Trebbi et al. (2008) have observed that (in the US):

Majorities tend to disenfranchise minorities through strategic manipulation of electoral rules. With the aim of explaining changes in electoral rules adopted by U.S. cities, particularly in the South, we show why majorities tend to adopt “winner-take-all” city-wide rules (at-large elections) in response to an increase in the size of the minority when the minority they are facing is relatively small. In this case, for the majority it is more effective to leverage on its sheer size instead of risking conceding representation to voters from minority-elected districts. However, as the minority becomes larger (closer to a fifty-fifty split), the possibility of losing the whole city induces the majority to prefer minority votes to be confined in minority-packed districts. Single-member district rules serve this purpose. We show empirical results consistent with these implications of the model in a novel data set covering U.S. cities and towns from 1930 to 2000.

Is this idea (well, it doesn't seem to have a name, let's call it "Trebbi's law") true in other parts of the world? Or are there substantial counter-examples outside of the US?

I looked at the full text of the paper, but they don't mention any other countries. My own hypothesis is that the level of covert racism (or even just ethnocentrism) in a country/region would be a significant predictor/covariate for where this electorally exclusionary strategy is employed...

  • 4
    There is nothie stopping this principle from applying to a non-racial situation in which the majority (say, a peach-loving majority) seeks to disenfranchise a minority (say, a bicycle-loving minority).
    – phoog
    Commented Aug 19, 2018 at 3:07
  • @phoog: yes, that's true and it has been proposed/found as well (I have a couple of links on that--stay tuned; one in Germany). The main difference seems to be that the presence of minorities is more easily quantifiable externally... whereas most of the other findings involve party share changes. Commented Aug 19, 2018 at 3:10
  • 1
    [continued] Older one world-wide: journals.sagepub.com/doi/10.1111/j.1467-9248.2005.00514.x Commented Aug 19, 2018 at 3:15
  • 3
    You should see what Baltimore has done to exclude GOP voters
    – user9790
    Commented Aug 19, 2018 at 10:19
  • 1
    Unfortunately, I can only read the abstract, The spirit of the question breaks this down as racial where there is nothing in the abstract that points to that and further as has been my experience, a minority race (aka non-white) can have the majority (Corona Queens for example) and disenfranchise other minority races (Koreans for example). The phrase "Particularly in the South" would bias the conclusions since this is a practice that occurs in all cities (as stated in the abstract), or worse, if that means he only looked at southern cities (south of what?) Commented Aug 20, 2018 at 15:44

2 Answers 2


It's definitely not applicable to my home country (the Netherlands) for the simple reason that our electoral system has no provision for it. As we don't have a district based system, the 'tools of the trade' so to speak are simply not available. Gerrymandering is tied one on one to a district system, so no districts, no gerrymandering. Also, since there is no registration to vote (all dutch citizens are registered by virtue of being a citizen and will get their voting pass over the mail, wheter they want it or not) this too can not be used/abused to encourage or discourage people to vote.

Our system of equal representation shows a different outcome: While a sizable chunk of our electorate is opposed to immigration and the growing presence of minorities, they don't use or see the electoral system as a mean to their ends. In stead they support, and vote for, political parties that represent these views. For instance, the dutch 'anti immigration' party, the PVV, has 20 of the 150 seats of parliament and will use their mandate and influence to support their cause, while our most outspoken pro-minorities party (Denk) will use the same means to their ends. In between there is a whole array of parties who take on more moderate positions, both to the left and the right of the spectrum. So while I have no reason to assume that 'covert racism' is any more or less a factor in the Netherlands, the system tends to include all voters to a large extend and does not lead to the phenomenon observed by Trebbi et al. Of course on a local (municipal?) level there are places where minorities are represented to a higher or lesser extend. (minorities tend to be represented more in urban area's, less in rural as a matter of demographics) but this does not effect general elections as our representatives are chosen directly*.

This is not at all saying that the Netherlands does not have their fair share of problems centered around the issue of minorities, or that the existing viewpoints differ fundamentally from those in the US, it is just that our system protects us from this particular evil.

(*) I am very much aware of that our senate (as opposed to our main parliament '2de kamer') is indeed chosen through a district system but I don't want to delve too deep into the Dutch political system here. There is afaik no reason to assume it can be used as described by Trebbi et al, let alone that it is.

Edit: Apparently, it is in fact not reasonable to assume the same applies to countries with a comparable system, as it seems that the Dutch system is unique in having no districts (or a single district for all voters, which amounts to the same thing). Thanks to @MartinTournoij for pointing this out.

  • Even though this doesn't really answer the question, it gives value by showing that there probably aren't that many counterparts to compare the system in the US (with gerrymanderable districts).
    – Communisty
    Commented Sep 18, 2018 at 11:15
  • The Netherlands has a district system for registering for the ballot. That's why some really small parties only appear on the ballot in some provinces. For example 'Jezul Leeft' was only on ghe ballot in a few places for the TK2 election.
    – JJJ
    Commented Sep 18, 2018 at 11:40
  • @Communisty OP asked for counter examples outside the US, which I hope I have provided.
    – Douwe
    Commented Sep 18, 2018 at 11:44
  • @Douwe oh sorry I missed that part. Good post anyways.
    – Communisty
    Commented Sep 18, 2018 at 11:47
  • 1
    According to Wikipedia: Among western democracies, only the Netherlands employs an electoral system with only one (nationwide) voting district for election of national representatives. This virtually precludes gerrymandering, so it seems that NL is the odd one out here. The biggest problem in the US btw isn't so much the distracting system, but that the legislature (i.e. the people you're voting on) are in charge of redistricting, which is a huge conflict of interest.
    – user11249
    Commented Sep 18, 2018 at 13:19

I think Trebbi's assertion is false. In the US, it is common knowledge that rural areas poll heavily conservative while urban areas poll liberal. Rural dwellers are by every definition minorities. Their population is much lower than urban dwellers. However the fact is that the electoral process gives far more representation to rural dwellers so the rural conservative minorities have actually gained power. We have had 2 elections in recent years where the popular vote went to Clinton and Gore yet the electoral college went to their opponents who got significantly fewer popular votes - this is due to the rural minorities ability to override the urban vote.

  • 3
    A somewhat interesting point, but a "rural minority" in nation-wide terms has to be a majority in many states (or at least the deciding factor) otherwise it could not decide anything, even with the two-stage voting system used for the US presidency. What you're talking about is simply the rural voters in the US having more voting power. But whether they qualify as a minority in the sense of Trebbi is pretty debatable. Commented Aug 19, 2018 at 9:03
  • Point of order... the term "voting power" is defined as the probability that a single vote is decisive. This is purely retrospective and in no way entails a special enfranchisement for the person who researchers say has "more voting power". Commented Sep 18, 2018 at 14:42

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .