The Clerk of the Privy Council's Blueprint 2020 exercise is a positive step in redefining the role and functions of the federal public service in Canada. It asks important questions about how the public service needs to evolve, what best practices it should adopt, and how the federal public service can come to represent a competitive advantage for the country. We laud the Clerk and the entire public service for this introspective initiative.

However, a key question omitted from the blueprint document is: what's the right size and scope of the federal government in 2020?

The fact is, the resources dedicated to public services can be reduced without a necessary trade-off in economic and social outcomes. Indeed, a smaller, more focused, and well-managed public service would produce better outcomes for Canadians in 2020 and beyond. To achieve this end, the federal government should undertake a systematic review of its programs and services using the Program Review experience of the 1990s as a model to ensure that it delivers on its core functions in the most efficient way possible.

In his recent annual report to the Prime Minister on Canada's public service, the Clerk stated: "to succeed, a leaner public service must be one that is continuously performing at peak productivity." Blueprint 2020 is an opportunity to deliver on this worthwhile objective.

The case for a focused, well-managed public service is not ideological. There is considerable empirical research on the size of government and its impact on economic and social outcomes. Economists have measured the effect of the size of government on economic growth and social outcomes such as life expectancy, infant mortality, homicide rates, educational attainment, and student reading proficiency. They have also explored the connection between the size of government and the efficiency of public sector outcomes across a host of indicators.

In a recently published study by the Fraser Institute entitled Measuring Government in the 21st Century, Lakehead University Professor Livio Di Matteo makes an important contribution to this literature. His research produced three important findings.

First, Professor Di Matteo finds there is a positive association between government spending and favourable social outcomes but much of the relationship is for lower amounts of spending with a leveling off of incremental improvements as the public sector's size exceeds a certain level. Specifically, he finds a positive correlation between government spending and social indicators like life expectancy and education attainment when the share of total spending goes from 20 percent to 35 percent of gross domestic product (GDP). But improvements in the performance indicators are generally limited once the size of government exceeds 35 percent. Thereafter any improvements are small and require substantial incremental increases in the share of total government spending in the economy.

Second, Professor Di Matteo uncovers a ceiling after which the size of government starts to have a negative effect on economic growth. After controlling for confounding factors through regression analysis, he finds that annual per capita GDP growth is maximized at three percent when government spending consumes 26 percent of the economy. Economic growth rates start to decline when relative government spending exceeds this level. In other words, there is a hump-shaped relationship between the size of government and economic growth with the sweet spot being achieved at 26 percent of GDP.

The premise that underpins these two findings may be a bit counterintuitive. It seems self-evident that government programs and services contribute to social and economic well-being. More police officers keep our streets safer. New infrastructure helps to promote commerce and trade. Additional funding for education helps children learn and grow into productive adults. The research shows that certain levels of public services do indeed contribute to the economy and help to achieve positive social outcomes. That is not in doubt. The fundamental question is whether there is a point at which incremental public spending impedes economic growth and only achieves higher social outcomes at great marginal cost.

Finally, Di Matteo estimates the efficiency of public sectors in 34 advanced countries. Efficiency is measured by accounting for the resources used to deliver various outcomes; higher efficiency results when fewer inputs (smaller government) produce greater outputs (better economic, health, and social outcomes). While Canada ranks eighth among the countries in efficiency, there is certainly room for improvement. According to Professor Di Matteo's calculations, if Canada's public sectors (all levels of government) were as efficient as South Korea's (the highest-ranking country), its total government expenditures would be 29 percent of GDP instead of the current 41 percent.

So what does this all mean for Blueprint 2020?

It means there is room to improve the efficiency of Canada's public sector and it can begin by using this exercise at the federal level to think about the government's core responsibilities and eliminate those activities that do not contribute positively to economic growth.

The blueprint document refers to regular reviews of programs and services 'to help the Government determine whether they are still required and whether adjustments are needed to ensure that they are effective, efficient, and focused on the needs of Canadians.' This is a good start but this type of review process implies tinkering on the margins rather than a more fundamental look at the right role and scope for the federal government.

The government should instead launch a process similar to the Program Review exercise of the 1990s to test the effectiveness of government organizations and their programs and services. Program Review was different than subsequent spending review exercises at the federal level in that its mandate involved asking questions about whether certain organizations and programs should be reformed, devolved to other levels of government or the private sector, or discontinued altogether. Departments applied six tests to all of their spending and activities in order to come to these decisions.

The exercise led to a significant structural change in the federal government's involvement in the Canadian economy, including a 60 percent reduction in subsidies to businesses and divestitures of major Crown corporations such as airports and the air navigation system. These reforms produced considerable fiscal savings, reduced the size and scope of the federal government, and ultimately helped usher in a period of sustained economic growth and job creation.

This final point is worth emphasizing: Canada's total government spending as a share of GDP fell from 53 percent in 1992 to 39 percent in 2007 and despite this 14 percentage point decline in the size of government the economy grew, the job market expanded, and poverty rates fell dramatically.

Professor Di Matteo's findings along with Canada's own experience tell us that an ambitious refocusing of the scope and size of government both federally and provincially can lead to improved services for Canadians and better economic outcomes. Blueprint 2020 should work towards that end.