Hillsborough County Commission and the Voting Rights Act

The Hillsborough County Commission currently has seven members, four of which are elected from districts, three elected at-large by the entire county. This is not a new arrangement, and the hybrid system of districts/at-large is not unique – in fact, my home city of Gainesville has the same four district/three at-large setup for their commission. A map of the districts is below (click for a larger view).


The county has seen racial change over the past few decades, especially when it comes to the Hispanic population. In the 1990 Census, 13% of the county was Hispanic, in 2000, 18%, and in the most recent Census, Hispanics made up 25% of the county. However, there is no majority-Hispanic district in the county. Instead, District 3 is a joint Hispanic/African-American majority district, with a voting-age population of 35.5% African-American and 23.4% Hispanic; this district currently has an African-American representative, and there is no Hispanic representative currently on the commission.

The Hispanic VAPs of the four districts in the map, in order, are 30.3%, 18.6%, 23.4%, and 19.4%. Is the reason that there is no Hispanic majority-minority district because the population is spread out across the county? Not really. The zoom-in of the map presented below has over half the Hispanic population in it (red blocks – more Hispanic, green blocks – less Hispanic, district lines in black). It turns out that the largest Hispanic area just happens to be where three districts cut their boundaries.


In any case, this map was pre-cleared by the Department of Justice under Section 5 of the Voting Rights Act, as was required at the time of passage. However, Section 5 is primarily concerned with retrogression – does the new map make minorities worse off than they were under the previous map? And indeed, a similar hybrid district was in place in the previous decade.

Section 2 of the Voting Rights Act, on the other hand, requires something slightly different. The case law surrounding the section is complex, but the essential takeaway is that if a reasonably compact district can be drawn where a racial or ethnic minority makes up a majority of the voting-age population, that district should be drawn. Under a four-district plan, however, both an African-American majority district and a Hispanic majority district is not possible – given their geographic distribution, one would lose out to the other.

But that’s assuming a four-district map is required. If the commission moved to all seats being elected from districts instead of three being elected at-large, is it possible to have a map with a majority-minority district each for African-Americans and for Hispanics? If so, there’s a parallel to 2006’s United States v. City of Euclid out of Ohio, where a similar hybrid four district/five at-large system was found to dilute the votes of African-Americans and not allow them to elect the candidates of their choosing (since then, they’ve moved to a system of eight districts and one at-large City Council President).

The answer turns out to be yes. I’ve drafted one example, which is pictured below (click for a larger view).


For calculating the African-American voting-age population, the Florida state legislature chooses to include those who not only respond “Black alone” for their race on the Census, but Black with any combination of other races, which has the effect of producing larger figures. However, the county’s official statistics for their maps use just Black alone, so I’ve done the same here. District 1 has a voting-age population of 50.1% African-American, while District 2 has a VAP of 50.4% Hispanic.

While Districts 1 and 2 are not especially compact, neither is District 3 in the current map, and they certainly beat District 5, for example, in the state’s current congressional map. Furthermore, the average compactness across all districts (using the relatively standard Polsby-Popper measure) is actually slightly better in the seven-district map than the four-district one currently in place – 0.300 versus 0.282, where higher is better. Finally, the population deviation in the seven-district map is lower than the four-district current map, at 7.5% versus 9.2%.

Thus, it seems pretty clear that Hispanic voters are losing out under the status quo, and that there are realistic alternatives where they do not. Whether or not this reaches a point of being illegal is another question, though, and a group with the resources to challenge the map would have to be the ones to push it.

Leave a Reply

Your email address will not be published. Required fields are marked *