Market dirigisme

Market dirigisme is the name I give to a style of public policy I think we ought to use more. The idea is pretty simple. Governments form preferences over how the polity ought to be but currently is not. Often, what governments should do is to explicitly purchase the changes in behavior they desire from the general public.

Put this way, it sounds a bit weird. But it’s not weird. As I write, Ohio is enrolling people who accept a COVID-19 vaccination in million dollar lotteries. Ohio is paying people a few dollars each, in expected value terms, to get vaccinated. Among people who see declining birth rates as a problem — “natalists” is sometimes the word — welfare state proposals like child allowances are often evaluated on an axis of whether and how much they might encourage childbearing. One can support child-attached benefits for other reasons (like reducing child poverty) even if one is “antinatalist”. But part of the coalition that supports these benefits is openly interested in using the public purse to purchase a higher birth rate from the citizenry.

I have views about vaccination and natalism, but this piece is not about those. This piece is about the style of policy. It is a good style of policy, we should use it more, and we should use it in place of, or at least to augment, another style of policy that is common but I think very fragile.

Often when a central government wishes to change the polity, it tries to induce changes at the level of subsidiary governments, rather than via citizens and households. The Federal government incentivizes states to expand Medicaid, for example. Elizabeth Warren proposes encouraging less restrictive land use policies by offering grants to local governments that adopt them.

There are deep problems with this style of center-to-subsidiary governance. It’s intuitively attractive, I think, due to a mistaken analogy between government subsidiarity to bureaucratic hierarchy. It feels “rational” or “logical” to work through the “chain of command” rather than have the center try to mess directly with with hundreds of millions of citizens about whose particular circumstances it knows little.

But subsidiary governments are not bureaus of the central state. They are separately chartered, separately elected, sovereigns. In the United States we often discuss this tension in terms of law. The Supreme Court decides whether some proposed inducement from the center is acceptable or an unconstitutional abridgment of powers reserved to the states. Policy entrepreneurs come up with clever arguments to persuade the Court.

But whatever the law decides of itself, the problem runs deeper. The central government is accountable to some weighting of the broad national public, and forms its goals and preferences accordingly. Subsidary governments are accountable to different weightings of different populations, and form very different preferences and priorities. When the center imposes inducements upon subsidiary governments, elected officials face often radically conflicting incentives. Perhaps their supporters desire the effect of single-family zoning, but the central government is incentivizing its abolition. How would, or should, local electeds behave? In practice, they will likely game the incentives. Local officials will seek ways they can both win the grants and deliver for their supporters, perhaps by eliminating the zoning but creating procedural pretexts for blocking multiunit development. We are left at best with a cat-and-mouse game, with the central government imposing ever more elaborate requirements, while local electeds invent ever more clever ways to have their cake and eat it. In practice, the center usually loses this game, or at best makes very slow progress, sewing cynicism about government capacity to deliver anything more than expressive victories.

A better approach is for the central government to alter the circumstances, and so the preferences, of the broad public. If we’d like denser communities, the central government can simply pay a subsidy to residents of communities growing denser. Elected officials of subsidiary governments no longer face conflicting incentives. If the subsidy is large enough to shift the preferences of the voters to whom local politicians are accountable, politicians will enact real change. If their supporters’ dispreference for density overwhelms the money, they won’t. The size of the subsidy can be set large enough to meet the central government’s objectives while still permitting some communities to opt out. While the aggregate effect is as political an outcome as any other form of state action, it would result from decentralized choices of citizens (either to support local government action to harvest the subsidy, or to eschew the subsidy in favor of the status quo) rather than direct compulsion by the central state. In the way that market outcomes do, either outcome will feel “natural”.

In general, the flow of incentives should be from the center to the edges, rather than from the center through subsidiary governments. Subsidiary governments are not subordinate governments, but distinct entities. They are accountable to citizens, not to a hierarchical superior. It is citizens that ought be the nexus between central and subsidiary governments.

