I worry that this sort of analysis puts process ahead of results.
In large-scale decisionmaking, such as regional or national politics, most voters are confused and inattentive. I think this is inevitable and even proper. The world is too complicated for most people to have informed and thoughtful opinions on most topics. As a result, I don’t particularly care if the process delivers results most voters want. Instead, I care if the process delivers decent results. And in particular, I want decent results for impatient voters and potentially-dishonest election apparatus. First-past-the-post has the important benefit that as a voter I have to indicate one preference, rather than an ordering. This requires strictly less input from me, and therefore probably less attention and thought, which is a Good Thing.
I would be interested to hear an argument for why all the voting theory stuff is useful in practice, given the constraints and goals of practical politics.
We have some examples of cities and countries that use systems other than first-past-the-post. Which of these actually are better governed as a result?
You seem to want some kind of a hidden technocracy where what the (ignorant and confused) voters say they want doesn’t matter much.
This is how most of the First World is run. I think Belgium’s 589 days without an elected government, during which Belgium was not much misgoverned at all, provide a nice natural experiment demonstrating this. To quote the blogger Foseti:
The odd thing about this period of no government was that Belgium’s government was very busy. For example, the non-government nationalized one of the country’s largest banks. Less importantly, your humble blogger had numerous meetings that were attended by representatives of Belgium’s non-government.
It isn’t great but I expect it outperforms popular input. In the third world in particular, experiments with having popular input into actual governance seem to end badly, though I am the first to admit there are important confounding variables.
When it comes to managing cities, the more technocratic the city is the more prosperous it tends to be, now of course correlation isn’t causation, but overall I’d rather live in a prosperous city than a non-technocratic one, and if something can scale for cities with millions of inhabitants, why not small countries?
Most people wouldn’t call the result a democracy.
Except most people do. And the case can be made it is a democracy in the same sense say Sweden is a monarchy.
Kinda but not really. The mandarins / professional bureaucracy aren’t a technocracy to start with and the voters aren’t quite that powerless. But that’s a fairly big discussion, probably for another thread.
if something can scale for cities with millions of inhabitants, why not small countries?
A large subset of political ideas/solutions/proposals suffer precisely from the problem that they scale badly. For example, democracy.
the case can be made it is a democracy in the same sense say Sweden is a monarchy.
Which is not an interesting or meaningful sense at all.
You seem to want some kind of a hidden technocracy where what the (ignorant and confused) voters say they want doesn’t matter much.
I want the voters to have neither too much nor too little influence. I don’t know how to characterize that amount other than by the results, unfortunately.
Which quickly goes to requiring zero input from you.
This is not the historical experience. Britain has had elections that matter, going back many centuries. In their system, the general election ballot is quite simple: “which of party’s candidate do you like?” In contrast, a US general election ballot can have several pages of officials, referenda, and so forth. In California, it’s routine to have 30 separate things to vote on every two years, at just the state and federal levels.
My sense is that Britain is better governed and has a more stable political system, I think in part because the voting avoids over-burdening the electorate. Simple elections don’t seem to degenerate to dictatorship, and more complicated systems don’t produce results I like better.
Wait until I get to explaining SODA; a voting system where you can vote for one and still get better results.
As for comparing different societies: there are of course societies with different electoral systems, and I think some systems do tend to lead to better governance than in the US/UK, but the evidence is weak and VERY confounded. It’s certainly impossible to clearly demonstrate a causal effect; and would be, even assuming such an effect existed and were sizeable. I will talk about this more as I finish this post.
I could just call you a nasty, evil person, but instead I’m going to argue against you on more “rational” grounds.
One of the most important facets of democracy for good decision-making is the feedback loop between electors and elected, and thus between decision-makers and those affected by decisions. Voting is an informational signal from the ruled to the rulers, and legislation is an informational signal from the rulers to the ruled.
Now, what do we know about standard-issue humans trying to make decisions under partial certainty or great uncertainty regarding time-varying and complex pleasure/pain signals to optimize a large system for the expression of a broad system of values (and thus available good/evil stimuli)?
If you answered “they can’t really do it well”, I disagree. My position is: they need swift, sharp feedback in order to do it well. Humans are actually remarkably good at decision-making when given swift, sharp feedback signals.
So what’s the problem with an undemocratic democracy, or a technocracy as such? It shatters the feedback loop that’s signalling:
To the voters: how well their expressed policy preferences effect their expressed values, and how well their expressed values in the voting booth match their actual values as they experience themselves on an everyday basis.
To the government apparatus: what the masses prefer in public policy, how well the government is effecting those preferences, what impact actual enacted policies have on the voting masses, and the stability of the collective whole of society.
No feedback loop means the system as a whole can and will eventual spiral out of control from sheer divergence, as the people in power no longer know or care about the experience of the masses, leaving the masses with no reason to support the system itself, setting the stage for either a mass exodus or a revolt and thus destabilization of the State.
If you answered “they can’t really do it well”, I disagree. My position is: they need swift, sharp feedback in order to do it well. Humans are actually remarkably good at decision-making when given swift, sharp feedback signals.
I agree with this. The place where I don’t follow your analysis is the assumption that voting is the primary means of feedback. We only hold elections every few years. If you want “swift” feedback, elections are not the tool for you.
Moreover, they don’t give very precise feedback in any real democracy. In, say, the UK, the election sends basically two bits of information per voter—LibDem, Tory, Labor, or other. Even in the US, the ballot just doesn’t encode a lot of information about voters; it’s single or double-digit bits per year. And each additional bit per voter is hugely expensive to collect. It costs millions of dollars to pay for all the analysis, campaigning, advocacy and so forth that goes into a national or regional election.
Happily, we don’t rely on elections as our primary mechanism for sending signals to the government. Functioning democracies have an amazingly rich set of mechanisms for getting prompt feedback from the public. We have opinion surveys. We have newspaper editorial pages. We have constituent letters and phone calls. We have donations. We have private meetings between government officials and lobbyists. We also have the whole panoply of economic indicators—politicians do notice when the stock market crashes or when bond rates spike.
I think trying to optimize the information content of the election system is missing the point. Elections aren’t the right tool for preference elicitation. Elections are there to give force to voter preference—they’re not only a measurement tool or even primarily a measurement tool. They’re a civic ritual to demonstrate the presence or lack of voter support for the government and they’re a system to designate a specific alternative in the event that the incumbents are voted out. it’s important that the election produces a clear outcome and an outcome with enough political support to sustain itself until the next election. It’s not important that it extracts the maximum information about the voter preferences on election day.
In conclusion: If you want swift and detailed feedback from the public, run a telephone poll. Rounding up millions of people and asking them about their preferences every few years is a lousy way to measure sentiment.
I agree with this. The place where I don’t follow your analysis is the assumption that voting is the primary means of feedback. We only hold elections every few years. If you want “swift” feedback, elections are not the tool for you.
Moreover, they don’t give very precise feedback in any real democracy. In, say, the UK, the election sends basically two bits of information per voter—LibDem, Tory, Labor, or other. Even in the US, the ballot just doesn’t encode a lot of information about voters; it’s single or double-digit bits per year. And each additional bit per voter is hugely expensive to collect. It costs millions of dollars to pay for all the analysis, campaigning, advocacy and so forth that goes into a national or regional election.
Well yes. This is a very good criticism of our current democracies.
I worry that this sort of analysis puts process ahead of results.
In large-scale decisionmaking, such as regional or national politics, most voters are confused and inattentive. I think this is inevitable and even proper. The world is too complicated for most people to have informed and thoughtful opinions on most topics. As a result, I don’t particularly care if the process delivers results most voters want. Instead, I care if the process delivers decent results. And in particular, I want decent results for impatient voters and potentially-dishonest election apparatus. First-past-the-post has the important benefit that as a voter I have to indicate one preference, rather than an ordering. This requires strictly less input from me, and therefore probably less attention and thought, which is a Good Thing.
I would be interested to hear an argument for why all the voting theory stuff is useful in practice, given the constraints and goals of practical politics.
We have some examples of cities and countries that use systems other than first-past-the-post. Which of these actually are better governed as a result?
You seem to want some kind of a hidden technocracy where what the (ignorant and confused) voters say they want doesn’t matter much.
Which quickly goes to requiring zero input from you.
Most people wouldn’t call the result a democracy.
This is how most of the First World is run. I think Belgium’s 589 days without an elected government, during which Belgium was not much misgoverned at all, provide a nice natural experiment demonstrating this. To quote the blogger Foseti:
It isn’t great but I expect it outperforms popular input. In the third world in particular, experiments with having popular input into actual governance seem to end badly, though I am the first to admit there are important confounding variables.
When it comes to managing cities, the more technocratic the city is the more prosperous it tends to be, now of course correlation isn’t causation, but overall I’d rather live in a prosperous city than a non-technocratic one, and if something can scale for cities with millions of inhabitants, why not small countries?
