Arnold Kling  

Fantasy Federalism

PRINT
Morning Commentary... Mood and Macro...

This is going to be a random daydream sort of post. It beats watching Super Bowl pregame or doing pointless snow-shoveling (you loaded 16 tons, what do you get? Another day older and deeper in debt).

I want the U.S. to return to Federalism. But some of our states are too large--California has a higher population than any of the non-U.S. countries that are in the top 10 of the economic freedom index. And some states are too small. Finally, some cities are ridiculously large, but they are not easy to break up. So here is my proposal:

1. Turn any area of 500,000 or more people within an area of 100 square miles or less into a city-state. City-states would be autonomous, other than their participation in the United States as a whole. Note that New York City would be too large an area. Manhattan would be a city-state by itself. So would Brooklyn. Los Angeles and Chicago would be broken up as well. On the other hand, Indianapolis and Columbus are too spread out to be city-states by this definition. My guess is that we would end up with about twenty city-states, including the multiple ones carved out of New York, Los Angeles, and Chicago. Once the initial set of city-states is determined, it does not need to be changed. There is no reason to suddenly create a city-state out of a city that just crossed the 500,000 person threshold.

Each city-state should be governed by a single elected individual, whose powers are limited by a Constitution, but who rules until death, resignation, or loss of a recall election. Recall elections will be held when, say, 5 percent of the eligible voters in the city sign a recall petition. If a governor loses a recall election, a new election for governor is held 30 days later. One need not reside in the city-state to run for governor, although presumably once elected governor one would choose to live there.

My thinking is that a legislative branch does not necessarily help in a city-state. They can be pressured by interest groups, especially public employee unions. Instead, a wise, benevolent governor is what you want. With about twenty city-states to observe, citizens can get a pretty decent idea of who is wise and benevolent and who is not.

2. Next, we want to create county-states. These would cover larger areas than city-states, and they would be governed the way states are today, with elected legislatures and governors. County-states would range in population from 500,000 people to 10 million people.

a. Of the remaining counties, turn any county with 500,000 or more people into a county-state. If there are many small counties adjacent to one another but not near a large county, then merge the small counties together.

b. Merge each county with a population under 500,000 with the adjacent larger county,. If there is more than one adjacent larger county, come up with some mechanism to pick the merger partner--it could be done by election, or by a judge. Any mechanism will do. The result of the mergers will be county-states with populations of at least 500,000 each.

c. Once we have the county-states, any two adjacent county-states with fewer than 5 million people have the option to merge with one another, based on a majority vote in both county-states. So, Montana could merge with Idaho, if both states wanted a merger. Actually, the optional merger concept is not necessary for this scheme. I am more focused on getting rid of polities that are too big, as opposed to worrying about polities that might be too small.

d. County-states would be governed as states are today, with elected legislatures and governors. My thinking is that because county-states would occupy large areas, it would be very difficult for unsatisfied people to move from county-states. Therefore, they should have more voice in their government than people who live inside city-states.

Note that county-states could choose to be highly decentralized. I do not mean to force the small counties of New Jersey to lose their autonomy. I just mean that the government of the state of New Jersey could be broken into between two and ten county-states.

3. Government responsibilities.

The central government would be responsible for national defense and for maintaining a sound currency. It would have no other responsibilities. The city-states and county-states would be state governments. Apart from national defense, each state government would have as much responsibility as Denmark. It would run its own Social Security system, its own health care system, and so on.

One issue concerns what happens when people move from one state to another I would propose giving people the option to choose whatever Social Security system or health care system they want, regardless of where they live. However, if you change your mind and want to pick a different system, you may have to pay an exit fee to the old system or an entrance fee to the new system. A court or bureaucracy would have to be set up to determine exit and entrance fees that are fair.

Another issue is that of regional overlaps of concern. For example, roads and other forms of transportation infrastructure may cut across states. This would require negotiation among states, just as it does now. However, because there would be more, smaller states, with more responsibility, there would be more need to negotiate. If this becomes a problem, it might be necessary for states to agree to submit disputes to a binding arbitration board of some sort.

I envision this system looking like the European Union, but with no Eurocrats, a more common language, and more individual mobility across states.


Comments and Sharing


CATEGORIES: Political Economy



COMMENTS (19 to date)
david writes:

That binding arbitration board will grow in power, I venture.

Regional overlaps are common; greater New York derives much of its tax revenue from businesses in New York City (and, likewise, many of the people who work in NYC live and consume social services elsewhere). Existing states moderate this by applying mobility restrictions, but Kling presumably does not want this. Something else has to give.

Free-rider problems - states benefit when a nearby state invests in some beneficial policy. Who needs expensive Hong Kong-esque schools when you can send all your children over the border? Or Singapore-esque carefully-maintained public housing when all its occupants go to work and generate tax revenue for the state next door? Excessively small states means that the benefits of good governance are mostly externalized and would consequently be under-provided, reducing the benefit of interstate competition. Again, existing countries do moderate this by applying restrictions, but this is again not an option here.

States are going to want to make binding agreements with other states, and find ways to agree to punish defectors - hence why that arbitration board will grow in power.

(and all these are just economic issues. One day some state decides to ban abortion when its neighbor decides to legalize everything right up to third-trimester abortion. Oh, boy. Cue the start of mobility restrictions)

Alex J. writes:

That Kling's system wouldn't encourage public housing or public schooling is a feature, not a bug.

Eli writes:

The central government should also enforce an internal free trade area, no?

David R writes:

How would the national defense be funded? I'm guessing Arnold didn't forget to mention his ideas for a national defense tax...so would states be required to pay a fee, perhaps per capita or per state GDP?

I suppose that potential wars could be decided by state legislatures voting yea or nay.

david writes:

Alex J., I thought that the point of this elaborate scheme is to permit policy competition. If we know a priori that public housing (Singapore: 80% of population) and public schooling (Hong Kong: 80% of all students) were bad ideas, we could simply drop the pretense of interstate competition and apply these everywhere.

Regardless, the countries on Kling's favored Heritage-top-10-most-free list do have such massively interventionist policies, so it's clear that governments can sometimes do it extraordinarily well. Note that Singapore has strong public schooling and Hong Kong public housing as well, I just mixed them up for variety.

david writes:

To add more fuel to the fire, note that of all the top ten nations on that Heritage list, exactly one doesn't have universal healthcare. Guess which one it is!

Mike Vine writes:

This is very similar to the original, anti-federalist vision of the more libertarian founders.

I disagree that the whole federation should be broken down by some utopian program. Right now, jurisdictional evolution is practically illegal because it requires the consent of the existing state's legislature and that of the federal Congress. Provide a route for peaceful and legal secession and you would see an organic end to behemoths like California and New York.

More importantly, your project is already sunk by reserving the currency as a federal power. A proper confederation (which seems to be what you're after) should be a club of sovereign states. Making the states dependent on a federal central bank means that the federal government gets all the advantage from inflation and leaves the states as beggars for the value of their own citizens' savings.

A final point is that the confederal legislature should have equal representation for each member state. Both the US and EU attempt to form a relationship directly with their member states' citizens (direct election, popular legislative chambers). The confederation is to states as the states are to the people. Any attempt to make federal subjects of the states' citizens renders that states as mere administrative subunits of the "national" government -- which is exactly what has evolved in the US.

You're on the right track. I am encouraged that anti-federalism is making a comeback in the blogosphere. Keep up the good work.

David R. Henderson writes:

@david
Actually, no country has universal health care.

david writes:

@David R. Henderson

Here's number 2 on that list, then. See the first sentence. I can't find an English-language Hong Kong equivalent.

I know, I know - what you actually mean is that no country offers universal health care by your definition, but I venture that we don't get to redefine words when we like it. ;)

david writes:

Hm, my reply got nailed by the spam filter. Hopefully it'll show up soon.

Regardless - to avoid quibbling in what "universal" really should mean, how's this: the other nine countries have dramatically more expansive government intervention in the healthcare and healthcare insurance markets. This seems to work out pretty well.

If smaller governments really do have better government, as per Kling's previous post, it seems reasonable to assume that some of these policies play a role in said better government (even if all these policies do, as Caplan has speculated on this blog, is "buy off" populist sentiment whilst market-friendly policies are enacted in more crucial areas).

(and hence, contra Alex J., that these policies are not possible under Kling's scheme of numerous small governments is a weakness, not a strength)

(David): "If we know a priori that public housing (Singapore: 80% of population) and public schooling (Hong Kong: 80% of all students) were bad ideas, we could simply drop the pretense of interstate competition and apply these everywhere."

The World Bank and OECD online resources put a majority of pre-college enrollment in "public" (i.e., government-operated) schools. Other sources (e.g., G.T. Kurian's World Education Encyclopedia, Postlethwaite's "The Encyclopedia of Comparative Education and National Systems of Education") assert otherwise. Here's Postlethwaite: Hong Kong "While the reliance upon private education as a profit-making venture has become less acceptible, the pattern of dependence on management by voluntary agencies as increased and, of the 537,000 children in primary schools in 1985, just over 37,000 were in government schools. All kindergarten schools are still operated privately. At the secondary level, the picture is similar in that, of the enrollment of 401,200 up to form 5 (grade 11), only 31,400 are in government schools."

