Republicans probably just lost at least 4 or 5 seats in the House.

This afternoon, the Pennsylvania Supreme Court unveiled its remedial Congressional map for the midterm elections in November. [See Featured Image] The Supreme Court Order imposing the map is here.

According to NYT data expert Nate Cohn, the new districts are arguably more advantageous to Democrats than the ones Democratic lawmakers themselves proposed a week ago. The map was actually drawn by Nathan Persily, a Stanford professor who is frequently retained by courts to “remedy” alleged gerrymandering.

Recall that on January 22, the Pennsylvania Supreme Court ruled—on a party-line vote—that the districts adopted by the legislature in 2011 intentionally “diluted” the votes of Democrats, in violation of the state constitution’s guarantee of “free and equal” elections.  The court commissioned a new map, ordering that districts be “compact and contiguous” and avoid dividing localities. The Republican legislature was given three weeks to devise such a map and get the Democratic governor to sign off on it. If no agreement was reached, the court promised that it would adopt its own map by February 19.

The state Speaker of the House and president pro tempore of the Senate asked the U.S. Supreme Court to intervene. It declined to do so, with Justice Alito turning down a stay application without comment or referring it to the full court.

Chastened, the Republicans in the state legislature proposed new districts, and even though no one denied that their map conformed to the court’s neutral geographic indices, analysts determined that Republicans still intended for the map to help their party. Governor Tom Wolf rejected the proposal.

The Cook Political Report’s Dave Wasserman explains that the court’s remedial map intentionally compensates for Democratic clustering in urban precincts.

Other commentators pointed out that many of the judgment calls the court made seemed to defy its own criteria.

This is to say that the court’s map does not simply remedy defects in the old districts, but goes further by ensuring that Pennsylvania’s House delegation more closely represents the statewide vote. Such compensation is not necessarily “unfair”, and it would probably be an exaggeration to describe this as a “Democratic gerrymander.”

Indeed, in a typical year, Republicans will still hold a 10 to 8 advantage. In this national climate, Democrats are expected to win at least 9 seats. Admittedly, this division hardly seems unreasonable in a swing state like Pennsylvania, and it certainly seems fairer than a 13–5 split in a state where parties almost never win more than 60 percent of the vote.

But this outcome is deeply problematic nonetheless. If the court faithfully applied its own order and only redrew cartoonishly-contorted (literally) districts like PA-07, Republicans, would, most likely, still have had the upper hand in a state like Pennsylvania, where Democrats are concentrated in Philadelphia and Pittsburgh. As the NYT explains:

In general, partisan balance is not usually a goal when redistricting. You could certainly argue that partisan balance and maximizing the number of competitive districts should be among the criteria, but, in general, they are not. Instead, a nonpartisan map usually means a partisan-blind map. It strives for compact districts that respect communities of interest, with little regard for the partisan outcome.

A decision to pursue partisan balance in Pennsylvania is particularly significant because Democrats are at a clear geographic disadvantage. They waste a lopsided number of votes in heavily Democratic Philadelphia and Pittsburgh; the Republicans don’t waste as many votes in their best areas, and so the rest of the state (and therefore its districts) leans Republican. As a result, a partisan-blind map will tend to favor the Republicans by a notable amount.

No one, to my knowledge, is arguing that such a natural geographic edge is illegal. Therefore, it’s not the court’s place to eliminate this advantage in the course of adjudicating a constitutional dispute.

But I’m sure the judges figure that, when confronted with a choice between constitutional government and “what seems fair,” the voters will pick “what seems fair.”

Here are the old and new maps:

https://en.wikipedia.org/wiki/Pennsylvania%27s_congressional_districts#/media/File:Pennsylvania_Congressional_Districts,_113th_Congress.tif

http://www.wtae.com/article/pennsylvania-supreme-court-issues-new-congressional-district-map/18238727?src=app

Indeed, if it’s really unconstitutional to have an unrepresentative delegation, then we should simply do away with districts and apportion seats on a statewide, at-large basis. That would be a lot more sensible— and look a lot less partisan—than PA-01 jutting into Montgomery County. Or northern PA-06 stretching around Reading.

It’s unclear how the legislature will proceed. This is a difficult PR battle to win—how do you explain to the average voter that, because of demographic patterns, a map drawn for compactness and non-divisibility will generally advantage Republicans, unless you actively make sure it doesn’t?

Expect federal court challenges by Republicans arguing state legislatures, not courts, have exclusive authority over congressional redistricting. But since Justice Alito already rejected such a plea, it doesn’t seem likely a federal court will get involved.

This post has been updated.