A point in favor of this style of policy, I think, is that it works through finance. Money is abstract, fungible, bloodless. Land is birthright, race and increasingly party are tribe. These are things to fight and die for. But taxes and subsidies are arcane. They can be applied in increments, leaving scope for individual and local choice in ways that explicit command does not. It has become fashionable for people with my politics to lament “financialization”, and I do lament financialization, defined as the increase in rents to the financial sector and the decrease in the degree to which financial flows in our economy are tethered to what I perceive as meaningful production. But the last few years of tribalized politics, of self-righteousness and conflict largely detached (in my view) from meaningful progress, has reminded me of the virtues of financial materialism. The tax system might be racist, but as Dorothy Brown puts it, she had to become a “detective” even to perceive that. What if we could the mechanisms that have so quietly riven us apart be retooled to quietly bring us together? Money isn’t everything, but our conflicts over values would be less existential I think if our material circumstances weren’t so divergent, if our material interests were better aligned. I want just outcomes, but I don’t want people fighting and dying for them. I’d rather we just set the taxes and subsidies right.

If neoliberalism counseled leaving everything to markets, market dirigisme would put democratic publics in command of markets, while retaining markets’ extraordinary capacity to naturalize outcomes, and so diffuse and diminish conflict. Purchases of changes in behavior by the government would be overt and accountably debated, not hidden in technocratic “nudges”. Just as you know McDonald’s is trying to get you to eat hamburgers, you’d know the government is trying to pay you to have kids, or to get vaccinated, or to accept new housing nearby, or to diversify your neighborhood. But it would still be your choice whether to participate or to abstain. You might agree or disagree with the national policy. But retention of local agency means you can protest by opting out, rather than by blowing shit up. Even people who think McDonald’s is kind of a conspiracy to give us all heart attacks rarely resort to that.

Update History:

  • 28-Apr-2020, 11:40 p.m. EDT: “Once One can support child-attached benefits for other reasons”

4 Responses to “Market dirigisme”

  1. Nicholas Weininger writes:

    This is kind of question begging though. If the central government has preferences that a bunch of state governments disagree with, why should the central government win? There are some situations where you can give a reasonable answer to this– large interstate externalities, say, or the need for countercyclical deficit spending that states can’t do– but I think those are relatively uncommon compared to the times when the national majority just arrogantly thinks the preferences of dissenting state majorities are wrong and uses federal power to overrule them.

  2. Peter Dorman writes:

    I generally agree with this post, but the argument isn’t new — maybe the framing it in terms of chain of command vs center/edges is. (And that summons counterexamples like France, where the chain of command is real and pervasive.)

    The main challenge is fiscal. With so many objectives to pursue, if policy mainly took the form of bribery — excuse me, incentives — the state would be overwhelmed. It’s necessary to have negative incentives, taxes, as well as positive ones, feebate systems for instance. Of course, now things are more visible and complicated. Oh well.

  3. Akhil Rao writes:

    Interesting way to frame this. I agree with Peter Dorman’s point about fiscal challenges: using incentives at such a large scale makes me wonder about revenue balance. But how big do the incentives actually have to be to purchase behavior changes? The Ohio example seems instructive: spend a million dollars to purchase (probably) more than that in tax revenues from increased business activity and averted costs of overwhelmed medical systems. Sometimes these incentives might be self-financing.

    Nicholas Weininger raises a good positive point too — subsidiary levels of governments could do the same. The federal government just comes to the party with a much bigger bank account (and the ability to print more, subject to some constraints). We could ask the same questions about municipalities, too: the aggregate preferences of the CA state government maybe don’t fully align with those of Simi Valley. Again, it seems like the level of government with the biggest tax base has the most power in these games.

    I guess all I really have to add is that this seems like an extreme kind of fiscal federalism. The Oates-style models seem to mostly consider two levels of government, but why not more? The Interfluidity federalism model just calls the lowest level of government a citizen.

  4. Detroit Dan writes:

    I hope you are planning on compiling these posts into a policy book. Deep thinking about democracy and governance free from vested interests would seem to be an underserved niche. And you get feedback on your ideas when you post them here, so you can rethink and refine your proposals with the light of such intelligent criticism.

    Do you have any thoughts on various think tanks? There must be a lot of them with money to spend on behalf of various interest groups, but some that are better than others in paying people to come up with ideas that are good for the country as a whole. I’ve long been a fan of the various MMT organizations. In the area of international affairs, I’ve been impressed with Richard Hanania who is a Research Fellow at the Saltzman Institute of War and Peace Studies at Columbia University.

    With regard to this post on dirigisme in particular, I like its objective and non-partisan nature, with attention to both political and economic factors. Governing a nation of 328 million is a challenge, and requires people willing to think beyond tribal politics.

    Kudos also to the commenters here. I hope people will share more details and references to preferred political/economic models and approaches. I’ve learned a lot in this manner over the years.