Yesterday’s New York Times had an article about the latest instance of Republican trouble- making as we head into the 2012 election year. This involves a proposal in the Pennsylvania legislature to award the state’s electoral votes by congressional district (plus two at-large), replacing the long-standing “winner-take-all" method, that is in use in all but two small states (Maine and Nebraska). The theory, apparently, is that, while no Republican presidential candidate has carried Pennsylvania in decades, such a change would award the Republican at least some of the state’s electoral votes, thus reducing President Obama’s tally – and possibly even denying him re-election.
Of course, the Pennsylvania legislature is legally entitled to do more or less whatever it wants in this matter. The Constitution leaves it up to each state legislature to determine hw a state’s presidential electors should be chosen: in such Manner as the Legislature thereof may direct (Article 2, Section 1). At the beginning, most electors were chosen by the state legislatures themselves, which was probably what the Framers themselves had expected. With the rapid democratization of presidential elections thanks largely to the quick development of organized political parties (something the Framers apparently didn’t anticipate and actually desired to avoid), the popular election of presidential electors soon became the norm. So also did the “winner-take-all” system – a system which, by and large, has served our country well. If nothing else, it has deeply entrenched our 2-party system (since it makes it so difficult for a 3rd-party or “Independent” candidate to win even a single electoral vote). For better or for worse, it has made carrying certain states more valuable than carrying certain other states, thereby structuring the orientation of presidential campaigns in certain predictable ways.
Any change in the electoral process presumably always has a partisan purpose. Like campaign finance reforms, they also often have unintended consequences. Court-mandated legislative reapportionment in the 1960s, for example, was widely hailed by those who sought to increase the power of Democratic constituencies in cities. It ended up facilitating computerized reapportionment as a form of bi-partisan incumbency insurance. In the case of the currently proposed Pennsylvanian plan, one unintended consequence could be that Democrats (deprived of the traditional incentive to maximize turnout in urban districts, e.g., Philadelphia) might concentrate their resources instead in swing suburban districts – and in the process possibly unseat some Republican congressmen in those districts. Precisely such unintended consequences have always served as a prudential check on such partisan, short-term electoral tinkering.
Such structural tinkering – electoral, or even worse constitutional – tends to bring out the Edmund Burke in me. Of course, even the constitution needs to be amended form time to time, but there is a good reason it has been done so rarely. Partisanship aside, it’s generally dangerous to subject inherited traditional institutions to too much rational analysis.
That applies even to the (admittedly less-than-rational) inherited institution of the electoral college. It think it’s safe to suggest that, if the constitution were being written today, the electoral college would not figure in anyone’s plan. Likewise, if Canada were being created today, it perhaps wouldn’t occur to anyone to make it a monarchy. But countries and their constitutional arrangements are historical products, which have helped form their societies in specific ways. Hence, Canada remains a monarchy for good historical reasons, and U.S. Presidents are chosen by the Electoral College for somewhat less compelling but nonetheless good historical reasons. There is nothing remotely rational about how our electoral college functions – except that it has served us well. Admittedly, not only does it distort the popular vote, but on some occasions (1888, 2000) it has actually nullified it. That said, the system still has served us well on balance. Its elimination would severely undermine the present 2-party system, for example. Maybe having multiple parties has much to recommend it; but we don’t know how the benefits of a multi-party system could be enjoyed in a presidential system, where a parliamentary-style “coalition government” is physically impossible. And it could conceivably create even more gridlock, even more of a policy impasse than we already have – the solution to some theoretical problem somewhere, no doubt, but not to the problems facing us right now. (Nor should we underestimate the “legitimizing” effect of the electoral count in an era of closely contested popular elections – especially in our increasingly polarized and litigious society).
In short, American society and politics have adapted well to how the electoral college (including the "winner-take-all" allocation of electors) functions. The best thing to do would be to leave it alone.
It was Chesterton, I believe, who supposedly said that if it is not necessary to change, then it is necessary not to change. The success of any electoral or other political arrangement depends in significant part upon its predictability and moral legitimacy. Change challenges that. Partisanly motivated change undermines it that much more so.