Monday, December 10, 2012

The Unintended Consequences of Piecemeal Electoral College Reform. Or How to Turn Pennsylvania into New Hampshire

So Pennsylvania Senate majority leader, Dominic Pileggi (R-9th, Chester/Delaware), is pushing a new bill to reformulate the way in which the Keystone state allocates its electoral college votes?

Instead of allocating the electoral votes in a winner-take-all fashion -- as is the case in 48 states and the District of Columbia -- the senator initially proposed shifting to a districted allocation similar to method utilized in Maine and Nebraska. Now, however, the newly amended approach proposes allocating the electors proportionally. The statewide winner would receive the two electors representative of the two federal senate seats and the remaining electors would be allocated based on the proportion of the vote each candidate received.

There is a lot going on here, so let's start with the basics.

For starters, the partisan intent here is to break up the electoral vote bloc from a reliably Democratic state. [Pennsylvania has voted for the Democratic candidate in every election since (and including) 1992. That is six consecutive cycles.] That is, to cut into the Pennsylvania and overall national electoral vote tally for the the Democrats. Both of the plans proposed by Leader Pileggi accomplish this, but the newer version is less beneficial to the Republican Party. Due to the way in which the Republican-controlled Pennsylvania legislature redrew the congressional district lines following the 2010 census, Mitt Romney would have won Pennsylvania, winning more districts and thus electoral votes under a districted allocation than Barack Obama. As Nick Baumann pointed out in the Mother Jones write up, that would likely have meant Romney winning the 13 (Republican) congressional districts and Obama taking the remaining five (Democratic) districts plus the two remaining electoral votes for winning statewide.

Fair or not, that would translate into Obama winning nearly 52% of the vote in Pennsylvania but taking only 35% of the electoral votes from the state.

Under the revised plan, Obama's 52% would translate into 55% of the electoral votes. The president and Mitt Romney would have evenly split the 18 non-statewide electoral votes and Obama would have won the remaining two statewide electors.1 That is likely to be marginally more palatable to Democrats, but not nearly as good as taking all of the electoral votes under the current distribution.

Of course, there is unified Republican control (across the legislative and executive branches) in Pennsylvania that would not necessarily require Democratic support. But the intent of the switch in plans seems to be to create an argument based on fairness; that this is a fair way of allocating electoral votes. Pennsylvania Republicans would be on firmer ground with that argument on behalf of the revised allocation plan than attempting to push a districted plan that produces an allocation that does not reflect the statewide vote. ...or produces a distribution that is so noticeably distinct from the statewide vote.

Let's take a national detour here and FHQ will revisit the situation within Pennsylvania momentarily.

Normally, FHQ would be ecstatic at the prospect of such a rules change. In fact, I spent the weekend mulling over the implications nationwide. But this is not something that the Republican Party would necessarily want employed nationwide in each state. Strategically, a party would want to keep states that are reliably, in this case, Republican and maintain a winner-take-all allocation of electoral votes there. That maximizes the number of electoral votes the party would receive. Ideally, the party would want to push this strategy in blue presidential states that are redder down-ballot. This means states like Florida, Michigan, Ohio, Pennsylvania and Wisconsin; Obama states where Republicans have unified control on the state level.

If we were to assume a proportional allocation of electoral votes across those five states based on the two-party vote (see footnote 1) that would have netted Mitt Romney an additional 40 electoral votes.2 That would still have put the former Massachusetts governor short in the electoral vote tally (Obama 292, Romney 246). It would be closer, but still shy of the 270 electoral vote threshold.

Granted, the calculus is more complicated than this. The Republican Party (nationally) would not want to employ this strategy in states that it could win outright in a political climate slightly more favorable than the one the party faced in 2012. That is cannibalizing the party's own potential electoral vote total. This probably eliminates states like Florida and Ohio. In turn, that reduces the potential electoral vote gain from 40 to 19 under the currently apportioned electoral college (Obama 313, Romney 225).

Now sure, if race had been closer in 2012, the addition of 19 electoral votes (or 40) could have mattered. It would also potentially have increased the likelihood of an electoral college winner different from the popular vote winner. If those are the rules, a win is a win, but FHQ is not entirely sure that increasing the likelihood of a popular vote/electoral college split is a goal for which we should be aiming.

This brings us back to Pennsylvania.

Under the right conditions, Pennsylvania adopting a proportional method of electoral vote allocation could increase the likelihood of the aforementioned split. This is even more the case if toss up to lean blue but Republican-controlled states like Michigan and Wisconsin follow suit. But let's focus on a scenario where Pennsylvania walks this road alone. The plan as outlined by Leader Pileggi passes the Pennsylvania legislature and is signed into law sometime before 2016. What impact does that have?

Well, we know that the resulting electoral vote tally for the two major party candidates is likely to be close. The math in a state like Pennsylvania is such that the non-statewide allocation will be tied unless one candidate -- over the last six cycles, a Democrat -- clears just under 53% of the two-party vote. Assuming the Democrat does clear that barrier, he or she would win Pennsylvania by four electoral votes. If not, then that candidate would win the Keystone state by two electoral votes.

Why would any campaign waste much of any time or money on a state where they would... 1) under normal circumstances gain two electoral votes and 2) have to spend a lot of money in media markets like Philadelphia and Pittsburgh?

The answer is they likely wouldn't. FHQ is not suggesting that the campaigns would not spend any time or money in Pennsylvania. Some resources would be spent there to be sure. But the motivation would perhaps be to spend money in a state where the electoral vote gain is going to be larger. There is a reason the Romney campaign turned to Pennsylvania late in the 2012 race. Ohio was not budging and even though Pennsylvania represented a state that was further toward the Democrats than the Buckeye state, it was one that was both largely proximate to Ohio in electoral vote strength and one that had to that point seen comparatively little campaign activity (There were potentially persuadable voters there.). With a diluted Pennsylvania, the next best options for the Romney campaign would have been to turn the focus to a similarly sized state even further out on the Electoral College Spectrum (like Michigan) or to spread resources across several states in an effort to recoup electoral votes lost in Ohio.

This plan just seems shortsighted. FHQ gets the perceived benefit in a national zero-sum game of winning electoral votes. And when couched in terms of fairly reflecting the statewide vote in the electoral vote tally the argument is even more convincing on the surface.

  • Yet, is it fair to Pennsylvanians to reduce the clout of their state in the electoral college -- to reduce the potential electoral vote prize there with the likely result of decreasing attention to the state? 
  • Is this reasonable when an exchange of nine electoral votes is very likely not going to alter the outcome of the electoral college? [The outcome is rarely that close. It would have made Bush's electoral college win wider in 2000, but would not have gotten him any closer to winning the popular vote. The intent in Pennsylvania anyway is to overturn a Democratic win in the electoral college. Those nine electoral votes would not have made a difference in any of the Democratic wins in either the 20th or 21st centuries. Even if you pull Michigan and Wisconsin in, it doesn't change anything. In some cases during Democratic electoral college wins those states were voting Republican.] 
  • Is is fair to Pennsylvanians to transform Pennsylvania into New Hampshire in the electoral college? [Yes folks, New Hampshire receives attention, but it receives the sort of attention a competitive state with four electoral votes would receive when compared to one with 20: less.] 

Why do that? Why dilute the value of a state in the electoral college for less than clear benefits?

Beware the unintended consequences of altering electoral rules.

1 There are a couple of interrelated items that remain unclear in the updated proposal. First, there is no mention of how fractional electoral votes will be treated. Second, there is no mention of third party candidates. If this is to be anything like delegate allocation in the nomination phase of the presidential election process, then there is likely to be a threshold that third parties have to reach in order to receive electoral votes. As it stands, a third party candidate has to win 5.56% of the vote in Pennsylvania in order to win one electoral vote. Even that is a high bar and may negate the need of a threshold. However, this has implications for the first issue: How and on what is the rounding of fractional electoral votes based? If no third party candidate crosses the 5.56% threshold, then that means the allocation of electoral votes would be based most logically on the two-party vote (or the percentage of the vote for each candidate who cleared 5.56%). As that count currently stands, Obama has 52.73% of the two-party vote which equals 9.491 (out of 18) electoral votes. That is not enough to round up to the 10 non-statewide electoral votes that Leader Pileggi's memo allocates to Obama. Either something is not right about that math or the rounding mechanism has not been adequately outlined.

2 The breakdown would look like the following:
Florida -- Obama: 16 (2 statewide + 14 non-statewide), Romney: 13
Michigan -- Obama: 10 (2 statewide + 8 non-statewide), Romney: 6
Ohio -- Obama: 10 (2 statewide + 8 non-statewide), Romney: 8
Pennsylvania -- Obama: 11 (2 statewide + 9 non-statewide), Romney: 9
Wisconsin -- Obama: 6 (2 statewide + 4 non-statewide). Romney: 4

This raises a couple of interesting points:
1) It is quite difficult to receive a high enough share of the two-party vote to round up and create any kind of cushion in the non-statewide electoral vote allocation. That means that the true difference in a moderately competitive state (Remember the targets are going to be blue states with Republican control. Those are typically going to be more competitive states on the presidential level.) will be equal to the two electoral votes allocated based on the statewide vote with one exception...
2) If things are extremely close, rationally acting candidates would prefer to push resources into states with an odd number of electoral votes. Winning a state with an odd number of electoral votes would produce a one electoral vote differential between two major party candidates in the non-statewide allocation plus the two electoral votes awarded to the statewide winner. These are exactly the types of calculations the Obama campaign was using during the 2008 Democratic primaries (and caucuses!).

Are you following FHQ on TwitterGoogle+ and Facebook? Click on the links to join in.

No comments: