Archived — Discussion Paper Number 11: Social policy and productivity growth: what are the linkages?

by Richard G. Harris, Simon Fraser University, May 2002


The equity versus efficiency argument has been the bread and butter of economic policy and social policy discussions since the emergence of the modern welfare state in the post World War II period. In virtually all aspects of policy, the twin goals of promoting economic progress and social justice stands as a hallmark of the modern industrial democracy. By the late 1960s, the general view was that a conflict existed between the efficiency objective and the equity objective, nicely summarized in Okun's famous 1975 book: Equality and Efficiency: The Big Tradeoff.1 In the 1990s, a new debate has emerged covering similar, although conceptually different, ground. Productivity growth is widely regarded as the major long-run determinant of per capita income growth in industrial countries. Over the last two decades, economists have been pre-occupied with understanding the sources of productivity growth, and slow productivity growth in Canada has been a major policy concern for several years. Prior to the mid-1980s, traditional economic analysis focused on the static effects of economic policy — the so-called size-of-the-pie effects. For example, when looking at the impact of taxes on labour supply, the analysis was concerned with the one-time effect an increase in wage taxes could have on the labour supply, rather than its effect on long-run economic growth. However, it is evident that, in the longer term, how fast the pie grows is more important. The reason is simple: a small change in long-term growth rates — on the order of 1.0 percent, or even less — has dramatically larger consequences than a similar percentage change in GDP. This explains the emphasis put, in both research and policy, on understanding the factors leading to higher, or lower, productivity growth, as opposed to other factors that do not have permanent consequences on growth. Social policy might well be one factor that could have an impact on growth. The expansion of the welfare state was heavily dependent on strong economic growth in the 1950s and 1960s. The fiscal repercussions of slow productivity growth, which had set in by the mid-1970s and were evident in a debt and deficit build-up by the mid-1980s, raised concerns about the sustainability of high social spending. For both of these reasons, the dynamics of social policy became inevitably linked with the issue of economic growth.

That growth depends on productivity is not a fact in serious dispute; but the long-run sources, or ultimate determinants of productivity growth, are not completely understood. At the most general level, this is Adam Smith's question: What are the sources of the wealth of nations? At a more restricted level, there is agreement on the proximate sources of productivity growth — new investment, human capital formation, new technology and product innovation. What drives these factors in an economy has been accounted for largely by economic determinants, that is those impinging directly on investment, innovation, education and trade, which appear to have a direct and medium-term impact on productivity growth. However, recent research has put forward the hypothesis that social factors may also be a major determinant of productivity growth. Social factors would include the distribution of income and wealth in an economy, the range of social policy interventions including health, education, labour market regulation, and a variety of income support programs. These social policies may be defined to include the tax-transfer system, which finances the social budget. The implications of this change of perspective are potentially quite powerful in making a case for social policy. If it could be established that social determinants are a quantitatively major factor in productivity growth, then the traditional efficiency-equity tradeoff may not exist. Social policies to promote equity could also be defended on grounds that they simultaneously increase economic growth. The tradeoff is replaced by a virtuous circle in which equity-enhancing policies also promote economic growth. This paper provides a critical evaluation of these arguments.

In the paper, we present a survey of the evidence and debate on the social determinants of productivity in the context of the Canadian productivity debate. The paper examines both the basic theoretical arguments and the evidence advanced by economists, and their relationship to what might be called modern social policy. Not all social policy is directly motivated by equity considerations. In particular, modern social policies in the area of education and health focused on promoting the growth of human capital represent one category where both the evidence and debate on the growth effects are qualitatively different than in other areas of social policy.

It is instructive to consider the context in which this often heated, and at times politically loaded, debate surrounding the impact of social policy on economic growth has taken place. Three trends have been driving the wider debate in industrial countries — all of which are noticeable in Canada. First, the slow growth in Europe, particularly of employment, had led many to put the blame on the welfare state.2 Eurosclerosis became the term employed to describe the slow growth and poor employment record of a number of European countries through the 1980s and early 1990s. A parallel debate in the Scandinavian countries has led many to the conclusion that the Scandinavian welfare state had similar consequences. Assar Lindbeck's critique is one of the most well known.3 Part of the European record was the perception that generous social programs were a major factor responsible for the poor growth record. This debate was fuelled in part by the famous OECD Jobs Study (1994), and an attack by all OECD governments on the growth of debt and deficits in the mid-1990s. It may well be that the factors behind the slow employment growth in Europe ultimately have little to do with long-term productivity growth; but in the popular debate, the impacts of the European welfare state on productivity, employment and fiscal policy tend to get lumped together. Canada is typically viewed as somewhere between the United States and Europe on the welfare state spectrum, so that these arguments have likewise played out here.

