ECONOMISTS are often accused of greeting some item of news with the observation, “That may be so in practice, but is it true in theory?” Sceptics too seem much more interested in superficially plausible theories about the diminishing power of the state than in the plain facts.

In practice, though perhaps not in theory, governments around the world on average are now collecting slightly more in taxes—not just in absolute terms, but as a proportion of their bigger economies—than they did ten years ago. This is true of the G7 countries, and of the smaller OECD economies as well (see chart 6). The depredations of rampant capitalists on the overall ability of governments to gather income and do good works are therefore invisible. These findings are so strange in theory that many economic analysts have decided not to believe them.

Tax burdens vary a lot from country to country—something else which is wrong in theory. Despite the variations, governments in all the advanced economies are well provided for. The United States is invoked by some European anti-globalists as the land of naked capitalism, the nadir of “private affluence and public squalor” to which other countries are being driven down. Well, its government collected a little over 30% of GDP in taxes last year: an average of some $30,000 per household, adding up to roughly $3 trillion. This is a somewhat larger figure than the national income of Germany, and it goes a long way if spent wisely.

At the other extreme is Sweden, despite its celebrated taxpayer revolt of the early 1990s. Last year its taxes came to 57% of GDP, a savage reduction of three percentage points since 1990. Next comes Denmark, on 53%, fractionally higher than in 1990. And here's a funny thing. Sweden and Denmark are among the most open economies in the world, far more open than the United States. Denmark's ratio of imports to national income is 33%, compared with America's 14%. And in common with other advanced economies, neither of these Scandinavian countries has capital controls to keep investment penned in.

Harvard's Dani Rodrik, one of the more careful and persuasive globalisation sceptics, has written: “Globalisation has made it exceedingly difficult for government to provide social insurance...At present, international economic integration is taking place against the background of receding governments and diminished social obligations. The welfare state has been under attack for two decades.” Sweden, admittedly, is reeling, its government now able to collect only 57% of GDP in tax. But plucky Denmark is resisting these attacks well, and so is most of the rest of Europe.

Even if taxes were falling precipitously, it would be absurd to claim, as many globalisation sceptics do, that companies are nowadays more powerful than governments. It is routine to be told, as in “The Silent Takeover”, a new book by a Cambridge University academic, Noreena Hertz, things like this: “51 of the 100 biggest economies in the world are now corporations.” Quite what that implies is never explained: readers are invited to draw their own conclusion about the relative power of governments and companies.

Before you even think about whether it makes sense to weigh corporate power against state power, you can see that this particular comparison, which measures the size of companies by their sales, is bogus. National income is a measure of value added. It cannot be compared with a company's sales (equal to value added plus the cost of inputs). But even if that tiresome, endlessly repeated error were corrected, there would be no sense in comparing companies with governments in terms of their power over people.

The power of even the biggest companies is nothing compared with that of governments

The power of even the biggest companies is nothing compared with that of governments—no matter how small or poor the country concerned. The value added of Microsoft is a little over $20 billion a year, about the same as the national income of Uruguay. Does this make it remotely plausible that Bill Gates has more sway over the people of Uruguay than their government does? Or take Luxembourg—another small economy with, presumably, a correspondingly feeble state. Can Microsoft tax the citizens of Luxembourg (whose government collected 45% of GDP from them last year), conscript them if it has a mind to, arrest and imprison them for behaviour it disapproves of, or deploy physical force against them at will? No, not even using Windows XP.

But those are specious comparisons, you might reply. Of course Bill Gates is less powerful than the government of Uruguay in Uruguay, but Mr Gates exercises his power, such as it is, globally. Well then, where, exactly, is Mr Gates supposed to be as powerful in relation to the government as the alarming comparison between value added and national income implies? And if Bill Gates does not have this enormous power in any particular country or countries, he does not have it at all. In other words, the power that Mr Gates exercises globally is over Microsoft. Every government he ever meets is more powerful than he is in relation to its own citizens.

