Does size (of government) matter?

August 17, 2004

When Canadians went to the polls last June they chose, in policy terms, new spending by the federal government over shrinking government through more tax cuts. In so doing, they rejected arguments that higher public spending would inevitably undermine Canada's economic performance, or conversely, that reducing the size of government by cutting taxes would boost the economy.

They made the right choice. Statistically, there is no correlation between economic performance and the size of government. And to the extent that studies are marshaled in favour of small government, closer inspection usually shows that the data have been tortured to wring out a confession.

Canada has tax revenues (an indicator of size of government) of about 36% of GDP, a bit less than the average for the most advanced economies of the OECD (38%), and much smaller than the European countries (42%).

Eleven countries in the OECD are in the same income range as Canada (just under US$30,000 per capita), but have larger public sectors. Topping the list are Sweden and Denmark, the countries with the largest shares of taxes-to-GDP in the OECD (54.2% and 48.8%, respectively).

If "big government" led to weaker economic performance, the Scandinavian countries, which have much larger governments than Canada, should all be economic basket cases. But this is not the case. These countries have among the highest productivity and living standards in the world.

Also in the same income range as Canada are Japan and Australia, countries with shares of taxes-to-GDP closer to the low end (27.1% and 31.5%, respectively). Interestingly, the country with the lowest level of taxes to GDP, Mexico at 18.5%, was second-lowest in terms of income. Luxembourg, with a very high average income of US$50,600, nonetheless had taxes that amounted to 41.7% of its GDP.

The same basic result holds for other measures of economic performance and other indicators of the size of government. In a detailed review of the evidence by US economist Peter Lindert, he asks why European welfare states have not had the negative effect on growth that many economists assume it should. He finds that the actual experience of countries with large public sectors has been towards implementing pro-growth taxation and spending policies.

On the tax side, these governments have tended to tax capital lightly to avoid capital flight. They also tend to rely more on consumption taxes, particularly those for gas, alcohol, and tobacco. These taxes, while considered regressive, were introduced as part of a social bargain that the proceeds would fund beneficial social programs.

The flip side of taxation is public spending. On the spending side, welfare state countries have invested in public services, such as health care and child care, that have pro-growth impacts.

A lesson for Canada is that we should press forward with a national early childhood education and care program. Such a program makes good economic sense, both in terms of the labour market impacts for women, and the cognitive impacts for young children. University of Toronto economists Gordon Cleveland and Michael Krashinsky find a two-dollar return for each dollar invested in child care.

The enemies of "big government" fight their battle on ideological rather than economic grounds. Economically, the key questions seem to be not how much tax is taken as a percent of GDP, but what tax mix is used, and what is the money spent on. These considerations dwarf the simple idea that big government is bad for growth.

There is no economic reason why Canada could not expand its public sector by a significant margin. This would not kill the Canadian economy, and could greatly improve it if new expenditures went to pro-growth investments.

What is refreshing about this research is that it reinforces the idea that public policy is about making choices. We can choose to engage in good social policy without fear that the economic sky will come falling down on us.

Marc Lee is an economist in the BC Office of the Canadian Centre for Policy Alternatives, and is the author of "Size of government and economic performance: What does the evidence say," available at