Three of our arguments drew no objection from Dr. Koza. One is that plurality rule—the heart of his proposed reform—can be anti-majoritarian. In our three-candidate example, Libby was the plurality favorite but Maude beat each of her rivals, Libby included, by a majority. A plurality can be a majority, of course, but the Electoral College has never reversed a popular majority.

Another argument is about battleground states. The winner-take-all rule favors them, Koza contends, with a bounty of “campaign events” and promises of pork. But that cannot be much of an advantage, we argued, else the other states would have given up winner-take-all for the division of its electoral votes in proportion to popular votes. Any state which did that would have received much more events and promises.

Our third unchallenged argument is that the proportional division of electoral votes in each state, or a particular version of that procedure, would solve the problems that exercise reformers while avoiding the new ones created by Koza’s proposal. Maybe he thinks the battleground states would never give up winner-take-all. But if most other states adopted proportionality the battleground states would lose their supposed advantage.

Koza’s counter-arguments are seven.

1. 1. Federal courts could enforce the compact under the Contract-impairment clause of the US Constitution.

No, the Compact would not be a legal contract without the Congressional consent required by Article I, Section 10. Yes, courts allow exceptions: when an interstate compact does not affect federal power or other states, consent is presumed (Virginia v. Tennessee, 148 US 503 [1893]). But the Compact would massively affect other states, and Congressional consent is unlikely. The two cases Koza cites are irrelevant. In both, courts said that certain supposed compacts were not binding: each state could unilaterally withdraw. Has there ever been a court decision blocking withdrawal from an interstate compact that lacked Congressional consent? If there were, we are sure, Koza would have found it. (And under the 11^th Amendment, how could contracting states sue a renegade state?)

Koza has stretched and struggled so much to claim enforceability because he thinks that would solve a prisoners’ dilemma: no state would give up winner-take-all “while receiving nothing in return.” Has he forgotten how winner-take-all deprives nonbattleground states of all those prized “campaign events” and porcine promises? If there is a prisoners’ dilemma it must be that (1) all states would fair better under plurality rule than at present, and yet (2) each state would fair even better from defecting to the present system if allowed. But Koza has not made a case for (1).

1. 2. A state cannot change how it appoints electors after the date set by Congress for appointing them, the first Tuesday after the first Monday in November.

Maybe so. But then the Compact is illegal. For it requires each state to appoint its electors well after that date, after all the states have reported their popular votes. The problem is not that it takes time to count a state’s popular votes. It is that votes would no longer constitute acts of appointment. Those acts would instead be performed by a state’s chief election officer based on the popular votes of other states—votes that could not be construed as acts of appointment by that state.

1. 3. Support for the Compact is not partisan.

We showed by several measures that support is heavily Democratic. To our evidence about states that have joined, Koza adds that of four states that rejected the Compact despite significant Republican support. But even there it would have passed but for Republican opposition.

1. 4. The Compact would create no new problem of close votes and nationwide recounts, and votes close enough to force recounts are actually more likely, albeit at the state level, under the present system.

But popular-vote margins of less than 1% have occurred in six presidential elections, including that of 2000. Referring to that election, Koza baldly asserts that such a margin is not close enough to justify a nationwide recount. But in state elections it often is by law, and Ansolabehere and Reeves (Reference Ansolabehere, Reeves, Hall, Atkeson and Michael Alvarez2012) showed that there has been about a 1% difference between initial votes and recounts when both are tabulated by hand, and still a half percent difference when they are optically scanned. Partisan polarization is making vote margins closer (Abramowitz Reference Abramowitz2014), moreover, and the Compact’s overwhelming incentive to seek every vote, regardless of location, would make them closer still. The problem is not that recounts would be more numerous under the Compact. It is that they would have to be nationwide: because no vote count is perfect, a big error in one state could easily be offset by dozens of small ones elsewhere. The obstacles to nationwide recounts are two: lack of time and lack of authority to compel them, especially in non-Compact states. Koza’s solution? Leave it to the courts! But what could they do? Maybe he thinks they would simply accept initial state reports of vote totals, however close. But the true totals would eventually be discovered by political scientists spurred by politicians and journalists. In a close enough election they could well show, several months after inauguration day, that the wrong candidate had been sworn in.

1. 5. Uncooperative electors are unlikely, and state laws can ban them by mandating their replacement.

No. Uncooperative electors would be more likely under the Compact because some electors, especially Republicans, would anathematize the Compact itself. And we cannot imagine those fancied state laws (even if passed in all Compact states) surviving judicial challenge, especially since they would fire and replace electors after the federal legal deadline for choosing them.

1. 6. Manipulation of turnout and vote counts already exists, especially in swing states.

True, but it is much easier in solidly partisan states. Under the current system, however, those states have no incentive to manipulate state-wide votes for president or anything else, thanks to the winner-take-all rule. The Compact would change that.

1. 7. Plurality rule would not encourage candidates to appeal to narrower, less diverse interests than they do today because in battleground states they already make broad appeals.

But that just means that those states are few, pivotal, and closely divided; every last vote must be sought in the tiniest hamlet and the remotest valley. So what? It is still true that, in a nationwide popular vote, plurality rule would reward candidates for accumulating ever more votes from ever fewer places: efficient nationwide media and ground campaigns would have to bypass a lot of hamlets and valleys in favor of dense population centers and closely connected interest groups.

We are glad to end by thanking our magnanimous foe for his titular agreement that the Compact would be somewhat calamitous and our six (actually five) kinds of mischief somewhat mischievous.