In a war between two countries, national income is relevant as a measure of available resources. If companies raised armies and fought wars, their wealth would count for something. But they don't, and couldn't: they lack the power. Big companies do have political influence. They have the money to lobby politicians and, in many countries, to corrupt them. Even so, the idea that companies have powers over citizens remotely as great as those of governments—no matter how big the company, no matter how small or poor the country—is fatuous. Yet it is never so much as questioned by anti-globalists.

Any power to tax, however limited, gives a country more political clout than Microsoft or General Electric could dream of. But how can a small, exceptionally open economy such as Denmark manage to collect more than 50% of GDP in taxes, in utter defiance of the logic of global capitalism? The answer seems inescapable: Denmark no longer exists, and questions are starting to be asked about the existence of many other European countries. At least, that is how it looks in theory; in practice, the theory needs to be looked at again.

The alleged squeeze on government arises from the fact that, in a world of integrated economies, again in Mr Rodrik's words, “owners of capital, highly skilled workers, and many professionals...are free to take their resources where they are most in demand.” The people Mr Rodrik refers to have high incomes. Through the taxes they pay, they make an indispensable contribution to the public finances. If economic integration allows capital and skills to migrate to low-tax jurisdictions, the tax base will shrink. Governments will find themselves unable to finance social programmes, safety nets or redistribution of income. Anticipating this flight of capital and skills, governments have to cut taxes and dismantle the welfare state before the migration gets under way. Markets triumph over democracy.

That is the theory. Experience largely refutes it, but it is not entirely wrong. In a variety of ways, economic integration does put limits on what governments can do. However, some of those constraints are eminently desirable. Integration makes it harder to be a tyrant. Governments have been known to oppress their subjects. Oppression is more difficult with open borders: people can leave and take their savings with them. In such cases, global markets are plainly an ally of human rights.

The affinity of totalitarianism and economic isolation was obvious in the case of the Soviet Union and communist Eastern Europe; it is still plain today in the case of North Korea, say. But democracies are capable of oppression too. It would therefore be wrong to conclude that integration is undesirable merely because it limits the power of government, even if the government concerned is democratic. One needs to recognise that some constraints on democracy are desirable, and then to ask whether the constraints imposed by markets are too tight.

These issues are rarely, if ever, addressed by the critics of globalisation: it is simpler to deplore the notion of “profits before people”. The sceptics either insist, or regard it as too obvious even to mention, that the will of the people, democratically expressed, must always prevail. This is amazingly naive. Even the most elementary account of democracy recognises the need for checks and balances, including curbs on the majoritarian “will of the people”. Failing those, democracies are capable of tyranny over minorities.

The sceptics are terribly keen on “the people”. Yet the idea that citizens are not individuals with different goals and preferences, but an undifferentiated body with agreed common interests, defined in opposition to other monolithic interests such as “business” or “foreigners”, is not just shallow populism, it is proto-fascism. It is self-contradictory, as well. The sceptics would not hesitate to call for “the people” to be overruled if, for instance, they voted for policies that violated human rights, or speeded the extermination of endangered species, or offended against other values the sceptics regard as more fundamental than honouring the will of the majority.

The possibility that people might leave is not the only curb that economic integration puts on government. The global flow of information, a by-product of the integration of markets, also works to that effect. It lets attention be drawn to abuses of all kinds: of people especially, but also of the environment or of other things that the sceptics want to protect. Undeniably, it also fosters a broader kind of policy competition among governments. This works not through the sort of mechanical market arbitrage that would drive down taxes regardless of what citizens might want, but through informing voters about alternatives, thus making them more demanding.

The fashion for economic liberalisation in recent years owes something to the remarkable success of the American economy during the 1990s: a success which, thanks to globalisation, has been seen and reflected upon all over the world. Growing knowledge about the West helped precipitate the liberation of Eastern Europe. But information of this kind need not always favour the market. For instance, the failure of the American government to extend adequate health care to all its citizens has been noticed as well, and voters in countries with universal publicly financed health-care systems do not, on the whole, want to copy this particular model. The global flow of knowledge creates, among other things, better-informed voters, and therefore acts as a curb on government power. This does nothing but good.

The anti-globalists themselves, somewhat self-contradictorily, use the information-spreading aspect of globalisation to great effect. Organising a worldwide protest movement would be much harder without the World Wide Web, but the web itself is merely one dimension of globalisation. The economic integration that sceptics disapprove of is in many ways necessary for effective resistance to the more specific things they object to—not all of which, by any means, are themselves the products of globalisation.

Still, all this is to acknowledge that economic integration does limit the power of government, including democratic government. The question is whether it limits it too much, or in undesirable ways. So far as far public spending is concerned, the answer seems clear. Given that even in conditions of economic integration people are willing to tolerate tax burdens approaching 60% of GDP, and that tax burdens of between 40% and 55% of GDP are routine in industrial economies other than the United States, the limits are plainly not that tight. These figures say that democracy has plenty of room for manoeuvre.

One puzzle remains: why are taxes not coming down? There are several answers. One is that international integration is far from complete, and is likely to remain so. Technology has caused distance to shrink, but not to disappear. National borders still matter as well, even more than mere distance, and far more than all the interest in globalisation might lead you to expect. For all but the smallest economies, trade and investment are still disproportionately intranational rather than international. Especially in the developed world, borders still count not so much because of overt protectionist barriers, but because countries remain jurisdictionally and administratively distinct. This is not likely to change in the foreseeable future.

For instance, if a supplier defaults on a contract to sell you something, it is much easier to get legal redress if your seller is in the same country (and subject to the same legal authority) than it would be if you had to sue in a foreign court. Because of these difficulties in contracting, trading across borders still calls for much more trust between buyers and sellers than trading within borders—so much so as to rule out many transactions. This remains true even in systems such as the European Union's, where heroic efforts have been made to overcome inadvertent obstacles to trade, suggesting that they will prove even more durable everywhere else.

You would expect the international mobility of capital to be especially high, given that the costs of physical transporting the stuff are virtually zero, yet it is surprising just how relatively immobile even capital remains. In the aggregate, the flow of capital into or out of any given country can be thought of as balancing that country's saving and investment. If the country invests more than it saves (that is, if it runs a current-account deficit), capital flows in; if it saves more than it invests (a current-account surplus), the country must lend capital to the rest of the world. Perfect capital mobility would imply that, country by country, national saving and investment would move freely in relation to each other. Very large inflows or outflows of capital in relation to national income would be the order of the day. In fact they are not. Nowadays, a surplus or deficit of just a few percentage points of GDP is regarded as big.

Still, capital is much more mobile than labour—and mobile enough, to be sure, to have given rise to some tax competition among governments. So far this competition has affected the structure of tax codes rather than overall tax burdens; total yields have been unaffected. In an effort to attract inflows of capital, and especially inflows of foreign direct investment, governments have been lowering their tax rates for corporate income and raising them for personal income, or relying more on a variety of indirect taxes, or both (see chart 7). But it is easy to exaggerate the extent even of this structural shift, never mind the effect on total taxation. This is because taxes on corporate income were small to begin with, so not much was at stake. In fact, heavy reliance on corporate taxes is bad policy even in a closed economy. Indeed, in a closed economy, you can make a respectable case on efficiency grounds for excluding corporate income from taxes altogether.

Taxes on company profits, the argument goes, are taxes on shareholders' income—ultimately, that is, taxes on a particular category of personal income. In the end, although it is politically convenient to pretend otherwise, “the people” pay all the taxes: companies are mere intermediaries. There is no reason to tax the income people receive as shareholders any differently from the income they receive as owners of bank deposits or as workers. In a closed economy, you might as well abolish the corporate-income tax and instead tax profits when they turn up as dividends in the incomes of individual taxpayers: it is simpler, and it is less likely to affect investment decisions in unintended ways.

