Update: Additional analysis on this issue shows that with an adjustment for uncontested races the Democrats would have either tied or taken a slim majority in the house.
As of today, the next session of the House of Representatives will continue to be led by the Republican party.
However, in the Democratic glow following the general election on November 6, 2012 one oft-repeated refrain has been that if the House of Representatives was elected based on national votes by party, that the Democrats would hold both houses of Congress in new session in 2013. Based on current vote totals collected from government and news outlet websites this would be true.
Nationally, Democrats earned approximately 600,000 votes more than Republicans. Since the Libertarian party is the strongest national third party I also tallied their votes separately. The Libertarian party earned 1.1 million votes nationally. This would result in a Democratic plurality in the House:
Although the Democrats would hold the most seats, it would not be unreasonable to expect the Libertarian party to caucus with the Republican party, therefore effectively resulting in the same Democratic minority in the House.
However, an additional 1.5 million voters voted for a different third party candidate or a write-in candidate. If these seats were allocated to an unspecified third party it would shift House seat allocations away from both national parties:
Again, Democrats would hold slim plurality that would be less numerous than a coalition of Republican and Libertarians representatives. I currently do not have the information available to accurately assume if these six additional seats would go to any particular third party. Even if that information was available, we cannot assume that voters would behave in the same manner if there vote counted towards a national tally rather than only impacting their district race. Additionally, several house races were uncontested. Uncontested races are not tallied by all states, and not all voters vote in uncontested races, making estimation of how these voters would break along party lines difficult to estimate. Further, voter behavior and national party candidacy may change if votes counted towards a national or statewide popular vote.
In the end this means little as we elect our congressional representatives to represent our local, or at least state level interests before the nation on Capitol Hill. In light of partisan driven gerrymandering, alternate proposals to elect congressional representatives on a statewide level have seen more discussion. Under these proposals the current district system would be tossed out in favor of a statewide vote based on party affiliation. Officials from that party would then selected to represent the party platform that voters preferred based percentage vote totals. This would allow popular election of representatives while still retaining state representation.
If enacted, alternate methods could be used to maintain a more local aspect to direct candidate selection while having the statewide popular vote as a kind of safety valve against partisan district drawing.
Assuming for now that representatives were elected based on a state basis using popular vote in that state, this would have a significant impact on House partisan make-up. However, since we can’t send partial people to Washington to represent us, this would present some difficult tie breaking issues for results that do not divide nicely among the number of representatives allocated to a state. If the 2012 election was tallied using this methodology, it would result in ten such seats needing a tie-breaking procedure.
Tie-breakers could be divided using a variety of methods. For a state New Jersey that has 12 representatives in Congress and a popular vote that split 55/45 percent along Democrat/GOP lines. Using generic rounding, would give us 6 Democrats and 5 Republicans with one uncommitted member. This could either be awarded to the Republicans to create a 50/50 split or to the Democrats since they won the majority. Ideally a system would be created that would award these tie-breaker seats based on objective measurements. More realistically, these ties would be decided by state governments. Based on the assumption that state governments would follow party lines, this would result in a GOP majority:
However, this lead is narrower than the actual 2012 election results. Here the GOP holds 220 seats instead of 233. The spread between the two major parties is 8 seats instead of 31. This smaller spread would allow moderate members from each party to switch sides on votes if they wish, rather than needing a unrealistically large swing block.
But in conclusion, despite how nice it would be for Democratic supporters to claim that the Presidency and both houses of Congress would be in their party’s hands without gerrymandering, the bare numbers don’t show that. Of course, if state popular vote did exist, an untold number of additional supporters might come out for either side. Ideally, with more seats up the air each candidate would have to work harder for their seat and need to moderate accordingly.
I plan to dive into specific state results in a future post.