Michael Lee, then Assistant to the Minister of Education, gave similar figures (about 90% in independent schools) 15 years ago (e-mail correspondence).

The World Bank and OECD online figures are identical, which implies that these are not independent verification but an echo. I suspect that some World Bank or OECD researcher used an idiosyncratic definition of "public" to justify deliberate misdirection.

(David): " Free-rider problems..."
Government does not solve free rider problems. Corporate oversight is a public good. The State is a corporation. Therefore, oversight of State functions is a public good which the State itself cannot supply. State assumption of responsibility for the solution of "free rider" problems transforms these problems but does not eliminate (or solve) them.

Eduardo Zambrano
"Formal Models of Authority: Introduction and Political Economy Applications"
Rationality and Society, May 1999; 11: 115 - 138.
"Aside from the important issue of how it is that a ruler may economize on communication, contracting and coercion costs, this leads to an interpretation of the state that cannot be contractarian in nature: citizens would not empower a ruler to solve collective action problems in any of the models discussed, for the ruler would always be redundant and costly. The results support a view of the state that is eminently predatory, (the ? MK.) case in which whether the collective actions problems are solved by the state or not depends on upon whether this is consistent with the objectives and opportunities of those with the (natural) monopoly of violence in society. This conclusion is also reached in a model of a predatory state by Moselle and Polak (1997). How the theory of economic policy changes in light of this interpretation is an important question left for further work."

Devin Finbarr writes:

Having a popular election choose one leader for an entire life time seems a bit scary. I don't trust the masses to choose a fully sovereign executive, and especially not a lifetime executive(although the recall possibility would perhaps help).

I would have an Electoral College for each city that chooses the King, I mean, governor. The Senate would include 24 members, each appointed for one term of ten years (or until the age of 75). The members comprise: six retired generals(with new members chosen by a vote of the army, with votes weighted by seniority). Six scholars, appointed by a vote of the faculty of the top universities. Six seats to represent the business class, sold via auction. Six seats chosen by a Venetian style combo election/lottery (with no campaigning allowed).

The only job of the College would be to appoint a executive, or to fire him. They could fire him for any reason. The point of the structure of the college is to align formal power with actual power. The structure is balanced between the major castes of a society (the soldier, the merchant, the scholar, and the average worker). But the numbers of the college are small enough to hopefully prevent major factional infighting.

Another option is to simply have each city run as a joint stock corporation. It works extremely well for companies, why not for city-states? The city-states are small enough that they will actually have an incentive to compete for talent.

david writes:

@Malcolm Kirkpatrick

There are many examples of privately-provided public goods that have been given on this blog; Kling has previously mentioned these and I believe he is sympathetic to the concept.

The mechanism is typically iterated checks and institutions permitting defector punishment - the root of many public-goods problems is often the prisoner's dilemma, so it is quite possible to rationally provide these under such changes. And states - which are more capable of strong-arming information out of a third party than private entities - are presumably best placed to perform such checks.

As for Hong Kong, yes, the researchers are using a idiosyncratic definition. Like Singapore, Hong Kong has a profusion of nominally religious and Chinese charity schools due to its cultural history. However, these schools are heavily subsidized, obliged to teach the government-prescribed syllabus, pay government-set wages, and charge government-set fees, so they are certainly not private. Their main distinction is being able to admit students by academic merit than by geographic locale.

To improve competition, there are selected schools that are permitted to set their own fees and decide on their own syllabus, but this is conditional on the government granting such permission, and such schools are still heavily subsidized anyway.

There are other, smaller schools which are genuinely private, charge non-subsidized fees, and are typically aimed at the children of expatriates (who might be studying the syllabus of, say, Canada rather than Hong Kong). But as noted, these are relatively few.

(David): "There are many examples of privately-provided public goods..."

No argument here. This observation weakens the case for State intervention, seems to me.

The "public goods" analysis is a mirror image of the "tragedy of the commons" analysis. A public good is usually the amelioration of a public bad (tradedy of the commons). The tragedy of the commons is an interated multiplayer prisoner's dilemma with memory.

(David): "Like Singapore, Hong Kong has a profusion of nominally religious and Chinese charity schools due to its cultural history. However, these schools are heavily subsidized (1), obliged to teach the government-prescribed syllabus (2), pay government-set wages, and charge government-set fees, so they are certainly not private (3). Their main distinction is being able to admit students by academic merit than by geographic locale. To improve competition, there are selected schools that are permitted to set their own fees and decide on their own syllabus, but this is conditional on the government granting such permission, and such schools are still heavily subsidized anyway."