Except most people do. And the case can be made it is a democracy in the same sense say Sweden is a monarchy.
Kinda but not really. The mandarins / professional bureaucracy aren’t a technocracy to start with and the voters aren’t quite that powerless. But that’s a fairly big discussion, probably for another thread.
A large subset of political ideas/solutions/proposals suffer precisely from the problem that they scale badly. For example, democracy.
Which is not an interesting or meaningful sense at all.
I want the voters to have neither too much nor too little influence. I don’t know how to characterize that amount other than by the results, unfortunately.
This is not the historical experience. Britain has had elections that matter, going back many centuries. In their system, the general election ballot is quite simple: “which of party’s candidate do you like?” In contrast, a US general election ballot can have several pages of officials, referenda, and so forth. In California, it’s routine to have 30 separate things to vote on every two years, at just the state and federal levels.
My sense is that Britain is better governed and has a more stable political system, I think in part because the voting avoids over-burdening the electorate. Simple elections don’t seem to degenerate to dictatorship, and more complicated systems don’t produce results I like better.
Wait until I get to explaining SODA; a voting system where you can vote for one and still get better results.
As for comparing different societies: there are of course societies with different electoral systems, and I think some systems do tend to lead to better governance than in the US/UK, but the evidence is weak and VERY confounded. It’s certainly impossible to clearly demonstrate a causal effect; and would be, even assuming such an effect existed and were sizeable. I will talk about this more as I finish this post.
I could just call you a nasty, evil person, but instead I’m going to argue against you on more “rational” grounds.
One of the most important facets of democracy for good decision-making is the feedback loop between electors and elected, and thus between decision-makers and those affected by decisions. Voting is an informational signal from the ruled to the rulers, and legislation is an informational signal from the rulers to the ruled.
Now, what do we know about standard-issue humans trying to make decisions under partial certainty or great uncertainty regarding time-varying and complex pleasure/pain signals to optimize a large system for the expression of a broad system of values (and thus available good/evil stimuli)?
If you answered “they can’t really do it well”, I disagree. My position is: they need swift, sharp feedback in order to do it well. Humans are actually remarkably good at decision-making when given swift, sharp feedback signals.
So what’s the problem with an undemocratic democracy, or a technocracy as such? It shatters the feedback loop that’s signalling:
To the voters: how well their expressed policy preferences effect their expressed values, and how well their expressed values in the voting booth match their actual values as they experience themselves on an everyday basis.
To the government apparatus: what the masses prefer in public policy, how well the government is effecting those preferences, what impact actual enacted policies have on the voting masses, and the stability of the collective whole of society.
No feedback loop means the system as a whole can and will eventual spiral out of control from sheer divergence, as the people in power no longer know or care about the experience of the masses, leaving the masses with no reason to support the system itself, setting the stage for either a mass exodus or a revolt and thus destabilization of the State.
I agree with this. The place where I don’t follow your analysis is the assumption that voting is the primary means of feedback. We only hold elections every few years. If you want “swift” feedback, elections are not the tool for you.
Moreover, they don’t give very precise feedback in any real democracy. In, say, the UK, the election sends basically two bits of information per voter—LibDem, Tory, Labor, or other. Even in the US, the ballot just doesn’t encode a lot of information about voters; it’s single or double-digit bits per year. And each additional bit per voter is hugely expensive to collect. It costs millions of dollars to pay for all the analysis, campaigning, advocacy and so forth that goes into a national or regional election.
Happily, we don’t rely on elections as our primary mechanism for sending signals to the government. Functioning democracies have an amazingly rich set of mechanisms for getting prompt feedback from the public. We have opinion surveys. We have newspaper editorial pages. We have constituent letters and phone calls. We have donations. We have private meetings between government officials and lobbyists. We also have the whole panoply of economic indicators—politicians do notice when the stock market crashes or when bond rates spike.
I think trying to optimize the information content of the election system is missing the point. Elections aren’t the right tool for preference elicitation. Elections are there to give force to voter preference—they’re not only a measurement tool or even primarily a measurement tool. They’re a civic ritual to demonstrate the presence or lack of voter support for the government and they’re a system to designate a specific alternative in the event that the incumbents are voted out. it’s important that the election produces a clear outcome and an outcome with enough political support to sustain itself until the next election. It’s not important that it extracts the maximum information about the voter preferences on election day.
In conclusion: If you want swift and detailed feedback from the public, run a telephone poll. Rounding up millions of people and asking them about their preferences every few years is a lousy way to measure sentiment.
Well yes. This is a very good criticism of our current democracies.