Francis Fukuyama, in the WSJ (July 28, 2004):

At the beginning of the 20th century, Britain and America took in less than 10% of GDP in taxes; by the 1970s, that figure had grown to over 50% for most of continental Europe and 40% in the U.S. Communism, of course, represented the ultimate effort of the state to take over the whole of the economy and civil society. Ronald Reagan and Margaret Thatcher presided over a successful political and intellectual revolution (spurred on importantly by this newspaper's editorial page) to reverse the trend toward an ever-larger state sector. Communism collapsed, and the U.S. and Britain today face brighter economic futures than most of the rest of Europe precisely because of their freer labor markets and lower levels of welfare-state obligations.

But the agenda for the age opening before us has shifted in important ways. The violent politics of the 20th century was dominated by great powers, states like Nazi Germany or the former USSR that were too strong. Today, it is instead weak or failing states that are the source of international troubles like poverty, disease, refugees, human rights abuses, and, as has been vividly clear since September 11, terrorism. Somalia, Bosnia, Kosovo, Haiti, Afghanistan and now Iraq are all places with weak or absent state institutions, where the agenda is not cutting back the state but its opposite, state-building.

This new agenda does not contradict the central thrust of the Reagan-Thatcher years, though it is sometimes interpreted that way due to intellectual confusion over two very different dimensions of stateness. The scope of a state refers to the number of different functions the government takes on, from providing necessary public goods like property rights, public safety and defense, to intermediate ones like the regulation of various markets, and ultimately to ambitious ones like industrial policy or large welfare-state social interventions. On the other hand, for any given function, a state may be more or less competent in enforcing laws or carrying out policies. Thus state strength can be seen as an attribute separate from its scope. States can in this sense be simultaneously large and weak, jailing dissidents and overregulating markets, while being unable to process business licenses or maintain safety on the streets of the capital.

From the standpoint of long-term economic growth, it is best to have a state that is restrained in scope, focusing on the provision of necessary public goods, but strong in its ability to enforce a rule of law. Unfortunately, developing countries almost always fall into one of two categories: Either they are very ambitious in the scope of what they seek to achieve but weak in execution -- Venezuela, Argentina and Turkey come to mind -- or else try to do little and fail to do even that, like Liberia, Haiti or Afghanistan.

The Reagan-Thatcher revolution was properly directed against excessive state scope, seeking to reduce the level of regulation and government interference with private economic activity. But applied to the developing world, it often had a perversely damaging effect because it undermined state capacity across the board, or else made things worse by promoting liberalization in the absence of strong basic state institutions. Elites in sub-Saharan Africa used IMF-mandated structural adjustment programs in the 1980s to cut back on core state functions while increasing the size of the patrimonial state. Russia went from being a totalitarian state to a weak one that could not collect taxes or control "mafias"; privatization in the absence of a rule of law tainted the whole process. Thailand liberalized its capital markets in the early 1990s before it had adequate banking regulation in place; its inability to handle the ensuing flood of liquidity led to the financial crisis of 1997.

Development economists have increasingly come to the conclusion that strong institutions are critical to economic growth, more important, indeed, than simply cutting back on the scope of the state. A recent study by William Easterly and Ross Levine shows that institutions trump other factors like economic policies, geography or resource endowments in explaining growth ....