In an open economy, however, company ownership is to some extent in the hand of foreigners, not just the citizens of the country where the company is based. This makes it more tempting to tax corporate income, because this allows the government to bring foreigners within the scope of its tax base. Seen this way, it is odd to blame globalisation for downward pressure on corporate-tax rates. Were it not for globalisation, there would be no reason to have corporate taxes in the first place. But it is true that once you are collecting corporate taxes, greater capital mobility limits your take. Economic integration rationalises, and at the same time limits, reliance on corporate-income taxes. The issue is subtler than it seems.

But what matters far more than corporate tax policy is that most people, skilled as well as unskilled, are reluctant to move abroad. Since workers tend to stay put, governments can tax them at surprisingly high rates without provoking flight. In all but extreme cases, the democratic constraint (the need to secure a broad measure of popular support for tax increases) binds governments long before the economic constraint imposed by international integration (the risk that groups facing very high taxes will leave). In the case of taxes on profits, it is true that the economic constraint will bind before the democratic one, and that globalisation serves to tighten the economic constraint further—but this does not matter. There is no need for high taxes on profits if people are willing to hand over 50% or more of what they produce in the form of taxes on income and consumption.

To simple-minded believers in the most desiccated branch of neoclassical economics, all this may seem surprising. Their theories regard people as “rational economic men”, narrow utility-maximisers with no ties to family, place or culture. Presumably, these ciphers would shop around for low-tax jurisdictions. Oddly, the same benighted view of human nature must be shared by many globalisation sceptics—otherwise, why would they fear taxpayer flight on a scale sufficient to abolish the European welfare state? But in real life, it is better to take a fuller, broader view of the human condition. Since people seem to choose to be tied down, indeed to relish it, governments, within broad limits, can carry on taxing them regardless of globalisation. If it seems prudent to cut taxes on profits in order to attract inflows of foreign investment, no problem. Taxes on people will still be sufficient to finance generous public spending of every kind.

Many anti-globalists have strangely little confidence in the merits of the policies they are anxious to sustain. Fearing what may be lost if globalisation continues uncurbed, Mr Rodrik writes:

If it was the 19th century that unleashed capitalism in its full force, it was the 20th century that tamed it and boosted its productivity by supplying the institutional underpinnings of market-based economies. Central banks to regulate credit and the supply of liquidity, fiscal policies to stabilise aggregate demand, antitrust and regulatory authorities to combat fraud and anti-competitive behaviour, social insurance to reduce lifetime risk, political democracy to make the above institutions accountable to the citizenry—these were all innovations that firmly took root in today's rich countries only during the second half of the 20th century. That the second half of the century was also a period of unprecedented prosperity for Western Europe, the United States, Japan and some other parts of East Asia is no coincidence. These institutional innovations greatly enhanced the efficiency and legitimacy of markets and in turn drew strength from the material advancement unleashed by market forces...The dilemma that we face as we enter the 21st century is that markets are striving to become global while the institutions needed to support them remain by and large national...The desire by producers and investors to go global weakens the institutional base of national economies.

The argument, presumably, is that international capital will flow away from countries with the high public spending and taxes that these highly developed institutions involve. One answer is that international investment, as already noted, is much less important in most countries than domestic investment. But a more fundamental question is this: why should foreign capital flow away from countries that have equipped themselves with these institutions, if, as Mr Rodrik emphasises, those arrangements have “boosted...productivity” and “greatly enhanced the efficiency...of markets”—so much so that the most ambitious period of national institution-building was also a time of growing and “unprecedented” prosperity for the nations that joined in?

If public spending boosts productivity, then competition among governments for inward investment is likely to favour more public spending (and the taxes needed to pay for it), not less. Suppose, as seems plausible, that public spending on education raises productivity by increasing the supply of skilled workers. Then you would expect international investment to be drawn to countries that invest heavily in top-quality schools and universities. Suppose, as may also be true, that public spending on social programmes such as health and welfare raises productivity, by producing a healthier and more contented workforce, with better labour relations and greater labour mobility. If so, again international capital will be drawn to countries that spend money on those things. Globalisation, surely, will not frown on policies whose net effect is to foster productivity and efficiency.