A second major element, of more recent origin, is the debate on the new economy in the United States in contrast with the slow growth in Europe. The long and extraordinary economic expansion in the United States throughout the 1990s was accompanied by high employment and strong productivity growth. While the sources of this growth remain a matter of discussion, the new economy hypothesis claims that it is driven by the impact of innovations in the information, communications and telecommunications fields, giving rise to an entirely new phase of economic development — the so-called Third Industrial Revolution. Prior to the recent surge in growth, beginning in the mid- to late-1970s but continuing into the 1980s, there was a significant rise in market income inequality in the United States and the United Kingdom. These trends have subsequently shown up in most OECD countries, including Canada, but in Europe particularly it appeared that inequality was not increasing to the same degree. The acceleration of growth in the United States during the 1990s led some to infer that inequality seemed to contribute to growth. The divergent U.S. and European growth patterns in the 1990s has brought the charge that the re-distributive and labour market policies responsible for eurosclerosis have also prevented Europe from experiencing the growth benefits of the new economy. Economic growth and the preservation of equality as seen through this debate appear to be conflicting goals, reinforcing the old view that equity and growth are in opposition with one another.

Thirdly, an intellectual challenge to the existence of an equity-efficiency tradeoff emerged at about the same time the eurosclerosis debate began. From the mid-1980s, economists began to seriously re-think the sources of economic growth, which led to both to the New Growth Theory4 and to a large empirical literature on the determinants of growth and productivity. The development of new data sets for a large number of developing and developed countries allowed researchers to pose new and interesting questions about the sources of growth. Much, if not all, of the intellectual impetus to discover links between social factors and growth are found in this literature on cross-country growth comparisons. In the early 1990s, a number of researchers identified a robust negative empirical correlation between measures of inequality and economic growth — lower inequality would be associated with higher growth. Other researchers began to look for other policy determinants of growth, many of which bear directly or indirectly on the issue of social policy, such as education and fiscal policy. Lastly, a voluminous literature emerged on the rising wage inequality in advanced industrial countries over the last two decades. While not directly about productivity and social policy, the wage inequality issue figures prominently in the productivity-social policy debate for a simple reason. Much of this literature adopts the opposite perspective — what is driving inequality is economic growth, which in turn is driven by technological change. From this perspective, understanding the consequences of any policy intervention on inequality and growth requires an understanding of the complex interaction between technological change, productivity growth, and its implications for wages and employment.

My purpose in this paper is to try to make sense of these often seemingly contradictory pieces of theory and evidence linking social policy to economic growth. Essentially the paper looks at four areas of research: 1) the growth and inequality debate; 2) the small but growing literature on the policy determinants of economic growth; 3) an examination of two specific social policies — education and health; and 4) the literature on major technological change, wage inequality and the new economy. To provide the context for this discussion, the paper also includes some background material on economic growth, productivity, and social policy in OECD countries.

By way of a caveat, the paper is focused specifically on issues that are pertinent to Canada, or at least to countries like Canada — a democratic, high-income, small, open OECD economy. Nothing in what follows is meant to prescribe what development strategies are, or are not, appropriate for the developing world. The paper does not discuss the other main objectives of social policy that are not directly related to growth. Lastly, the paper does not discuss two areas of social policy that do have growth effects but are not directly related to the productivity issue. These are: a) the consequence of social security reform on savings — a very active debate driven by the aging population issue; and b) the effects of labour market regulation on employment, which have been extensively discussed since the release of the OECD Jobs Study.5

My main conclusion is in the form of a non-conclusion. This is one case where strong policy conclusions are well ahead of both theory and evidence. Neither provides conclusive support for the proposition that either a) policies directed at reducing inequality will increase productivity growth or b) increased social spending will raise productivity growth. Both advocates and opponents of such policies will find little comfort in these conclusions. Advocates, for the obvious reason that they are left in the position of dealing with the charge that equity and efficiency are often conflicting goals. Opponents, because the evidence is often sufficiently indecisive to leave ample room for a priori reasoned arguments to the contrary. Lastly, it is important to stress that most of the research is relatively recent. It is entirely possible that the balance of evidence may shift one way or the other as new studies are published.

Date modified: