What to expect from an 18th-century democracy
Saturday I was walking through a neighborhood a few kilometers from my home when I passed a house with an almost-unheard-of collection of political signs. They had a sign for a candidate in every office up for election this year, half of which were affiliated with one of the major political parties and the other half of which were affiliated with the other major political party. Beside this mixed-party signage with another sign which read “End Single-Party Rule”.
I sympathize with this unusual signage. I believe everyone is best served by consensus governance, by policies and laws that people of differing outlooks can agree are worth supporting. But I also believe that the oscillation of power back and forth between two parties is an all-but-inevitable outcome of the way in which we tally votes.
The form of voting used in almost every voting body in this country is formally called “first past the post” (FPP). The name comes from the analogy of a competitive foot race: it doesn’t matter how fast you are or how much distance is between you and your competitors, only that you past the end post before any of them. FPP elections are characterized by there being one race (votes cast) and one winner (getting the most votes).
FPP votes result in winners with less than half the votes, meaning outcomes most voters dislike, unless there were only two options to begin with. For propositions with a yes/no vote FPP works fine, but for electing representatives it is fundamentally broken. But people find a way to make do with what they have, and the way we make do with FPP elections for representatives is with coalitions.
In a coalition-backed FPP voting system you an I get together before the election and agree to work together. I may prefer one candidate and you another, but we both agree that our first-choice candidates have no chance of winning and there’s a third candidate that we’re both OK with and that we can probably convince a lot of our friends to vote for too. We thus form a coalition, pledging to both vote for that compromise candidate instead of each voting for separate first-choice candidates
Once one coalition forms, others form too in order to get enough backers behind a single candidate to bet the other coalitions. Provided the population is sufficiently homogeneous without subpopulations that refuse to ever coalesce, this continues until there are each coalition has contains half the voters.
A two-party system is the only stable option here. A coalition with less than half cannot win on its own, so it must court additional members to have its views represented. A coalition with more than half either ignores some views on its fringes, giving ground to the other coalition; or forms internal coalitions, the smaller of which eventually get frustrated and leave the coalition.
Two-party systems are the steady state of FPP voting for representative democracy, but is itself unstable both in the platforms of the parties and the balance of power between them.
Parties differ from a simpler coalition in that, while voters generally understand that a vote for neither party is a vote wasted, they do not have consensus buy-in on the parties platforms. This leads to two types of parties: those whose platforms and candidates are dictated by a power base who are able to provide some guarantee of voters (such as organized crime, military dictators, media moguls, and dogmatic religions) and those whose platforms are selected by a jostling for voters.
Jostling for voters involves mostly small movements, but also at least two sources of sudden shift. Sometimes a single issue arises around which people realign their party allegiances. And sometimes a single persuasive voice arises which pulls voters regardless of platform. This latter process was particularly acute in ancient Greece, leading Aristotle to state that democracy always devolved to rule by demagogue; we’ve not seen it quite that extreme in modern times, but there are plenty of “party of name”-type examples where a single demagogue redefined a party platform.
But platform change is by far secondary to the oscillating power that two-party democracy almost inevitably creates.
Many issues of governance fall on continua. What’s the right balance between an unfunded government (i.e. anarchy) and a government that controls all assets (i.e. totalitarianism)? How harsh should punishment be? How should policy balance property rights and helping the disadvantaged? And so on.
Because these are continua, views of voters lie at all points along them. To help form coalitions, parties tend to draw a line somewhere in the middle where each party angles for half of the voters. And then each party picks a candidate that will appeal to most of its members; that is, somewhere in the middle of its half of each continuum.
Result: the views of the median voter are almost never reflected in the elected official. If we model opinions on some issue with numbers between 0 and 100, the average person lies somewhere around 50 but most representatives are either around 25 or 75: the average of their half of the position. Sometimes parties do some amount of moving toward “electability” and move a little closer to the average voter’s view, but a candidate that actually represents the average voter is on the fringe of both parties and highly unlikely to be selected by either.
This leads to an expected state where whichever party has control of government is doing things too extreme for most voters. The longer that lasts, the more voters lean toward voting against the too-extreme actions instead of for the party that comes closer to approximating their views until in a coming election power swaps. But after the swap policies are too far the other way, so it will swap again, and again, and again…
One of the side effects of frequent power swaps is an ever-increasing heap of laws as each newly re-empowered party seeks to add rules that will mitigate the impact of the last party’s rules; but that’s a topic out of scope for this post.
If we decide on the first past the post voting system (where each voter casts one vote and the candidate with the most votes wins) then coalitions will form and gravitate toward two parties each with roughly half the voters.
The two-party 50/50 split is a structural outcome of the voting system and survives interruptions by platform-defining issues and charismatic demagogues.
Two-party systems provide candidates who represent the 25th and 75th percentile of opinion, not the 50th percentile. Thus, most people feel their representatives don’t represent them and wish laws were closer to the views of the out-of-power party.
A force pulling toward the middle coupled with a system that prevents people in the middle from gaining power results in oscillations as people get fed up with the current extreme and vote for the alternative extreme.
And these forces are all present regardless of the specific people and platforms: if the population is sufficiently homogeneous to represent a continuum of views and if FPP voting is used, this power oscillation will emerge.
Looking for comments…