But what about policies that do not serve those goals? Many would argue, for instance, that welfare policies, especially if too generous, encourage idleness and reduce economy-wide productivity. Suppose that is true. Also suppose that, knowing it to be true, most people want such policies anyway. You might feel that they are entitled to that opinion, and in a democracy they are entitled to get their way. Another example might be policies to limit working hours. Suppose that they reduce productivity, but that people vote for them anyway. Must globalisation overrule democracy?

The answer even in this case is no—and to see why is to understand why so many of the fears about globalisation and democracy are groundless. Policies that reduce productivity do, in the first instance, cut a country's feasible standard of living, narrowly defined in terms of GDP per head. But what happens after that? If a country that is open to international trade and capital flows adopts some such policies, perhaps on the ground that they will raise living standards according to some broader definition, wages and profits will fall relative to what they would otherwise have been. Next, investment will fall and the capital stock will shrink, again compared with what they would otherwise have been. This will continue until the scarcity of capital drives the rate of profit back up, at the margin, to the rate prevailing in the global capital market.

All this time the economy will grow more slowly than if the policies had not been followed. Once the economy has adjusted, however, it remains as “competitive” as it was at the outset: lower wages have restored labour costs per unit of output, and a smaller stock of capital has restored the return on capital. The economy has grown more slowly for a spell. It is less prosperous than it would have been. But in due course, once wages and profits have adjusted, the economy will again be as attractive, or unattractive, to foreign investors as it was at the outset. The government's adoption of policies that compromise efficiency is not punished by excommunication from the global economy, or with an accelerating spiral of decline; the only penalty is compromised efficiency and lower measured incomes, which is what the country chose in the first place.

Would the economy have fared any better without globalisation?

Would the economy have fared any better without globalisation? Had it been closed to international flows of goods and capital, could it have adopted those productivity-cutting policies and paid no price at all? The answer is no. Even in a closed economy, policies that reduce productivity would cause wages and profits to fall, as in the open-economy case. The return on capital would be lower, so saving and investment would decline, relative to what they would have been (there would be no cross-border capital flows in this case, so saving and investment must always be equal). The capital stock would shrink and growth would be held back until the scarcity of capital drove the return back up. As in the open-economy case, the result would be a spell of slower growth and a standard of living permanently lower than it would otherwise have been.

The main difference is probably that in the closed-economy case, the losses would be subtracted from an economy that is already very much poorer than its open-economy counterpart, because it is closed. Conceivably, this would make further losses politically easier to sustain. But that is the most you can say in defence of the view that globalisation forbids social policies which jeopardise productivity. “Stay poor, because once you start to get rich you may find that you like it.” Not exactly compelling, is it?

You might well conclude from all this that globalisation, if anything, will lead to higher rather than lower social spending. As argued earlier, globalisation raises aggregate incomes but at the same time increases economic insecurity for certain groups. Both of these consequences tend to raise social spending. Generous social spending is a “superior good”: as countries grow richer, they want to spend more of their incomes on it, and can afford to. At the same time, quite separately, greater economic insecurity directly spurs demand for social spending.

Given that globalisation increases the demand for social spending; given that it does not rule out any decision to increase such spending which harms productivity, any more than a closed economy would; given that increases in social spending which raise productivity will be rewarded with inflows of capital; given all this, should globalisation and the generous social spending that democracies favour not go hand in hand? They should, and indeed rising social spending alongside faster, deeper globalisation is exactly what the figures for the past several decades show.

Governments in rich countries need to look again at their social policies, partly to make sure that temporary and longer-term losers from globalisation, and from economic growth in general, get well-designed help. But there is no reason whatever to fear that globalisation makes social policies more difficult to finance. In the end, by raising incomes in the aggregate, it makes them easier to finance. It creates additional economic resources, which democracies can use as they see fit.

Copyright © 2001 The Economist Newspaper and The Economist Group. All rights reserved.

The OECD provides statistics on taxation in its member states. Dani Rodrik, a globalisation sceptic, posts articles online.