How To Fix Gerrymandering
... with Federal Guidelines
-- David Swink, FCTA board member / 2025-11-08
Updated 2025-12-01
A few months ago, I opined that, since the 50 states' US Congressional apportionments closely match their state legislative chambers, US representation seemed reasonably 'fair' in spite of gerrymandering of many states' US Congressional districts. I failed to see that the correlation was because local state districts were also highly gerrymandered, giving a false illusion of fairness. But now, as red and blue states try to outdo each other in gerrymandering their US districts to retain power in Congress, the gerrymandering issue is a serious concern.
In 1964, US Supreme Court Justice Potter Stewart described his threshold test for 'obscenity' thusly: "I know it when I see it". That phrase similarly has seemed to describe 'gerrymandering'. However, we now have ways of objectively measuring 'compactness' of an area -- the antithesis of gerrymandering, which uses extreme 'un-compactness' to draw district boundaries in a way to gain political advantage.
The Polsby-Popper (PP) test is a simple algorithm used to numerically rate the 'compactness' of an area relative to the length of its circumference or outer boundary -- (A*4pi/C**2). The PP test rates a circle at 1.0; a square at 0.78; and a 1-by-2 rectangle at 0.70 -- the latter of which would seem to most people to be near the outer limit of 'fairness'. The same test would index many gerrymandered districts at close to zero. So every current political district in each state could retroactively and objectively be rated for 'fairness', though one would be hard-pressed to find such numbers posted anywhere. But just imagine how voters might react if such numbers were made available to them! -- Talk about 'obscene!'
The Federal government currently leaves it to the states to draw US Congressional district boundaries as the states see fit -- which leads to such acts of gerrymandering as we see today. But while the Feds cannot tell the states how to draw their local state legislative districts, the Feds have every right to place guidelines on how state apportionments for US office are implemented. Here is how an Anti-Gerrymandering Bill might be written:
This Bill applies to states with two or more US Congressional districts.
Each US district redrawn by a state shall contain, within 1%, the same number of constituents as every other US district redrawn in that state.
States with uncompact external boundary protrusions (panhandles, peninsulas) and/or uncompact inner boundaries (mountain ranges, navigable rivers) ... shall draw those 'peripheral' US districts first -- with a clean merger into the 'main body' -- before then drawing remaining 'main body' districts in a compact manner.
All US 'main body' districts shall be drawn with a Polsby-Popper rating not less than [perhaps 0.60, thus allowing each such district to be drawn utilizing existing but less-efficient county boundaries].
Each multi-districted state shall make their total redrawn US district compactness ratings freely available, and subject to the Federal government for final approval.
If a state fails to implement these guidelines for US Congressional districts, the Congress shall recognize no more than half the total elected Congressmen for either political party for that state.
Item #4 at first looks too simple to be effective. It's true that a lone district in the 'main body' void -- even a perfect circle -- could be maneuvered to favor one group over another. But that void must be filled with the remaining allocated districts, and so it becomes an entropy problem. Like soap bubbles across a small surface area, they all must cooperate by collective compactness to reach a lower energy level. And all resulting 'main body' districts will rate above whatever minimum compactness index is specified in #4 -- no gerrymandering allowed.
To summarize, the Federal government has the power to quantitatively
identify political gerrymandering and demand corrections in how
US Congressional districts are drawn, thus restoring a degree of trust
in our Congressional elections. But even without a Bill to address
gerrymandering, simply measuring and posting current PP ratings for
all 435 Congressional districts would go a long way toward eventually
making gerrymandering unacceptable.
Update: The Census Bureau provides current area data for each Congressional district (in square meters) in their "boundary" files, but omits perimeter lengths for same -- the web site even carefully warning NOT to imply any such measurement. So one must look elsewhere for perimeter information in order to calculate Polsby-Potter.