what they were intended to do: select a nominee. In an echo of state laws that attempt to bind the votes of Presidential Electors, whose role has been vestigial since the first contested election in 1796, most states have adopted laws binding delegates to vote for the candidate they are pledged to. The Never Trump movement among some Republican Delegates claimed that party rules that permit convention delegates to vote their conscience trump state laws. The effort was not successful, in part because party rules also include statements about bound electors.

Both parties also make use of caucuses instead of primaries in some states. These are essentially elections that require those participating to be present at a specific time. They always have much lower turnout than primaries and generally attract only the most committed activists. Despite all the media attention given to primary elections, only 28.5% of eligible voters participated in 2016, a higher participation rate than years where one candidate is an incumbent president (such as 2004 and 2012). Nevertheless, even with active contests in both parties, turnout in primary elections is half what it is in general elections, which is already on the low extreme of international norms for developed countries. One reason for the low turnout is that the elections do not happen at the same time, and many candidates drop out before most voters have had the opportunity to participate.

Non-Major Candidates in the General Election

But isn’t is true that the primaries are merely the way for the major parties to select their candidates? Other parties are free to compete in the general election.  State ballot access laws give the Democratic and Republican parties a superior status not granted to small or new parties. At the same time, state laws effectively grant open access to any potential candidate or participant in the nomination process of either of the two “major” parties.

Therefore serious presidential contenders have no business mounting a campaign except as a candidate of one of these quasi-official parties. The obstacles to working outside these two are insurmountable:

• ballot access is regulated by state law only, and even with some recent easing, about half a million signatures were required in 2016 to appear on the ballot in every state.
• many of the deadlines for ballot access occur before the Democratic and Republican conventions are complete, that is, before the opposing candidates are known.
• non-official candidates are not invited to participate in the quasi-official national debates run by an entity controlled by the official parties. This is in contrast to the primary campaign, where all candidates, no matter their performance in poling, are invited to participate in televised debates. Following his inclusion in national debates in 1992, Ross Perot received almost 19% of the vote, the highest share for a non-official candidate in 80 years.
• the lack of a majority requirement means that supporters of a non-official candidate are highly reluctant to show that support, lest their least favorite candidate be elected.
• with all of the above factors, there is a strong media and cultural bias against supporting non-official candidates, even as fewer Americans identify with or enroll in the two official parties.

Given that there are no obstacles to participating in the primaries, it is no wonder that self-styled socialist Bernie Sanders chose that route, or that Trump (despite his earlier Democratic registration) chose to run as a Republican, as did Libertarian Party leader Ron Paul before him.

It is also not surprising that Sanders, having come close to winning the Democratic nomination and having mobilized millions, decided not to contest the actual election. Similarly, conservative organizations who were unconvinced that Trump shared their convictions, or who were concerned that he was unelectable or incompetent, made no effort to put forward an alternative candidate for the general election.

Propaganda, the Media, the Right, and the Russians

How could the United States have elected someone who routinely tells falsehoods in the manner of a con artist? Besides all the failures of democratic norms already recounted, another factor is that many voters were ready to believe even his most obvious lies. Trump came into a political climate where conservative forces had already been working for decades to spread half-truths, distortions, and outright falsehoods to build support, particularly among voters who would generally not benefit from the policies that conservatives frequently wish to implement. This topic is worthy of a separate post, but for now we can note that in this century the conservative / Republican propaganda machine managed to convince millions of people that Al Gore claimed to have invented the Internet, that John Kerry is a flip-flopper, and that Barack Obama is a Muslim and not born in the United States (Trump was of course a major agitator behind that last calumny). Continuing its 80-year campaign against universal health care, the right propagated a long list of untruths about the Affordable Care Act. These highly successful propaganda campaigns are just a few examples of what has been a decades-long rightwing media strategy to present an alternative to fact-based journalism (which they rebranded as the “liberal mainstream media”), as described in The Republican Noise Machine: Right-Wing Media and How it Corrupts Democracy.

On top of all this, we now know that Russian President Vladimir Putin has a personal grudge against Clinton, a desire to destabilize democracies, and possibly a longstanding relationship with Trump. In the months leading up to the 2016 election, hackers working for the Russian government infiltrated email accounts and systems belonging to U.S. political parties and politicians, used trolls and bots to spread pro-Trump propaganda on social media, and attempted to hack into the voter registration systems of 20 states. It’s not clear how harmful the emails stolen by the Russians and released by WikiLeaks were to the Clinton campaign–probably not as harmful as FBI Director Comey’s unwarranted late-October announcement of further investigation into Clinton’s emails, nor as harmful as the amount of completely unwarranted coverage given to the email server matter. The effectiveness of the trolls and bots is currently being investigated.

The All-Powerful President

In summary: Trump won the U.S. presidency by exceedingly narrow margins in a handful of key states, even though he was not the choice of a majority of voters in those states or nationally, and was the plurality loser by nearly 3 million votes nationally. Further, about 6 million citizens were denied the right to vote in the election, and 94 million more chose not to vote — or were dissuaded from voting by photo ID laws or long waits.

This insufficiently democratic election result will have major impacts on the world because the U.S. presidency has become so powerful. I have previously described how the current power of the presidency is at odds with constitutional design and original intent. Most of the vast Federal bureaucracy is ultimately under the control of the President; there are only a handful of congressional agencies. In parliamentary systems, by contrast, government agencies are ultimately responsible to the legislature, and, in the case of coalition governments, may be headed by members of different parties. Furthermore, the U.S. President has the power to launch a unilateral nuclear strike and nominates members of the Supreme Court.

According to the original constitutional design, Congress was to be the most powerful arm of the government. Today it is only to be hoped that Congress can serve as a check to unwarranted executive actions. Despite the explicit constitutional provision that only Congress can declare war, the President has become virtually the sole power in foreign affairs. There is no mechanism whereby Congress can recall an unpopular or incompetent President. The only possibility is to remove the President for criminal offenses, which has never happened in U.S. history, or to replace him in the event of physical or mental incapacity under the 25th Amendment.

The other problem with Congress’s failure to be a democratic check on the President is Congress’s own lack of democracy:

• Midterm Congressional elections, which are equally as important as Presidential years, generally attract no more than 40% of eligible voters.
• The U.S. Senate is probably the most malapportioned powerful legislative body in the world.
• U.S. House elections are uncompetitive, with incumbent re-election rates routinely exceeding 95%.
• U.S. House district lines are drawn by politicians and may be (and generally are) drawn to benefit the majority party and/or to protect incumbents generally.
• U.S. House seats are not proportional to votes. In the 2016 elections, Republican House candidates had 49% of the vote total, just over the 48% that went to Democrats, but received 55% of the seats. In the 2012 elections, Republicans had 48% of the vote, less than the 49% won by Democrats, but received 54% of the seats.

We need to understand the defects of democracy before we can fix them.