Tax competitiveness: was Charles Tiebout joking?

The Tiebout model of cross-border human migration

From the Fools’ Gold project on ‘competitiveness’:

Do nations or states ‘compete’ with each other in a meaningful way? We have already explored the thinking of Paul Krugman, Adam Smith, Robert Reich, and the Tax Justice Network on this question. Their answers are, to summarise broadly: ‘no – or at least not in the way people commonly suppose.’

This ‘competition’ between states, we’ve argued, bears no economic relation to the microeconomic competition between firms or companies in a market. The shortest way to illustrate this, perhaps, is to note that a failed company is one thing: a failed state is another beast altogether.

But there are influential people who disagree. Take, for example, a speech by then Swiss president Ueli Maurer, to the World Economic Forum at Davos in 2013. As he put it:

“Locational competition exists within our own borders. This leads to good infrastructure, to restraint in creating red tape and to low taxes. And individuals benefit just as much from all of this as businesses. . . Diversity stimulates competition. That is not only the case in business, but also in politics.”

In the area of international tax this is a common world view: a quick search on a news site for “Charles Tiebout,” the subject of today’s article, yields many results just for the last few months: not bad for a theory that is nearly six decades old.

So where did this way of thinking come from?  Does it have any real intellectual credentials?

A pure, distilled theoretical model of local spending

We can trace the idea behind Maurer’s words to a 1956 paper by Charles Tiebout, a then little-known U.S. economist. Entitled A Pure Theory of Local Expenditures, it has since become one of the most widely cited articles in economics, and it may be the most widely cited in the tax area over the long term.

Forgive us, but we are going to be a little rude about this paper today.

Under Tiebout’s model, voters vote with their feet – and this has profound implications.

In short, each state offers different packages combining taxes with a bundle of public services, and people move about between these jurisdictions according to which mix of tax and public goods they like best. People will, in the jargon, maximise their personal utility by moving to the best jurisdiction for them. Governments, for their part, always find it hard to know what people want, but under Tiebout’s magic process of voting-with-the-feet people ‘reveal their preferences’ at last, and governments can ‘discover’ an equilibrium provision of local public goods against taxes. The result is what has been called the sorting of populations into ‘optimal communities’. All this is ‘efficient’.

Tiebout even invokes Adam Smith:

“each individual, in seeking as a competitive buyer to get to the highest level of indifference subject to given prices and tax, would be led as if by an Invisible Hand to the grand solution of the social maximum position.”

(Also see Matthew Watson’s article that looks at Adam Smith in a ‘competitiveness’ context.)

You cannot be serious

To summarise Tiebout’s central premise, communities behave like firms, and voters (or “voter-consumers,” as Tiebout put it), behave like consumers. Countries ‘compete’ in the same way that companies do.

This directly contradicts to the way we see things here at Fools’ Gold.

It doesn’t take an awful lot of head-scratching to see that this model is – as  Tiebout himself may have felt until he got carried along on a wave of others’ enthusiasm (see box) – a joke.  Not only that, but Tiebout’s paper itself was very cautious about whether it worked in the real world. As he puts it:

“Those who are tempted to compare this model with the competitive private model,” he observes in his conclusion, “may be disappointed.”

Those pesky assumptions

Now consider some of the assumptions that are required for this model to work. Tiebout himself recognised many of these, explicitly or implicitly, and it’s not for nothing that his paper’s title contains the words “a pure theory.” A clash with the real world leaves the Tiebout model in tatters. Here they are.

First, it assumes that hordes of citizens will flit back and forth costlessly between different jurisdictions, ripping their kids regularly in and out of schools, at the drop of a tax inspector’s hat. There is no kickback from people with baseball bats who don’t like immigrants, and no need to show your passport at the border.

That assumption alone, in my opinion, basically kills Tiebout’s model stone dead. But there are others.

