Is Economic Growth the Engine Driving Improved Life Expectancy?

Richard A. Easterlin, University of Southern California

Since the late 18th century the phenomenon of modern economic growth (MEG) has been sweeping the world. Formally, MEG may be defined as a rapid and sustained rise in real output per head and attendant shifts in the technological characteristics and economic structure of a nation. Since the late 19th century, there has been a doubling or more of world life expectancy at birth, from around 35 years to 70 years. The question addressed here is whether economic growth is the cause of this great improvement in life expectancy. The suggested answer is no - neither facts nor theory support this supposed association. Rather, economic growth and improved life expectancy are best understood as parallel developments, stemming respectively from major breakthroughs in production technology and health technology.

The following first summarizes some of the broad facts on the improvement in life expectancy - called here the "Mortality Revolution" - noting the association, or lack thereof, with MEG. It then turns to theoretical analysis of the relationship, and of the driving forces behind the Mortality Revolution.1

I.

On the factual side, two broad considerations can be cited in support of the view that MEG is responsible for the improvement in life expectancy. First, historical demographers find an early phase of mortality reduction in some of the leading countries in MEG, most notably England and France, from the late seventeenth to early 19th centuries, a period prior to that emphasized here. Second, the geographic pattern of diffusion of the improvement in life expectancy since the late 19th century - the real "Mortality Revolution" - is similar to that of MEG. Broadly speaking, the Mortality Revolution spreads from northwestern Europe and its overseas descendants to eastern and southern Europe and Japan, then to Latin America, followed by the Middle East, Asia, and northern Africa, and finally Sub-Saharan Africa.

Against these observations supporting the view that MEG is the cause of improved life expectancy, one may set the following. First, in the leading areas of MEG life expectancy ceased to improve in the central decades of the 19th century, despite continuing economic growth. (Indeed, in Great Britain the improvement in life expectancy up to the first part of the 19th century was little more than a return to the level prevailing in the Elizabethan period.) When life expectancy started once again to increase in northwestern Europe, in the latter part of the 19th century, the rate of improvement was much greater than earlier - about three times as rapid. This rapid improvement in life expectancy continued throughout the period from World War I to World War II, despite stagnant economic growth in many of the leaders in MEG.

Second, although the worldwide geographic pattern of diffusion of the Mortality Revolution is similar to that of MEG, the Mortality Revolution, despite its later start, spreads much more rapidly. Around the middle of the 19th century among the major regions of the world life expectancy at birth for both sexes combined fell in a band extending from the low twenties to the low forties. By 1990, the range extended from the high fifties to the high seventies. The exception is Sub-Saharan Africa, but even there, the last area in which the Mortality Revolution has taken place, life expectancy had broken out of the lower band and by 1990 was almost fifty years. Because of the rapid spread of the Mortality Revolution, worldwide differences in life expectancy have been narrowing since World War II. In contrast, international differences in MEG have widened since World War II. If the Mortality Revolution were simply an effect of MEG it would be hard to explain its later start, its more rapid spread, and the post World War II convergence in life expectancy differentials while income differentials have widened.

Finally, one may note a number of areas in the Third World where the Mortality Revolution has occurred under conditions of zero or negative economic growth. Before World War II this appears to have been true in Korea, British Guiana, Cuba, the Philippines, Sri Lanka, and Taiwan. In the post-World War II period, in a number of countries of Sub-Saharan Africa life expectancy has improved noticeably, despite declining living levels.

II.

In considering the theoretical determinants of life expectancy, it is helpful to follow the analytical scheme of demographer Samuel Preston, which parallels in a striking way Robert Solow's seminal analysis of the determinants of economic growth.2 Preston postulates a positively inclined function relating life expectancy (eo) as the dependent variable to MEG, the independent variable, under given conditions of health technology. He further postulates that this function shifts upward as health technology improves, because better knowledge, say, of how to control disease may reduce mortality and improve life expectancy, at a given level of economic growth. The substantive analytical issue thus becomes the role in the observed increase in life expectancy of (a) economic growth, a movement along a given function, versus (b) improved health technology, an upward shift of the function.

Consider now the likely slope of this function, that relating life expectancy to MEG, in the first part of the 19th century. The rise in real per capita income associated with MEG, by improving food and nutrition, clothing, and shelter, would increase resistance to disease and thus tend to raise life expectancy. This is the basis for the positive association between eo and MEG widely assumed by many economists and economic historians. But in the disease environment prevailing at the onset of MEG, another systematic feature of MEG, urbanization, affected life expectancy adversely. In the early 19th century urban mortality rates were much higher than rural. As a result, urbanization of the population increased exposure to disease and tended to reduce life expectancy. Schofield and Reher put it this way:

"[T]he rapid process of industrialization and urbanization in nineteenth-century European society created new obstacles to improved health. Towns had always been characterized by higher mortality rates due mainly to greater population densities which facilitated infection and filth; and during the nineteenth century increased proportions of the population were living in these urban centers. The poor living conditions of the age were probably one of the principal reasons why mortality ceased to improve during most of the central decades of the century."3

Thus, in the conditions prevailing in the first part of the 19th century, MEG had a twofold effect on life expectancy - one increasing resistance to disease and one increasing exposure. The net balance of these two effects is uncertain a priori. It seems likely, however, that the mid-19th century stagnation of life expectancy among the leaders in MEG, reflects the essentially offsetting effects of these two tendencies. The implication is that the slope of the relation at that time between eo and MEG was close to zero - that is, that MEG had little positive impact on life expectancy.

