Why I Am Not A Statist

E. Glen Weyl

January 28, 2019

This post is the second in a series I am planning articulating how my views and perhaps the views of many in the RadicalxChange movement differ from those of other easily identifiable ideological camps. In this post I address an ideology that I will call “statism”. The type of state planning that largely went out of fashion, at least the Anglo-Saxon world, in the 1970s is returning to the politics of wealthy countries with a vengeance. Many of these ideas are incompletely formed and represent a general set of political inclinations rather than a fully worked-out philosophy. Thus, I will not focus on the particularities of proposals made by leaders such as Jeremy Corbyn, Alexandria Ocasio Cortez or Bernie Sanders, though I believe that many admirers of these leaders have inclinations towards the political tendency from which I am here dissenting. After all, most practical political movements represent a combination of philosophical tendencies and I do not want to call out any mainstream leader, given that I would hope to persuade them to move closer to a Liberal Radical outlook.

Instead, therefore, I will focus my attention on a thinker who consistently and coherently exposes the perspective I dissent from: Matthew Bruenig and his People’s Policy Project. To be clear, there are many aspects of the statist critique of capitalism I wholeheartedly endorse. Bruenig’s critiques of private property and much of his argument that much more of what is currently private property should be commonly owned are spot on. His observation that many “free market” societies like Singapore and Scandinavia have significant levels of commonly owned property are correct. His arguments that private (health and unemployment) insurance are subject to rampant adverse selection, making it a tremendously wasteful way to administer insurance is surely right. His analysis of contemporary American capitalism as having degenerated into the rule of small elite of institutional investors systematically coordinating capital to expropriate workers is central to my critique of the status quo. And his fervent desire for a more egalitarian and diverse world is one I deeply share.

It is not then socialism in the sense of common ownership and certainly not the desperate need for collective action that I wish to challenge. It is instead the valorization of The State, or more generally some historically derived but ultimately quite arbitrary national polity, as the locus of collective action that I think is untenable, reactionary and, if taken to extremes, quite dangerous. It is this that I will call “statism” and I will do my best to clearly limit my critique to this ideology. Because this, like the other articles in the series, is purely critical, I will not discuss proposed alternative frameworks or solutions, but anyone familiar with RadicalxChange ideas may find it useful to contrast my critiques of statism with the ideas of Liberal Radicalism and polypolitanism that I have recently been developing.

Many ideas for increased collective action at the level of The State (such as much higher marginal income or carbon tax rates funding a social dividend or free university tuition) may be more or less appealing (I often think they are of limited efficacy in any case), but I would not flag them a deeply statist in the sense I critique. There are other “leftist” policies in current debate that I am deeply sympathetic to and do not consider statist at all (see more below), such as some kinds of dramatic expansions of the scope of antitrust policy based on clear formal rules, attempts to empower the growth of new types of labor movements and so forth.
What really concerns me is policies that aim to transfer very large amounts of discretionary power over and often-unprecedented scope to present democratic polities through a mixture of representative and direct democracy without simultaneously proposing equally detailed and novel methods for both intra- and internationally checking the power this creates. Examples include:

To take a specific example, one of my favorite love-to-hate proposals is Bruenig’s Social Wealth Fund. The idea is that a fund managed by the United States federal government would come to be the dominant, voting shareholder in most publicly-traded American corporations. Either a democratically chosen administrator or the public directly would vote the shares (which would be held, roughly, equally by all Americans) and the fund would come to manage something like 30-70% percent of US public equities. When I need to “pick on” a policy at length below, the Social Wealth Fund will be the target.
To be clear, I am *not * critiquing a policy that is similar where, say, 30% of equities are owned by a Social Wealth Fund and are passively managed with shares not being voted on and no corporate governance involvement. Bruenig explicitly cautions against this option and while I think it would have some challenges, I could imagine versions of it being superior to the status quo. My focus is on the power over corporate governance and the economy more broadly this would place in the hands of a simplistic form of democracy, not on a high effective tax rate on capital.