Second, it assumes that people consume all their public goods where they pay their taxes. There is no corporate profit-shifting, and there are no tax havens.  There is no commuting from one jurisdiction to another, and no poaching of skilled workers educated at other taxpayers’ expense. Students all work and pay taxes in the state where they originally studied, and don’t choose the high-tax states with good education funding, before moving on to work and earn income in a lower-tax state. (Or, to use a more mischievous example, there is no room in Tiebout’s model for people like former UK Finance Secretary Nigel Lawson, who spent his working life cutting and opposing ‘high’ taxes – then moved to spend much of his retirement in high-tax France.) Public goods, such as clean air and water, don’t spill over into other jurisdictions; externalities such as cross-border pollution and crime are brushed aside. And on, and on, and on.

Again, this bundle of issues singlehandedly demolishes Tiebout’s thesis.

Third, everyone in these ‘optimal communities’ is blessed with perfect knowledge of what’s out there: a complete understanding of the tax system and its future evolution, and of the intricacies of government spending. (Again, GIGO: garbage in, garbage out.)

Fourth, there are loads of communities to choose from, and everyone understands them all – perfectly. No moving hassles, and no tears.

Fifth, to assume away all restrictions imposed by jobs and their availability, everyone is living on dividend income alone. (In other words, nobody does any real work.) Good grief.

Sixth, there is an optimum community size, and communities know what this is and strive to attain this ‘optimum’ number of residents. (He appears to have inserted this assumption partly in order to be able to make this optimum size, as he put it, “closely analogous to the low point of a firm’s average cost curve.”)

It is true that a number of papers since 1956 have tried to finesse Tiebout’s model, to get around these objections. But these assumptions are so fundamental – particularly the first three – that it’s impossible to lose their murky bathwater without losing the baby too.

It is also true that Tiebout said these processes are more likely to work at the local level than at the international level: in other words, you’ll be more likely to move from Canton Zürich to Canton Schwyz than to Timbuktu. But this can actually make some of the other assumptions even more problematic: such as the idea that there is no commuting, or that nobody consumes the public goods (or suffers spillovers and externalities) created in a neighbouring state.

Never mind the nonsense: here’s the economics

But these multiple stakes to the heart never killed this theory. On the contrary, it flourished.

To those who, like the acolytes of Friedrich Hayek, were becoming increasingly hostile towards government, the Tiebout model was manna: a non-political solution to the problem of free-riding, which got away from the need to enter into ugly, difficult processes such as international co-operation and co-ordination. It would also become hugely attractive to a community of academic economists enamoured with the possibility of shoe-horning the kerfuffle of the real world into elegant mathematical models.

Complexities be damned: you can simply them assume all away.  And hey Presto! Competition between jurisdictions is just like competition between firms – and this is a good thing, right? It’s a race to the top, not a race to the bottom, after all.

And there are clear winners from this, what one can only call wilful myopia on the part of those who have taken the time to study it. As Will Davies argued recently:

“What the language of ‘national competitiveness’ does, in ways that are not simply mercantilist, is to offer a powerful analogy: the nation state is a giant corporation. Those that benefit most from this analogy are corporations themselves, although the handful of ‘super-rich’ individuals who now move between a number of global cities also benefit.
. . .
In my book, The Limits of Neoliberalism, I define neoliberalism as the “disenchantment of politics by economics”. Carefully and deliberately, the language of politics (of law, democracy, the public, the state) has been replaced by that of economics (of efficiency, management, incentives, governance). This is facilitated through the application of measurement tools (such as competitiveness evaluations) to criticize public institutions.

An inversion takes place, where states become ‘like giant corporations’, rather than corporations being ‘like mini states’. Subjecting laws and regulations to forms of cost-benefit analysis, as if their purpose were indistinguishable from efficiency, is a symptom of this. This then opens the door to private consultants to advise on how such efficiency might be increased, while concealing the question – ‘efficiency for whom?’”

At the end of all this, we shouldn’t be too hard on poor old Tiebout himself. He didn’t always take his own model very seriously, and he clearly had a sense of humour, as you can see from a letter he wrote a decade before his death in 1968:

“By the time of publication I shall not be, “also of Northwestern University.” I shall be at UCLA. The reason is obvious. The level and pattern of public goods provided in Santa Monica is preferable to those in Evanston.”

From what I can gather, Tiebout himself was a pretty nice guy. But his model is nonsense.

And it is the worst kind of nonsense: dangerous, influential nonsense.