Should our elections offer fine-grained representation of geographic units, or fine-grained representation of ideas?

If we wanted fine-grained representation of both land and ideas we would need to elect 50,000 people to the legislature, which is not practical.

I saw this question on Twitter: What would it take for Social Democrats to win in the USA?

My thoughts:

Even in Europeans countries it has been fairly rare, since 1945, for Social Democrats to gain absolute majorities. Typically, they rule as part of a coalition, often with other left-wing parties. In Germany, about 75% of the time that the SPD has been in power, it has been in a coalition and, one of the odd things about politics in Germany, much of that time the coalition has been with the moderate right-wing conservative party (the CDU).

Even in countries such as Sweden, and with groups that have urgent needs for resources that the government can provide, it takes both strong institutions (such as labor unions) and some decades of participation, before the right kind of broad political coalition can be put together. This describes Sweden in the early 1900s:

In line with the existing literature, we show that electoral socialism depended on both the extension of the suffrage to its ‘natural’ electorate, that is, the urban working class, and the organizational capacity of trade unions and other civic associations. In addition, we show that socialist support was not uniform within the working class – even for a highly homogeneous society. Instead, the social-democratic vote was initially stronger among low- to middle-income workers, only expanding to poor voters later in time.

And that was in a multi-party system that allowed the creation of flexible coalitions.

My point is, a 2 party system like in the USA makes Social Democracy almost impossible. The closest the USA has come was the broad coalition built in 1932, which included Southern conservative racists and also some very progressive left-leaning activists, including life-long Socialists such as Eugene Debs and Harry Hopkins.

So how could Social Democrats win in the USA?

The first step has to be to change the voting systems to get away from the 2 party system. I think making changes at the Federal level is hopeless, so things would have to start at the state level. And rather than simply imitating the systems which exist in Europe, which are also known to have flaws, we should push to build new forms of democracy that incorporate all that Political Science has learned over the last 80 years, to build new, creative, robust forms of democracy that will be the envy of the world.

So the next three steps are:

1. deciding what a better version of democracy looks like

2. picking a state that might be open to change (Vermont? Massachusetts? Washington?)

3. convincing the public that the new version of democracy will bring real benefits (or at least solves some of the existing problems, such as gerrymandering in the USA and over-representation of rural areas in much of Europe)

The systems of democracy that we have now, in the USA and in Europe, grew out of the European Enlightenment and continue to embody ideals from 3 centuries ago, and as such they tend to establish a center of gravity that favors property owners, such that progressives need to put together overwhelming majorities to gain mandates for big changes, while by contrast conservative parties, backed by corporate power, can push through big changes while holding narrow majorities.

In the USA, some progressives have recently committed to Rank Choice Voting. They feel it will fix some of the problems in the system. To be clear, the only reason to chose Rank Choice Voting, over a run-off election, is to save money. Rank Choice Voting produces the same result as holding a run-off election, but it saves the tax payers some money because the country has just one election instead of several elections. Rank Choice Voting means that when you have an election, you ask the voters up front, "Hey, if we need to have a run-off, who would be your choice?" and then the voter specifies their choice, given all the possible two-way competitions that might result from the multiple candidates who are currently running for office.

A side note: We should consider the possibility that having a run-off election, or multiple run-off elections, would produce different results than Rank Choice Voting, because presumably the candidates would hold more debates, and if the voters first choice was knocked out of the competition, those voters would now be considering some candidates whom they had never seriously thought about before. But for the sake of this essay, lets assume that Rank Choice Voting and multiple run-off elections are basically the same, and both have the same problem.

What is the problem?

The problem is rank of any kind. Back in 1953 the economist Kenneth Arrow was able to prove this mathematically in his Impossibility Theorem. He later won the Nobel Prize, in part for this work. He showed that rank voting will tend to fail to express the true preferences of the voters. He suggested that approval voting would do a better job of aggregating the real preferences of voters. Here is a simple summary of his work on voting:

https://democracychronicles.org/kenneth-arrow/

In approval voting, the voters get an infinite number of votes, and the only finite limit is on how many legislators get elected.