The structure of my argument will be:

  1. Simple democratic structures should be expected to lead to good outcomes only under relatively limited conditions, the most important of which is when the topic addressed by the democracy has a natural polity (a set of people affected by knowledgeable about a set of decisions) that coincides with the electoral polity of the democratic state.
  2. In the past, when democratic states have had great power over topics whose natural polity lines up poorly with their boundaries, bad and even disastrous consequences have frequently followed.
  3. Most of the collective action problems we face today have a natural polity that lines up poorly with existing states, being smaller, larger or most commonly cutting across state borders.
  4. Thus, giving dramatically increased discretionary powers to existing simple democratic states to address present collective action problems is dangerous.
  5. While these difficulties could be overcome by co-designing new governance structures along with these increased powers that would help ensure they are used rather than abused, the statist tendency I criticize is one that generally ignores or minimizes the problem of governance, and to the extent they don’t they are largely not statists of the type I dissent from.
  6. Statism is thus a seductive, but ultimately thin, deus-ex-machina that would in most cases worsen the already extreme dangers facing communities and the planet, while leaving the collective action problems it purports to solve neglected.

First, we should expect simple democracy, based on some version of one-person-one-vote (1p1v) and without extensive “liberal” constraints (such as federalism, checks-and-balances, strong protections of individual rights, international treaties, etc.), to lead to good outcomes only when “the majority is right”. Such rules are based on the principle of majority rule, which will be appropriate on issues where the preferences and knowledge of a typical citizen are reasonably representative of the overall good of society. This is commonly the case in reasonably homogeneous societies, where differences of interest and opinion are relatively “randomly distributed”, or on issues where this is the case assuming that all the relevant individuals for a decision are given the franchise. It will tend to fail dramatically when there are deep, consistent, and politically salient “fractures” that divide societies, at least on certain important issues, into aggrieved minorities. These minorities are either divided against the majority or are disenfranchised, such as in cases where many of the individuals most affected by important political decisions do not have a vote.

A natural way to think about this is whether the boundaries of the “actual polity”, those participating in the voting process, lines up naturally with what I will call the “natural polity” of those individuals who are most and roughly equally concerned with the outcome of the collective choice. If the actual polity diverges in a severe way from the natural polity (many examples soon to follow), we should expect simple democracy to lead to very bad outcomes. While, to keep matters simple, I am focusing on inclusion or exclusion from the polity as a 0-1 variable, in most cases in practice the actual and natural polity will both be weighted (some will have more effective voice than others and some will be more or less affected by a decision), but this complication only further strengthens the argument because simple democratic mechanisms based on 1p1v have a hard time dealing with this subtlety.

Second, this issue is not merely one of theory or academic speculation. The failures of simple democracy when actual and natural polities diverge is one of the most persistent themes of political history. It is useful to divide historical cases into three buckets:

  1. Actual polity>natural polity (AP>NP): this is the case of minority or local group oppression, when the actual polity is much larger than the natural polity for an issue.
  2. NP>AP: this is the case where important individuals are disenfranchised from the decision.
  3. NP≠AP: this the generic case when the natural polity cuts across the boundaries of the actual polity and is neither smaller nor larger; some members of the actual polity are in the natural polity, but not all, and there are members of the natural polity disenfranchised form the actual polity.

I limit myself to two examples of disastrous outcome of simple democracy historically in each of these cases, focusing on contemporary examples where those who tend to favor statist policies will clearly agree that the policy was disastrous, and which should clearly illustrate a much broader phenomenon.
Disastrous examples of AP>NP:

  1. The Rohingya genocide: The end of military and the emergence of democratic rule in Myanmar was widely celebrated around the world. But, as has been widely noted, it appears to have only fueled long-standing sectarian divisions between the Buddhist majority and the Muslim Rohingya minority, precipitating an on-going genocide. Here the NP is local Rohingya communities and the AP is the Burmese nation state.
  2. Underrepresented communities in US cities: Many US cities have a variety of substantial minorities (viz. African American, various immigrant communities, artist communities, etc.) that are systematically underrepresented and neglected by majority-elected city administrations. Here the NP is these communities and the AP is the city.

Disastrous examples of NP>AP:

  1. Global change: Anthropogenic climate catastrophe is one of the greatest threats facing the world. Yet the effects being global, many nation states have strong incentives to free ride on mitigation efforts, leading to our present stalemate. Here the NP is essentially the cosmopolis and the APs are a bunch of nation states.
  2. Migration and global inequality: Inequality across countries is 2-3 times greater, depending on how you measure it, than is inequality within countries and millions of people around the world would be willing to take a substantial chance of death to have the opportunity for safety and economic advancement that wealthy countries offer. Yet wealthy countries are increasingly violently slamming their doors shut not just against the desperately and unjustly impoverished by even against those fleeing mass murder. Here the NP is the set of people who would like to live in wealthy countries and the AP is those currently living there.

Disastrous examples of NP≠AP:

  1. International waterways: Nature does not respect the arbitrary boundaries of nation states. Two of the largest rivers in the world, which supply crucial public goods and can easily be degraded by exploitation, are the Amazon and Nile, which each cross at least half a dozen countries. The free-riding by and conflicts among the nation states that lie along these rivers has contributed to systematic degradation, poisoning and deforestation. Yet the NP here is not simply the whole of these states, but instead those who live along and directly benefit from the river, which in most cases is a minority or even a small minority of each polity. Thus, the NP is the union of several national minorities, while the APs are a collection of national polities.
  2. The War on Drugs: The war on drugs in the Americas has been roughly, a war waged against poor and especially black and brown citizens of the Americas allegedly for the benefit of these same groups. Putting aside the opioid crisis (the policy response to which has been, unsurprisingly, very different) it wouldn’t be far off to argue that there has been virtually no drug enforcement against the well-off white Americans who many studies suggest consume most cocaine. The brunt of the drug war has fallen not just against black and brown US citizens who have been incarcerated at literally record-setting levels, but on the tens of millions in Central and South America whose lives have been shattered by resulting violence. The NP here is, I would thus argue, the black and brown minorities of the United States plus the (sometimes minority, sometimes majority, depending on country) of Latin Americans directly impacted by drug violence. Yet the AP of American citizens have consistently supported doubling down on policies destroying the lives of much of the NP.

In short, it seems quite clear that not just in theory but in practice, significant divergences between the AP and NP tend to lead to very poor outcomes from statism.

Third, nearly all major policy issues we face today are ones where the NP and the AP of existing established states diverge sharply. The examples above are all leading contemporary issues, but a few others will be illustrative:

Fourth, it seems clear that many of the sort of disasters that have occurred in previous mismatches between natural polity and actual polity are likely to occur if the discretionary authority of actual polities are dramatically increased in the ways that current statists are advocating. As a detailed example, consider the Social Wealth Fund proposal where, remember, either a government-appointed board or a democratic majority would effectively become the controlling shareholders of most US public companies. The natural polity here would be the workers and product consumers of these corporations and the actual polity would be the United States electorate. What could go wrong from this divergence?

  1. Many employees and customers of US companies are abroad. Given extreme recent nationalism and protectionism, it seems quite likely that US national democratic control of these companies would lead to protectionism by owner preference, probably of such an extreme form that the US’s status as a “market economy” under WTO rules would be suspended. Observe the behavior, prior to checks from the European Union, of European state-owned national champions.
  2. Public companies would likely do everything within existing laws to stop hiring undocumented and perhaps even certain types of documented migrants.
  3. Efforts of US companies to combat climate change, which are increasingly gaining steam, would likely be stopped in their tracks given the widespread skepticism of the American public about climate change.
  4. Efforts at diversity and inclusion by corporations would likely be significantly reduced, based on current public opinion on these issues.
  5. Investments in research and development on speculative or unpopular projects would likely be squashed, while investment in wasteful white elephant schemes that are attention-grabbing would likely be greatly increased.
    In short, the corporate sector would come to inherit all the dysfunctions of the public sector (though, admittedly, some of the dysfunctions of the corporate sector, such as chronically low investment, low wages, and exploitation of US consumers, would be thereby eliminated). But now, rather than these two dysfunctions checking and balancing each other, every US company would come to look like the current US government and would be roughly equally likely at any time to be ruled by a nativist white supremacism, capitalist conspiracies, or a moderately progressive agenda.

This strikes me as quite a disastrous outcome. For all the monopolization and cruelty of the current corporate economy, at least companies seek workers and customers in a reasonably flexible and adaptive manner that makes the base they serve, even if poorly, somewhat heterogeneous and diverse. Furthermore, to the extent that not all companies have precisely the same set of investors, the capitalists in control are, while all wealthy elites, at least a range of different power centers rather than a single actual polity. Effectively nationalizing all of these would eliminate that tendency and homogenize the control of much of the world’s economy under a single wildly unrepresentative AP.

Now, one might reasonably argue (as I have) that this apparent diversity is mostly a mirage and that in fact a few investment companies (e.g. Fidelity, BlackRock, State Street) with a small number of corporate governance officers control roughly ¼ of all US equities at present and mostly coordinate these companies in the interest of the capitalist class. Wouldn’t it be better to coordinate things in the interest of the broad US public? I would argue not:

  1. This plan would, rather than solving the monopolization of industry in the US as can easily be done by simply enforcing antitrust law against financial companies, permanently lock in the rigidity and concentration of power at present.
  2. There are other ways to redistribute the associated wealth that dramatically increase decentralization of power and would grow the overall pie.
  3. While competition in the US economy is dramatically reduced by investment funds, it is far from absent at present (there are plenty of privately-owned companies, or companies with a large concentrated shareholder), as it would be for anyone not in the US polity under the Social Wealth Fund proposal.

A natural response would be, it works in Alaska and Norway, why not in the US? The answer should be obvious: those countries are a tiny part of the global economy, while the US is a huge part. Thus, effectively those vehicles are purely for investment, and they are small investment funds relative to any of the problematic funds like those mentioned above. Their natural polity is thus their investors (the Alaskan and Norwegian polities) and pure investment services (with limited voting power) have the property that they are reasonably scale invariant…once you have a few million members, most scales work as the basic investment advice is quite straightforward and there are few economies to be gained by growing further. To the extent that there are harms from this concentration, they are felt by consumers and workers around the world who experience reduced competition, not by Alaskans and Norwegians in particular, so even if there are large harms, they would be hard to see. In short, the fact that things have gone reasonably well for Alaskans and Norwegians because of their large SWFs is simply a non sequitur with regards to the key problems with a US SWF.

While I will not go into other examples in detail for brevity, there are quite clear similar risks with other statist plans. Placing global platforms like Facebook and Google under the discretionary control of the US democratic polity would likely result in far more rampant politicization of content (such as news), further erode civil society and could even be used as a weapon against the remaining governments of the Left in Europe. If this sounds like a fantasy to you, look at the voices most loudly advocating for this idea in the US. A single payer system for the entire US economy, which provides most of all funding for pharmaceutical research, would be qualitatively different than in other countries. Such a system would be very likely to regulate drug prices down to levels like those in other single-payer systems and thus cut off most of the funding for pharmaceutical research at present. A “Green New Deal” could easily become a protectionist football, used to favor domestic green projects over more-efficient Chinese ones, further escalating trade tensions with China.

Now, fifth, I am not claiming there are no ways to structure policies of these general sorts that would not lead to disastrous outcome: there are ways to effectively “nationalize” a large fraction of US wealth that would further decentralize power, for example and as mentioned above. The crucial thing is that these designs are decidedly not based on the idea that we should “nationalize first, worry about governance second”, or worse, simply turn things over to an unchecked majoritarian decision process. Successful exercises of state power have co-evolved tightly with detailed governance structures that help align actual polities and natural polities through checks and balances. Examples include:

While such systems are described by extreme capitalists as “statist” or “central planning” they are not statist in the sense I use the word: they involve simple, transparent, widely understood legal or quasi-legal regimes that decentralize power, in a way that is quite different from, and I would argue much more effective than, simple capitalism. They constrain the discretionary power of the democratic state, as well as that of private wealth. There are also international equivalents, such as restraints on nuclear weapons, rules of war and trade barriers, that try to deal with NP>AP and NPAP cases; while these have been relatively ineffective, there are increasing experiments with creative new means of international cooperation, such as blockchains and open source software collaborations. Anyone interested in developing efficacious versions of such systems will find an ally in me.

What worries me, and what I label statism, is the tendency to allow idealistic schemes to address issues like inequality and the environment to be championed without parallel developments in effective governance mechanisms. While every citizen should judge for herself the extent to which statism thus-defined manifests in any given political movement or proposal, such sentiments have a real purchase on the political imagination of some prominent intellectuals and political leaders. Take again Bruenig’s Social Wealth Fund proposal: of 62 pages describing the design of the system, 1 is devoted to how what could well be most of the productive economy is to be governed. Majoritarianism, either direct or elected, is the only governing principle. In fact, Bruenig has repeatedly argued decentralization of power by institutions is meaningless and that every policy regime is equally centralized as all ultimately emanate from the authority of the state (I have no direct quote for this as Bruenig deletes his tweets, but I do not think he will dispute this characterization).

Attitudes like this are dangerous. If they capture the popular imagination, they will not solve the primary collective action problems we are facing. Instead they will further cement the hegemony of existing mismatched actual polities over ever more aspects of our rich and diverse collective life that resent the violence of such simplifications. They are thus a recipe for worsening existing conflicts and oppression.

Note: I am grateful for detailed comments on this post to Vitalik Buterin, Zoë Hitzig, and especially Aksel Braanen Sterri, though I make no claim they endorse it.