(1) The "public goods" argument implies subsidy and regulation, at most, not government operation of an industry.
(2) The State cannot subsidize education without a definition of "education".
(3) I make less of the public/private distinction than most and less of the for-profit/non-profit distinction than most.

MernaMoose writes:

I like what you're playing with here Arnold.

But I think, the Fed needs a little more power than you gave it. Immigration control for one thing. And the power to specify and enforce the Ts&Cs that make the whole thing a giant free trade / free movement of people zone internally.

Perhaps also another thing or two that isn't coming to mind just now.


The idea of having to buy your way in and out of jurisdictions for welfare purposes -- hmm. Maybe with the right T's&C's this could work. Could we have state funded welfare/medicine competing with private solutions, within the same locale? That'd be ideal, and we could maybe find out what really is the most cost effective solution to the problem.


And dictator (king, whatever) for life in the big city regions? A bit scary. I can sort of see the logic that says you don't need as much representation in a big city environment, but I somehow see the Dictatorship being in a state of perpetual recall. This needs a little more thought.

As far as legislatures, how about appointing an Upper and Lower House on terms something like jury duty (maybe a little easier to get out of if you aren't interested or consider yourself not up to it). There should be requirements (education, income/asset levels, or similar) to be eligible for appointment to either House. But the whole idea behind an Upper and Lower House should be retained.

This would at least help mitigate some of the inherent flaws in democracy. Maybe provide mandatory training (with an exam you must pass?) at the front end of your House duty. To insure the members understand their responsibilities, duties, restrictions, etc.

As far as voting, across the board, those who pay little to no taxes should not get votes of equal weight to those who pay lots of taxes. To prevent the masses from voting themselves big welfare entitlements, for one thing. To recognize the fact that those who are footing the lion's share of the bills, should have more to say about it. Although, maybe they just naturally get it that way, anyway?


Finding the right balance between Fed and local government is tough. The Fed needs enough power to make it One Nation. But not so much power that it stifles competition and innovation.

In any case, once we get this balance figured out -- and it will probably shift over time! -- then we face the bigger problem: how to restrain the inevitable growth of the bureaucracies that will necessarily come to exist? Assuming jury duty appointments to the legislatures kill off most of the inherent problems with elected politicians.

Cassander writes:

Life tenure made much more sense in an era when people tended to die quickly rather than slowly. In the modern world of degenerative illnesses, it makes much more sense to appoint leaders to either a fixed term or until a fixed age then give them exceedingly generous severance pay.

Plans to Prosper writes:

This strikes me as an excessively centrally-planned approach to making things less centrally planned. I'm inclined to agree with Mike Vine. Make it easier for jurisdictions to change on their own, when they want to, and over time we'll see organic movements to aggregate small-population jurisdictions and break up large-population ones.

Arnold assumes apparently a priori that New York City is better off separated into multiple city-states. But why 500,000 people in 100 square miles? Why not 450,000 people in 112 square miles, or one million people in 33 square miles? If you're going to arbitrarily choose a number, why does it have to be a nice round number? It seems to me that the people of Manhattan, Brooklyn, Columbus, etc. would all be better off deciding for themselves what sort of jurisdiction they would form, and how large, and whether it's separate or not from nearby jurisdictions. What works for New York may not work for LA or Seattle or anywhere else for that matter. In my opinion, that's the primary advantage to federalism in the first place.

Jeremy, Alabama writes:

Arnold is not trying to limit the power of all government.

He is trying to limit the SCOPE of the federal government (ie the part that individuals cannot escape), and introduce competition into all other levels of government (where individuals at least have a choice).

Each person, individually, gets to choose what level of taxes/services they want. If "free" health care and education is your bag, there is bound to be a polity that you find attractive. Rich people can already escape the federal government, because they can afford to live elsewhere. Arnold would like the middle class to have these choices.

For me, it's not a question of whether an idealized "conservative" government is better than an idealized "liberal" one, but whether an incremental change should be more-govt or less-govt. Less, please.

Liberals seem to be moving the opposite way. More powers, more taxes, more global.

Max writes:

I made an argument similar to Arnold's here:

http://athousandnations.com/2009/10/20/towards-youtopia-are-all-public-good-providers-earthbound/

It's time to start a serious conversation about the idea of radical federalism, poly-centric government and a "right of exit."

Comments for this entry have been closed
Return to top