Kenneth Arrow provides some formal math for us to think about the problem of voting, but I'd like to try to talk about this issue using plain English. I think the basic issue is that most of the forms of voting that we current use are too coarse-grained to actually capture voter preferences. We should want to use fine-grained systems of voting, to better capture the actual preferences of voters. Parliamentary systems don't actually solve the coarse-grained problem. Having 5 parties is better than having 2 parties, but it is still coarse-grained, and it still mostly fails under Kenneth Arrow's Impossibility Theorem.

I'd like to see the Left take the leading role in looking at what Political Science has learned over the last 80 years and then using that to fashion a new kind of democracy that breaks with the traditions that we inherited from the 1700s. Let us instead create a new kind of democracy that's really robust.

We need to break with the idea of a legislature that is composed of representatives who each represent a small geographic area. Some parliamentary systems already do this, they have pure party systems at the national level, with no geographic representation (Israel, for instance). Adding in segmentation by geography makes some kind of rank voting almost mandatory, because if you have 500 geographic units, you can really only have one representative from each of those geographic units. If you have 500 geographic units, then you have fine-grained representation of land, but you have coarse-grained representation of ideas. If we want to have fine-grained representation of ideas, then we have to have coarse-grained representation of land (again, as in Israel). Because otherwise, if we have, say for instance, 100 representatives each from 500 geographic regions, we end up with a legislature with 50,000 people in it, and I think we can all agree that such a large legislature would be unreasonable. (There is a counter-argument that I am sympathetic to, that we would benefit from an extra level of voting, that is, the public should elect a pre-legislature (a massive mass legislature) who can then elect the real legislature.)

In the USA, geographic units are a primary division of politics. The House of Representatives divides America into 435 oddly shaped geographic units, with each geographic unit having 1 representative. By contrast, in Israel there are no geographic divisions represented at the national level, instead the whole of Israel votes for the whole of the legislature, and many parties compete for votes. In the last election there were 39 different political parties that registered to compete in the Israeli election (for a legislature that only has 120 seats!). So Israel has some fine-grained competition of ideas, but it has coarse-grained representation of geography (there is only one geographic unit in the election -- the whole of Israel).

A large country will have some regional differences of culture and language. How much should those differences be modeled by having explicit geographic regions, versus modeling those differences by having parties that represent those cultural differences? Some people feel this is important. In the early days, the USA had significant regional differences, but that has lessened with time. Still, regional differences are very important in a place like Spain. But does Spain benefit from having rigid and explicit geographic boundaries around cultural areas? The main problem with that approach is that people in the modern world, looking for jobs, leave those areas and move to a few big cities. If someone from Basque Country moves to Madrid looking for work, but they still want to vote for Basque culture, wouldn't it be better to have a national party that fights for Basque culture, rather than trying to limit the voting for Basque culture to a fixed geographic area? I think the idea of geographic representation is mostly a holdover from the 1700s. Back then, people were attached to the land, it was an agricultural society. Nowadays, people move big distances looking for jobs, and these people end up in multicultural urban areas surrounded by people from dozens of other cultures -- and obviously one cannot chop up the city to give geographic division to those cultural differences. What if you end up with an apartment building that has tenants from Thailand, Sweden and Nigeria? Do you draw the boundaries of the voting districts so that each apartment in the building is actually in a different district? I think we can agree that such an effort would be ridiculous. So in our modern and multicultural world it is probably better to treat all national elections as being about ideas, with the whole of the public voting in a unified way for the legislature. That is, I favor fine-grained representation of ideas, rather than fine-grained representation of geography. If there is a cultural factor that would cause all Italian people to vote one way, while all Japanese people vote a different way, let those cultural differences express themselves in a national election for a national legislature, with fine-grained representation of each group's different cultural beliefs. We don't need fine-grained representation of geographic units to achieve this fine-grained representation of people's belief, indeed, the opposite is true: a fine-grained representation of geographic units makes it difficult to have a fine-grained representation of ideas and beliefs.

I do think it should be possible to create a new version of democracy that is really exciting to people. Perhaps we need to start in a single state, such as Vermont or Washington. And if the experiment is successful in one state, then we can use that as a springboard to advocate for its adoption elsewhere.