I noted a couple days ago that Maryland’s Sixth Congressional District is unconstitutional under the Benisek plaintiffs’ district-specific test, while Maryland’s congressional map is also (probably) unlawful under the Whitford plaintiffs’ statewide test. (The Benisek test requires discriminatory intent with respect to a particular district, as well as a discriminatory effect in the form of that district flipping from the opposing party to the line-drawing party. The Whitford test requires discriminatory intent with respect to a plan as a whole, a discriminatory effect in the form of a large and durable partisan asymmetry, and a lack of a legitimate justification for this asymmetry.)
Generalizing a bit, the Benisek and Whitford theories can be expected to point in the same direction when (1) the previous plan was symmetric (or favored the opposing party); (2) the current plan is asymmetric in the direction of the line-drawing party; and (3) the current plan’s asymmetry was achieved exclusively by flipping districts from the opposing party to the line-drawing party. In these circumstances, there is both discriminatory intent (the aim to flip at least one district) and a discriminatory effect (the actual flipping of at least one district) under the Benisek test. All of the elements of the Whitford test are satisfied as well: a partisan asymmetry that is deliberate, severe, persistent, and unjustified.
However, when these criteria are not satisfied, the Benisek and Whitford approaches can be expected to diverge. Start with the symmetry of the previous plan. What if it was already highly skewed in favor of the line-drawing party? What if, for example, Maryland Democrats already controlled seven out of eight congressional seats throughout the 2000s? Then there would be no liability under the Benisek test, because Maryland Democrats could design a highly asymmetric map without flipping any districts from Republican to Democratic control. All they would have to do is maintain their hold on their seven existing districts. Conversely, there would be liability under the Whitford test, because the new map would be intentionally, significantly, durably, and unjustifiably asymmetric. That it happened to be as asymmetric as its predecessor would be legally irrelevant.
Next, consider the asymmetry of the current plan. What if it’s not skewed in favor of the line-drawing party? What if, say, Maryland Democrats previously controlled three out of eight congressional seats, and then designed a new map that would enable them to win four seats? That Maryland Democrats are going from three to four seats (not from six to seven) would make no difference under the Benisek test. There would still be the intent to flip a seat, as well as a seat actually flipped, and hence liability. On the other hand, the current map’s overall fairness would make a dispositive difference under the Whitford test. A plaintiff would not be able to show that the plan is skewed toward the line-drawing party if it actually treats the major parties symmetrically.
Lastly, take scenarios where the current plan’s asymmetry was achieved in more complicated ways. Assume, for instance, that Maryland Democrats went from six to seven congressional seats not just by flipping the Sixth District from R to D, but also by flipping the First District in the same direction and by enabling Republicans to win a seat that was previously held by Democrats. Then there would be three viable claims under the Benisek test: one by Republicans in the Sixth District, another by Republicans in the First District, and yet another by Democrats in the seat that was deliberately flipped from Democratic to Republican control. Under the Whitford test, in contrast, there would be just one claim and by just one party’s supporters. Republican voters throughout Maryland could contend that the map as a whole is biased against them.
The point of these examples (which could be multiplied many times over) is that the Benisek and Whitford theories are not equivalent in their scope. Sometimes (as in Benisek itself) the two tests do converge. But in several other situations—whenever flipped districts are not tantamount to an asymmetric map—the two approaches yield different conclusions.