tl;dr: Yes, this is another post promoting ranked choice voting.

Like other registered voters in Howard County, I recently received my postal ballot for the 2020 presidential primary. Some of the decisions I have to make are easy (yes, I’ll vote for the presumptive Democratic nominee for President) while others are hard (gee, there are a lot of Democratic candidates for the US House of Representatives, and I have no idea who most of them are). And then there’s the question of who I should vote for in the Howard County Board of Education election.

In recent days there have been posts on the Board of Education race by Jenny Solpietro and Jason Booms, both progressive Democrats, commenting directly or indirectly on the lack of BoE candidates in District 5 that match their own political views. If you’re a progressive Democrat in District 5 then presumably no one you like will end up representing you on the BoE. This is reminiscent of past complaints by conservative Republicans living in (say) County Council District 2 that there was little to no chance of their being represented by someone who shared their own views.

These complaints support my firm opinion that electing Board of Education members (or County Council members) by district makes no sense. Why, do you ask?

First, it causes the problem we see in the current Board of Education race and past County Council races: if your views are at odds with the majority of people in your district, you don’t have a realistic chance of electing someone who reflects your views.

Second, in my opinion electing members by district does not do a very good job of what it was originally justified as doing, namely giving people more of a say in what happens in their “local” schools, i.e., schools in their part of the county. That’s because the district boundaries are arbitrary divisions.

For example, in past times the boundary line between District 1 and District 5 ran down the middle of the street in front of my house. If someone moved across the street to a different Board of Education district, would their concerns as a voter, a taxpayer, and a current or future parent of school-age children change in any significant way?

No, because they’d still be Howard County residents and taxpayers, and they’d still be dealing with a county-wide school system with a single budget, curriculum, and set of policies. The schools that their children attended (or would attend in future) probably also wouldn’t change, since the Board of Education districts are based on County Council districts and don’t necessarily have any relation at all to how HCPSS draws its own district boundaries.

Third, electing BoE members by district ignores the other ways in which voters might differ in their interests, beyond just the part of the county they happen to live in. For example, suppose that you have an interest in seeing more advanced math and science classes, or more funding for arts programs, or increased choices for vocational education, or you have concerns about disorderly classroom environments, or want more attention paid to issues around social justice and equity, or in general have strong opinions concerning any of a host of other issues that might arise in running a large public school system.

Electing members by districts may do little or nothing to ensure that your views are represented on the Board of Education: while there may be a lot of people across the county who share your views, there may be few like-minded individuals within the district you happen to live in.

So, what would be a better approach? The first step would be to abandon election by districts as a well intentioned approach that unfortunately does not address the true problems of representing the views of a diverse county population, and go back to electing all Board of Education members on a county-wide basis. (Under the current scheme only two at-large members are elected county-wide.)

But we can’t stop there. Using the traditional way of electing Board of Education members has its own set of problems. If we simply count the votes for each candidate and elect the five (or seven, or whatever) people with the most number of votes, then we run the risk of having minority views not be represented at all:

If 51% of voters think one way, and 49% of voters another way, the 51% of voters will vote for their preferred candidates and the 49% will vote for their (different) candidates. The result will be that the 51% of voters will get all of their candidates elected, and the other 49% will have no one on the Board of Education representing their views.

(Incidentally, the roles of the majority and the minority can vary. In the past this approach of electing the top set of candidates was used in many US jurisdictions as a way to systematically suppress the votes of African Americans—who would always be a minority in a city- or county-wide election—and elect all-white city or county councils. As a different example, in Howard County in the 1970s electing county council members at large resulted in Democratic dominance of the council, with no Republicans at all elected to the council at one point.)

So what’s the alternative? The alternative is to combine county-wide election of Board of Education members with a voting scheme that better takes into account the full range of voters’ interests, by allowing them to rank candidates in order of preference. I’ve previously written about this “ranked choice” scheme in the context of Howard County Council elections, proposing it to the Charter Review Board and then addressing various follow-up questions about it.

I want to particularly focus on how ranked choice voting can actually do a better job of providing “local” representation on the Board of Education than election by districts:

Suppose you live in (say) Elkridge and think that that area of the county has been neglected by the Board of Education and the Howard County Public School System. Even though in the scheme I’ve proposed the candidates would be running county-wide, nothing would prevent a candidate from running as the “champion of northeast Howard”. This candidate could then be elected as follows:

First, people living in that area could designate that candidate as their first preference. Note that in doing so it wouldn’t matter exactly where these voters live: there would be no artificial district boundary dividing Elkridge from Ellicott City, for example, such if you live just on the Ellicott City side of the boundary line then you don’t get to vote for the “northeast Howard candidate”. This increases the size of the voter pool that the candidate can attract votes from.

Second, it’s possible that other people elsewhere in Howard County might think that northeast Howard needs more attention by the BoE, even if they don’t see it as the most important issue. Those people could vote for the candidate that best represents their own perceived interests, designating them as their first preference, and still vote for the “northeast Howard candidate” as their second preference. If enough people do so, that candidate could be elected even if the number of their supporters in Elkridge or Savage might not be enough to get them elected just on “first preference” votes.

What about the concern that parents wouldn’t know who on the Board of Education to contact if they had concerns about their local schools? That’s easily solved: just have the Board of Education itself assign members to different parts of the county, to serve as the official liasons to parents and students in those areas.

BoE members would naturally nominate themselves to serve those areas that are their natural constituencies, as in the “champion of northeast Howard” example above. And because the Board of Education would be doing this assignment as a matter of board procedure, and not in strict accordance with state law, it could assign members to areas that would be as closely aligned as possible to the actual school catchment areas, and not have to go back to the Maryland legislature in the event that those boundary lines change.

What about the concern that it would be more expensive to campaign county-wide, and that election by districts makes it easier for “local” candidates to run in their own districts without having to raise a lot of money? I think what we are seeing now is at odds with that ideal: since the number of voters in each district is relatively small, and the turnout for a primary is going to be relatively low, it’s possible for wealthy donors inside or (as Julia McCready has complained) outside a district to have an outsized influence in that district’s election.

I’ll close by noting that even though the Board of Education elections have in the eyes of many become more politicized and partisan, adopting a county-wide ranked choice voting scheme for the BoE or County Council should not be a point of partisan contention. In the context of this year’s Board of Education election ranked choice voting would help progressive Democrats who live in western Howard. In the context of the 2022 County Council election ranked choice voting would help conservative Republicans who live in eastern Howard.

That’s because ranked choice voting is designed to better reflect voters’ preferences in general, and to do a better job of ensuring that preferences held only by a minority of voters can nonetheless lead to those voters having representation on elected bodies like the Board of Education or County Council. Who will be in the minority will vary by time, place, and the issues involved, so ranked choice voting offers something for everyone regardless of their particular political views.

Further exploration

For more on this topic see the following: