Most communities select the members of their town or city council by one of three methods:
1) electing one representative per district;
2) electing representatives at-large by plurality vote; or
3) electing them at-large by proportional representation.
Each of these methods has its own pros and cons. So let’s consider them one at a time.
In one-winner-per-district elections, each council member ends up speaking for all the residents of a geographic area. The prime advantages of this arrangement are that each council member can personally interact with a substantial number of his or her constituents; he or she can have a good grasp of his neighborhood’s needs; and he can therefore be a forceful advocate for his neighborhood.
However, since each council member represents the entire district, he or she is speaking for many groups that often disagree with one another, including senior citizens, the middle-aged and the young; business owners, workers and the unemployed; singles, couples, families and one-parent households; liberals, conservatives and most points in between. Each of these groups has its own needs and interests — which frequently collide with other groups’ needs and interests.
So anything a council member says or does on a controversial issue can inflame many groups in his or her district. If a council member proposes detailed solutions to the community’s most difficult issues, many blocs of voters are likely to oppose his stands, perhaps enough to want to remove him from office.
Many council members therefore choose a safer strategy. The safest tactics are to reduce the hard issues to slogans, and to blame the community’s problems on political opponents.
What about voters? One-winner-per-district elections put voters in an equally troubling position. That is, every voter in the district — every senior citizen, young single, middle-aged parent, entrepreneur, blue-collar worker, liberal, conservative, moderate, and so on — has to share the same representative. Every district resident — despite their vast differences and disagreements — has to share the same spokesperson on their town or city council. So the typical voter cannot get a representative who shares his own concerns. As a result, with one-winner-per-district elections, most citizens don’t vote and most who do vote don’t scrutinize the candidates.7 In most cases, the majority simply vote for the name they know best: the incumbent.
Therefore, in one-winner-per-district elections, incumbents win reelection at very high rates regardless of how well they do their jobs.
Why then do so many communities use one-winner per district elections? Largely out of habit. When the United States was forming in the 1700s, 80 percent of Americans were farmers, which meant that neighbors had almost identical concerns. So a representative for a geographic district could understand nearly all of his constituents’ needs. Each representative could therefore explain to his voters how his actions were serving their interests, if in fact his actions were. If voters didn’t find his story convincing, they could boot him out of office.
Today’s one-winner-per-district elections no longer work that way: Voters in each district usually disagree among themselves about what their representative should be doing on the most serious issues. So voters often cannot hold their representative to account in a meaningful way.
What about at-large plurality (ALP) elections? With ALP, every citizen gets as many votes as there are seats on the town council and must cast each of those votes for a different candidate. So if a council has five members, each citizen gets five votes.
The main advantage of ALP is that each council member, instead of focusing on just his or her own neighborhood, is more likely to think about the needs of the whole community. That’s the theory, anyway.
In practice, though, with ALP a substantial number of community residents end up unrepresented. That is, if 51 percent of voters choose the same candidates, those candidates will win all the seats. The other 49 percent of voters would have no effective voice in local government. While that would be an extreme case, with ALP elections, minorities often end up without representation.
Plus, with ALP elections, a voter cannot point to a specific person as his or her representative. So if the town council does a poor job, who does each voter hold responsible? It’s unclear.
Then, there’s proportional representation, in which each political party gets seats in proportion to the number of votes cast for that party. That guarantees minorities a fair share of council seats.
However, under existing forms of proportional representation, each citizen is represented by a bloc of legislators, not an individual. So a typical citizen cannot point to a specific representative whom the voter can hold to account for legislative decisions.
Twenty-two American cities have in fact tried a proportional system known as the “single transferable vote” or STV. And voters in 21 of those cities eventually repealed the new method.
As we see it, STV's prime defect is that a voter does not get a specific council member whom the voter can hold to account for policy decisions. (For more information about STV, click here.)
With personally accountable representation (PAR), however, each voter gets one person as their representative, someone who shares their political outlook, someone they can hold directly responsible for making sound decisions, someone they can easily replace if he or she does not make genuine progress on the major issues.
Of course, PAR has its own drawbacks. In fact economist Kenneth Arrow won a Nobel Prize for proving that no election system is ideal because any system that meets one criterion very well is likely to fall short by other standards. For instance, at-large plurality and one-winner per-district elections produce a more decisive city council than one elected by PAR, because a PAR council includes a wider range of views than a council elected by these other methods. However, at-large-plurality elections effectively disenfranchise a large portion of the community, while one-winner-per-district elections allow incumbents to win reelection repeatedly regardless of what kind of work they do.
PAR, by contrast, gives every council member a strong incentive to find practical solutions to the most pressing problems, a stronger incentive than under any other system. By that standard, PAR is superior to the alternatives.
7 See The Cure for Our Broken Political Process by Sol Erdman & Lawrence Susskind (Potomac Books, 2008).