The lottocracy
Elections are flawed and can’t be redeemed – it’s time to start choosing our representatives by lottery
Alexander Guerrero is an assistant professor of philosophy, medical ethics and health policy at the University of Pennsylvania.
It is easy to feel that what you do won’t make any difference. Recycle that can, bike or drive, buy from this company not that one, march in the streets against the factory closing or the looming war. It’s never enough: the forces are large and anonymous, and there aren’t enough of us. Or there are too many of us. Vote, petition, protest. It can all feel pointless: a kind of precious dancing around, keeping a low causal profile, with an eye on some imaginary Future Judgment. How clean my hands are! How little of the world’s horror has been made by them!
But we don’t care about that. We care about the horror: the steady-warming planet; the children born into hard, sad futures; the millions of homeless, and hungry, and unjustly imprisoned; the growing gap between the rich and the poor in Philadelphia, Kansas, and Kentucky, in Moscow, Ghana, and Paris. The problem, at bottom, is that we feel that we can’t make a difference. Ethically and politically, we are ghosts in a machine.
The celebrity comic Russell Brand is gesticulating wildly, urgently, in a hotel room, under the bright lights of a television interview. ‘Stop voting, stop pretending, wake up. Be in reality now. Why vote? We know it’s not going to make any difference. We know that already.’ He is responding to his interviewer, Jeremy Paxman, who is taking him to task for never having voted.
We are brought up to think that voting is important, that it is a necessary condition of being a politically serious person, that we can’t complain about politics if we don’t vote. This last principle has echoes of the more reasonable parental admonition, said of lima beans or cauliflower: don’t knock it until you’ve tried it. But that principle is based on sound epistemological grounds: you might, for all you know, like cauliflower or lima beans. The voting thing is, as Brand argues, stupid. There are ways of participating in public affairs other than voting. For example, one can become a celebrity and call for revolution in a television interview.
More to the point, the inference from not voting to not caring is a poor one. As Brand points out, you might care a lot about what happens and what the system is doing, but still realise that voting doesn’t affect what happens or what the system does. In most elections, the chance that your vote will make any difference to who wins is much smaller than the chance that you will be hit by a car on the way to cast your vote. But people still turn out to vote. They even drive through snow, miss work, and wait in line for hours.
This has puzzled political scientists and economists. Why do people vote? This is an empirical question; it concerns our actual motivations. Many answers have been given: we vote because we enjoy it; because we think others will think badly of us if we don’t; because we want to express ourselves; or cheer for our team; or because we believe that we have a duty to do so. One worry about all of these answers is that they seem disconnected from what makes voting seem so morally significant, something that it might be worth fighting and dying for the right to do.
In the modern world, we often find ourselves in the following situation. I know that whether I do X rather than Y won’t make a difference by itself. I also know that everyone else knows this about me and about themselves. I also know that if all of us do X, rather than Y, it
will make a difference. And everyone else knows this, too. So it’s striking and surprising that a celebrity such as Brand would come out and say, to millions, ‘Don’t vote,’ rather than ‘Vote for X.’ That was the revolutionary part of the interview. A thousand lefty celebrities have gone on TV and advocated for causes. Very few have gone on TV and said ‘Don’t vote.’ Very few have gone on TV and said, essentially, X and Y can both go fuck themselves.
One reason not to vote is that your vote — your one vote — is unlikely to make a difference to who wins the election. Another reason not to vote is that
it doesn’t matter who wins the election, that there is no difference between X and Y, republican and democrat, Tory and Labour. An extreme version of this thesis — which is obviously false — is that there is
no difference between our Xs and our Ys. Much more plausible versions of this thesis are that there is
not enough difference between our Xs and Ys, or that with respect to
some important issues there is no difference between our Xs and Ys.
Brand’s view is clear: ‘I’m not [refusing to vote] out of
apathy,’ he says. ‘I’m not voting out of absolute indifference and weariness and exhaustion from the lies, treachery, deceit of the political class that has been going on for generations.’ Brand says that many of us don’t engage with the current political system, because we see that it doesn’t work for us, we see that it makes no difference. ‘The apathy doesn’t come from us, the people,’ he says. ‘The apathy comes from the politicians. They are apathetic to our needs. They are only interested in servicing the needs of corporations.’
Is this true? Why would this be? Wasn’t the whole point of democratic elections to ensure that power would be in the hands of the people?
