http://www.realclearpolitics.com/articles/2013/05/16/why_the_gop_lost_the_vote_but_gained_house_seats__118430.html

The author concludes that the Republicans would still have won the House, mainly because of the concentration of the Democratic vote in urban areas. He makes the point that this doesn’t hurt the Democrats as much in the western states, hence they did well with the Arizona and California maps.

In fact, there were many close districts that could have swung under the different campaign effects of an actual district-based presidential contest. Romney won 27 of his districts by five percentage points or less. Obama won 21 of his by such a margin. (Want a tighter margin? Romney won 12 by two or less, Obama 10.)

It is not necessarily implausible that a popular vote lead of nearly four percentage points could be reversed through single-seat districts absent gerrymandering. If one party tends to win by large margins, due to concentration in urban areas, it can fail to win a majority of fairly apportioned districts even with a majority of the vote.

Note that I am not claiming US House districts are fairly apportioned. I am simply noting that this result is not in itself evidence that gerrymandering is decisive.

In any case, I hope to have a chance to “play” with the dataset at some point in the (near?) future.

However, it appears that Romney carried a majority of districts, 226 to 209. Obama’s percentage of the vote (as calculated by Wikipedia) appears to have been 51.1%. This was also a 3.9% margin over Romney. Reversal of a 3.9% margin is implausible without gerrymandering.

The presidential vote is important because it puts Democrats in a position to needing to convince some Romney voter to switch tickets down-ballot. They needed Romney -congressional Democratic voters in 2012. Note that this is a normal situation for the Democrats. Democratic presidential candidates received a majority of the popular vote only four times since FDR died, and only carried a majority of congressional districts once or twice.

However, they fell short of a congressional majority by 16 seats, so maybe they lost a dozen due to gerrymandering and another dozen to having their vote concentrated in the cities. Republicans did better in the national congressional popular vote than in the national presidential popular vote, but not by enough it seems to make a difference.

In NSW a string of government agencies are required to notify the electoral commission of changes in status and address. Example, the education department notifies the commission of students who turn 18. The commission enrols people as it becomes aware of them and there is an easy review process to resolve administrative errors.

Almost a quarter of a million more electors are now on the NSW state roll than the NSW federal roll. The federal parliament is moving to adopt the same system for federal enrolment, in part because of sheer embarrassment at the divergence between the state and federal rolls.

Not getting a majority on 50.6% is a reversal, but something that could feasibly happen in a system using single member plurality without gerrymandering. However, without gerrymandering such a reversal would mean the Democrats falling only a few seats short.

Had the congressional Democrats gotten 52% of the national popular vote, the gerrymandering would not have cost them the majority, as evidenced by Obama winning a majority of the districts (and Obama’s performance indicates getting to 52% was and is feasible), it just would have been a much narrow majority than normal for that sort of percentage.

Its fair to note that in U.S. politics, votes for the congressional party usually lag votes for the winning presidential candidate, in fact in 2012 there was a smaller gap than usual. Also, Carter and Gore failed to carry a majority of the congressional districts in their elections despite winning the national popular vote, so there is a small anti-Democratic bias with using single member districts.

Also, the Republicans “won” redistricting for this decade, but their gains are still limited by the fact that its done state-by-state, so the redistricting winner’s gain is limited by states where the minority party is able to gerrymander, states with bi-partisan, judge-drawn, and commission drawn lines, and various instances where other political circumstances keep parties from exploiting this tactic. If an American political party ever is able to do this on a nationwide scale the effect would be formidable, though hopefully so would the backlash.