The theory of modern representative democracy goes something like this. Each of us is fundamentally autonomous and of equal moral worth, so that we have a claim to self-government, self-rule, to the extent that such self-government is compatible with an equal right to self-government of others.
This suggests something like direct democracy, in which each of us would have an equal say in determining whether we go to war, what policies and laws to adopt, what should be taxed and how much taxes should be, and so on. But — we quickly realise — modern politics is very complex; it is a full-time job to be even modestly well-informed about political issues. Ideally, one would spend all of one’s time doing it, in addition to having staff and resources to help. This suggests a move from
direct democracy to
representative democracy, where we would each have an equal vote in choosing that individual whom we think will best represent our interests and views. That person will act as our representative — and not as an elected tyrant — because to stay in power, she or he will have to be re-elected. If our representatives do things that we don’t like, we can vote them out. That’s the theory, and its simplicity and power — and the successes of actual electoral representative democracies — have led representative democracy to be the ascendant and unrivalled political system around the world.
So, what’s the problem? The problem is that despite the elections, elected representatives are not actually accountable, not meaningfully accountable, to those over whom they govern.
There are logistical hurdles to keep poor, marginalised citizens from successfully registering to vote
Even in established democracies there are concerns about the openness and fairness of elections. There are huge financial barriers to running for office, and considerable advantages to incumbency. Corporate money and television advertising have an outsized influence. There are logistical hurdles to keep poor, marginalised citizens from successfully registering to vote, and gerrymandering reduces competition, considerably. These difficulties all reduce how accountable our representatives are to us.
Even if these problems were addressed, they would succeed only in making elections fair. But meaningful accountability requires not just open and fair elections; it also requires that we are capable of engaging in
informed monitoring and evaluation of the decisions of our representatives. And we are not capable of this. Not because we are stupid, but because we are ignorant: ignorant about what our representatives are doing, ignorant about the details of complex political issues, and ignorant about whether what our representative is doing is good for us or for the world.
Our ignorance means that representatives can talk a good game, and maybe even try to do a few things that benefit the majority of us, but the basic information asymmetries at the heart of the representative system ensure that, for many issues — defence manufacturing and spending, policy that affects the insurance and pharmaceutical industries, agribusiness policy and regulation, energy policy, regulation of financial services and products — what we get is what the relevant business industries want. In the presence of widespread citizen ignorance and the absence of meaningful accountability, powerful interests will effectively capture representatives, ensuring that the only viable candidates — the only people who can get and stay in political power — are those who will act in ways that are congenial to the interests of the powerful.
These concerns are brought to the fore if we think about how little we know about most of what our representatives do, how little real choice goes into the election of our representatives, and how much deference to the goodwill of our favoured political party is required. Even when we step outside partisan information streams, most issues are complex, and much of what we believe about them is a result of information provided by a few dominant media institutions. But there might be a way to overcome these difficulties, if we rethink the fundamentals of democracy itself.
One response to these problems is to go small. In a small community, collective action problems are less prevalent, and can be solved organically. We can detect and shun violators or freeloaders. And information asymmetries disappear: I know the issues and problems that affect us, as do you. We understand their complexities. They are within our daily life. If we need to use representatives for some reason, we will know them personally, as friends or neighbours. We can easily see what they do.
One difficulty with this response is that it is not obvious how to go small. We know we can make a difference by connecting to people in more direct ways: talking to people we see during our day, providing food and shelter for local families, teaching in a prison. But it can be hard to see how our political communities can be made smaller in this way. And many of us are hard-pressed for time, energy, and the resources to make these efforts.
Worse still, the going small strategy can seem inadequate when compared with the forces at hand, the foundations of the horror. We are globally connected now. We can’t roll back the technological development and population increases that threaten the planet, and make it so that my small choices and your small choices all have such large, global effects. This is where the political system is supposed to be of help, but our system is broken.
Political systems are a kind of technology, inventions of human beings to bring about things we care about: peace, prosperity, freedom. Representative democracy is old technology. It dates back to the Roman Republic. Russell Brand says don’t vote, the system is broken, and I think he’s right: we do need a new system. But it is important to stress that in saying that, one needn’t be committed to the view that everything is awful. It’s not. Modern democratic governments do many things well, even if imperfectly: food safety and quality control, traffic safety and road maintenance, regulation and enforcement of building and zoning codes, public health crisis response, air-travel regulation, antitrust and market competition regulation, hospital and health care support, energy and telecommunications regulation, civil court systems, public libraries and basic public education, police and fire protection, support for basic and applied scientific research.
It’s true that for each item I just listed, there are 20 legitimate, serious complaints that could be made about the way some particular government handles that responsibility. It’s also true that modern governments collect an extraordinary amount of money in taxes, so it should be no surprise that some things get done. Still, it would be a mistake to think that representative democracy is a disaster. It’s good, but that shouldn’t keep us from trying to make an even better system by paying attention to the ways in which it falls short.
Electoral representative democracy has undergone a great many changes since it came on the scene. We’ve seen a steady increase in constitutionalism and proportional representation. We’ve seen multi-member districts and non-geographic districting become popular, along with publicly financed campaigns in some places, and the rise of the administrative state. These changes have offered substantial improvements, but it is now time to reform the heart of the system: the election. Modern policy is too complex for there to be meaningful electoral accountability. Electoral capture is too easy and too important for powerful interests. So, what’s the alternative? Get rid of elections. Use lotteries to select political officials.
There is historical precedent for this kind of method, also referred to as ‘sortition’. There are also a number of academics who have argued for a role for lotteries in the selection of political officials, including C L R James, Oliver Dowlen, and Peter Stone. In ancient Athens, the birthplace of democracy, lottery-selection was used to choose political actors in three of its four major governmental institutions. Selection of political officials in late medieval and early renaissance Italy incorporated selection by lot. More recently, Citizens’ Assemblies (in which citizens were chosen at random to serve on the assembly, and in which citizens heard from experts prior to coming up with their own proposals) were used in the Netherlands to reform election law, and in Canada (in British Columbia and Ontario). Randomly chosen citizens were also brought into the process of constitutional reform in Iceland in 2010, but nothing of the scope that I am envisioning has been tried before.
There are hard questions about how exactly to structure a political system with lottery-selection at its heart. Here’s one approach, which I am in the process of developing, that I call
lottocracy. The basic components are straightforward. First, rather than having a single, generalist legislature such as the United States Congress, the legislative function would be fulfilled by many different
single-issue legislatures (each one focusing on, for example, just agriculture or health care). There might be 20 or 25 of these single-issue legislatures, perhaps borrowing existing divisions in legislative committees or administrative agencies: agriculture, commerce and consumer protection, education, energy, health and human services, housing and urban development, immigration, labour, transportation, etc.
People would not be required to serve if selected, but the financial incentive would be significant
These single-issue legislatures would be chosen by lottery from the political jurisdiction, with each single-issue legislature consisting of 300 people. Each person chosen would serve for a three-year term. Terms would be staggered so that each year 100 new people begin, and 100 people finish. All adult citizens in the political jurisdiction would be eligible to be selected. People would not be required to serve if selected, but the financial incentive would be significant, efforts would be made to accommodate family and work schedules, and the civic culture might need to be developed so that serving is seen as a significant civic duty and honour. In a normal year-long legislative session, the 300 people would develop an agenda of the legislative issue or two they would work on for that session, they’d hear from experts and stakeholders with respect to those issues, there would be opportunities for gathering community input and feedback, and they would eventually vote to enact legislation or alter existing legislation.
Single-issue focus is essential to allow greater learning and engagement with the particular problems, especially given the range of backgrounds that members would bring to the institutions, and the fact that these individuals would be amateurs at the particular task of creating legislation. Lottery-chosen representatives would have more time to learn about the problems they’re legislating than today’s typical representatives, who have to spend their time learning about every topic under the sun, while also constantly travelling, claiming credit, and raising funds to get re-elected. In the lottocratic system representatives will be — at least over a long enough run — descriptively and proportionately representative of the political community, simply because they have been chosen at random. But they will not have in mind the idea that they are to represent some particular constituency. Instead, they will be like better-informed versions of ourselves, coming from backgrounds like ours, but with the opportunity to learn and deliberate about the specific topic at hand.
No pure lottocratic system has ever existed, and so it’s important to note that much could go wrong. Randomly chosen representatives could prove to be incompetent or easily bewildered. Maybe a few people would dominate the discussions. Maybe the experts brought in to inform the policymaking would all be bought off and would convince us to buy the same corporate-sponsored policy we’re currently getting. There are hard design questions about how such a legislative system would interact with other branches of government, and questions about the coherence of policymaking, budgeting, taxation, and enforcement of policy. That said, it’s worth remembering the level of dysfunction that exists in the current system. We should be thinking about comparative improvement, not perfection, and a lottocratic system would have a number of advantages over the current model.
The most obvious advantage of lotteries is that they help to prevent corruption or undue influence in the selection of representatives. Because members are chosen at random and don’t need to run for office, there will be no way for powerful interests to influence who becomes a representative to ensure that the only viable candidates are those whose interests are congenial to their own. Because there is no need to raise funds for re-election, it should be easier to monitor representatives to ensure that they are not being bought off.
44 per cent of US Congresspersons have a net worth of more than $1 million; 82 per cent are male; 86 per cent are white, and more than half are lawyers or bankers
Another advantage of lotteries over elections is that they are likely to bring together a more cognitively diverse group of people, a group of people with a better sense of the full range of views and interests of the polity. Because individuals are chosen at random from the jurisdiction, they are much more likely to be an ideologically, demographically, and socio-economically representative sample of the people in the jurisdiction than those individuals who are capable of successfully running for office. As a point of comparison, 44 per cent of US Congresspersons have a net worth of more than $1 million; 82 per cent are male; 86 per cent are white, and more than half are lawyers or bankers. Recent empirical work by Scott Page and Lu Hong has demonstrated that cognitively diverse groups of people are likely to produce better decisions than smarter, or more skilled, groups that are cognitively homogenous.
Elections lead elected officials to focus on those problems for which they can claim credit for addressing, and to ignore or put on the back burner those problems with a longer horizon or those solutions for which it is harder to get credit. This negligence is made possible by voter ignorance and made inevitable by the perverse short-term incentives that elections provide. Lottery selection can help us to avoid this problem.
Perhaps the most urgent issue we face is climate change, a complex collective action problem that will almost certainly require a political solution to solve. But many of the worst effects of climate change won’t be realised for decades, and so politicians are unlikely to pay the short-term political cost given that they won’t see the longer-term political benefits. Even when there are clear steps that need to be taken, many elected officials will avoid acting out of fear of the immediate consequences. Individuals chosen at random won’t be hamstrung by these skewed incentives. If there is agreement on a viable solution, to climate change or to the myriad other issues that affect our children and grandchildren, lottocratic representatives will have the luxury of looking beyond this week’s poll or next week’s fund-raiser.
This task of radically redesigning government is usually dismissed as utopianism, but there is no reason to think that electoral representative democracy can’t be improved upon, just like every other kind of technology. Of course, one must be aware of limitations in the materials; we must think critically and carefully about what we know, what we have learned from psychology, economics, history, political science, law, and philosophy. And we have to be mindful of the dangers that attend our tinkering. Some of the worst horrors of the 20th century were the result of political design projects gone terribly wrong. So, we must tread carefully and take small steps. But we can’t continue to stand still.
Observation
The lottocracy
Elections are flawed and can’t be redeemed – it’s time to start choosing our representatives by lottery
Alexander Guerrero is an assistant professor of philosophy, medical ethics and health policy at the University of Pennsylvania.
The celebrity comic Russell Brand is gesticulating wildly, urgently, in a hotel room, under the bright lights of a television interview. ‘Stop voting, stop pretending, wake up. Be in reality now. Why vote? We know it’s not going to make any difference. We know that already.’ He is responding to his interviewer, Jeremy Paxman, who is taking him to task for never having voted.
In the modern world, we often find ourselves in the following situation. I know that whether I do X rather than Y won’t make a difference by itself. I also know that everyone else knows this about me and about themselves. I also know that if all of us do X, rather than Y, it
will make a difference. And everyone else knows this, too. So it’s striking and surprising that a celebrity such as Brand would come out and say, to millions, ‘Don’t vote,’ rather than ‘Vote for X.’ That was the revolutionary part of the interview. A thousand lefty celebrities have gone on TV and advocated for causes. Very few have gone on TV and said ‘Don’t vote.’ Very few have gone on TV and said, essentially, X and Y can both go fuck themselves.
Brand’s view is clear: ‘I’m not [refusing to vote] out of
apathy,’ he says. ‘I’m not voting out of absolute indifference and weariness and exhaustion from the lies, treachery, deceit of the political class that has been going on for generations.’ Brand says that many of us don’t engage with the current political system, because we see that it doesn’t work for us, we see that it makes no difference. ‘The apathy doesn’t come from us, the people,’ he says. ‘The apathy comes from the politicians. They are apathetic to our needs. They are only interested in servicing the needs of corporations.’
...Here’s one approach, which I am in the process of developing, that I call
lottocracy. The basic components are straightforward. First, rather than having a single, generalist legislature such as the United States Congress, the legislative function would be fulfilled by many different
single-issue legislatures (each one focusing on, for example, just agriculture or health care). There might be 20 or 25 of these single-issue legislatures, perhaps borrowing existing divisions in legislative committees or administrative agencies: agriculture, commerce and consumer protection, education, energy, health and human services, housing and urban development, immigration, labour, transportation, etc.
These single-issue legislatures would be chosen by lottery from the political jurisdiction, with each single-issue legislature consisting of 300 people. Each person chosen would serve for a three-year term. Terms would be staggered so that each year 100 new people begin, and 100 people finish. All adult citizens in the political jurisdiction would be eligible to be selected. People would not be required to serve if selected, but the financial incentive would be significant, efforts would be made to accommodate family and work schedules, and the civic culture might need to be developed so that serving is seen as a significant civic duty and honour. In a normal year-long legislative session, the 300 people would develop an agenda of the legislative issue or two they would work on for that session, they’d hear from experts and stakeholders with respect to those issues, there would be opportunities for gathering community input and feedback, and they would eventually vote to enact legislation or alter existing legislation.
No pure lottocratic system has ever existed, and so it’s important to note that much could go wrong. Randomly chosen representatives could prove to be incompetent or easily bewildered. Maybe a few people would dominate the discussions. Maybe the experts brought in to inform the policymaking would all be bought off and would convince us to buy the same corporate-sponsored policy we’re currently getting. There are hard design questions about how such a legislative system would interact with other branches of government, and questions about the coherence of policymaking, budgeting, taxation, and enforcement of policy. That said, it’s worth remembering the level of dysfunction that exists in the current system. We should be thinking about comparative improvement, not perfection, and a lottocratic system would have a number of advantages over the current model.
The most obvious advantage of lotteries is that they help to prevent corruption or undue influence in the selection of representatives. Because members are chosen at random and don’t need to run for office, there will be no way for powerful interests to influence who becomes a representative to ensure that the only viable candidates are those whose interests are congenial to their own. Because there is no need to raise funds for re-election, it should be easier to monitor representatives to ensure that they are not being bought off.
Another advantage of lotteries over elections is that they are likely to bring together a more cognitively diverse group of people, a group of people with a better sense of the full range of views and interests of the polity. Because individuals are chosen at random from the jurisdiction, they are much more likely to be an ideologically, demographically, and socio-economically representative sample of the people in the jurisdiction than those individuals who are capable of successfully running for office. As a point of comparison, 44 per cent of US Congresspersons have a net worth of more than $1 million; 82 per cent are male; 86 per cent are white, and more than half are lawyers or bankers. Recent empirical work by Scott Page and Lu Hong has demonstrated that cognitively diverse groups of people are likely to produce better decisions than smarter, or more skilled, groups that are cognitively homogenous.
Perhaps the most urgent issue we face is climate change, a complex collective action problem that will almost certainly require a political solution to solve. But many of the worst effects of climate change won’t be realised for decades, and so politicians are unlikely to pay the short-term political cost given that they won’t see the longer-term political benefits. Even when there are clear steps that need to be taken, many elected officials will avoid acting out of fear of the immediate consequences. Individuals chosen at random won’t be hamstrung by these skewed incentives. If there is agreement on a viable solution, to climate change or to the myriad other issues that affect our children and grandchildren, lottocratic representatives will have the luxury of looking beyond this week’s poll or next week’s fund-raiser.
This task of radically redesigning government is usually dismissed as utopianism, but there is no reason to think that electoral representative democracy can’t be improved upon, just like every other kind of technology. Of course, one must be aware of limitations in the materials; we must think critically and carefully about what we know, what we have learned from psychology, economics, history, political science, law, and philosophy. And we have to be mindful of the dangers that attend our tinkering. Some of the worst horrors of the 20th century were the result of political design projects gone terribly wrong. So, we must tread carefully and take small steps. But we can’t continue to stand still.