How, then, to explain the onset of rapid improvement in life expectancy toward the end of the 19th century? The answer is major advances in health technology associated with the interrelated development of public health measures and the breakthrough in medical knowledge resulting from validation of the germ theory of disease. In terms of Preston's analytical scheme this breakthrough had a two-fold effect on eo. First, and most important, it shifted the function relating eo to MEG upward, raising life expectancy at given levels of economic development. Second, it rotated the function in a positive direction, because the new technology was first introduced in urban areas. Thus, by eliminating the previous excess of urban over rural mortality, the advance in health technology enabled the positive effect on eo of higher living levels associated with MEG to prevail. It is the spread of this new health technology throughout the world that accounts for the distinctive patterns of life expectancy improvement - the much more rapid diffusion than MEG, the convergence in life expectancy differentials, and the occurrence of marked increases in life expectancy in the presence of zero or negative economic growth in both developed and developing countries.

III.

One may ask whether this analysis overlooks another possible link between MEG and life expectancy. The higher income accompanying MEG, it might be argued, was needed to finance increased private and government expenditure necessary to implement the improved health technology; because of this, MEG may, in fact, lie behind the advance in eo. Although this is a reasonable hypothesis, in fact, the funds needed to implement the new health technology appear to have been quite modest, especially since World War II. If they were not, then Third World countries would have been hard put to fund public health programs in the last four decades without substantial external aid. Although such aid existed, its quantitative significance was trivial - an assessment published by Samuel Preston in 1980 concluded that "total external health aid received by LDC's is less than 3% of their total health expenditures."4 Thus, despite low levels of economic development, poor countries have been able almost entirely on their own to finance the advances in health technology needed to raise life expectancy.

Was MEG, then, perhaps responsible for the advance in medical knowledge associated with improved health technology? The answer to this question leads one into the much broader issue of the determinants of the pattern of advancing knowledge over the past few centuries - from astronomy and mechanics in the sixteenth and seventeenth centuries to chemistry and electricity in the 18th and mid-19th centuries, followed by medicine and biology at the end of the 19th century. To the extent economists and economic historians have touched on this question, the tendency has been to invoke "the market" as the explanation -the pattern of advancing knowledge was dictated by the changing structure of human needs. As Nathan Rosenberg has cogently observed, however:

Many important categories of human wants have long gone either unsatisfied or very badly catered for in spite of a well-established demand. It is certainly true that the progress made in techniques of navigation in the sixteenth and seventeenth centuries owed much to the great demand for such techniques in those centuries, as many authors have pointed out. But it is also true that a great potential demand existed in the same period for improvements in the healing arts generally, but that no such improvements were forthcoming. The essential explanation is that the state of mathematics and astronomy afforded a useful and reliable knowledge base for navigational improvements, whereas medicine at that time had no such base. Progress in medicine had to await the development of the science of bacteriology in the second half of the 19th century. Although the field of medicine was one which attracted great interest, considerable sums of money, and large numbers of scientifically trained people, medical progress was very small until the great breakthroughs of Pasteur and Lister.5

If one accepts this view, the implication is that factors internal to the evolution of knowledge, rather than the market, account for the later breakthroughs in the medical sciences compared with those most relevant to production technology - factors such as the innate complexity of the problems posed in each field, the ease with which the human mind can grasp the underlying phenomena, the technological requirements of scientific research, and the internal logic of scientific inquiry itself.

IV.

To sum up, the present analysis suggests that both the advance in health technology underlying the Mortality Revolution and the advance in production technology underlying MEG reflect the pattern of advance in human knowledge. The Mortality Revolution occurs later than MEG because progress in medical knowledge comes later. The Mortality Revolution spreads more rapidly than MEG, and life expectancy differentials narrow while those in MEG do not, because the requirements needed to implement the new health technology are much more modest. The Mortality Revolution is thus best understood, not as an effect of MEG, but as a development with its own technology and pattern of diffusion, similar to but largely independent from the spread of MEG.

In the analysis of MEG, a distinction is commonly drawn ˆ la Gershenkron between leader and follower countries, and the possibility noted of higher growth rates in the followers due to the "catch-up" advantage that they enjoy. The same distinction applies to the Mortality Revolution. The leading areas in the Mortality Revolution are the same as those in MEG, and for a time they enjoy a growing advantage in life expectancy over other countries. Among the leaders, however, the rate of improvement is constrained by the rate of advance in health technology - indeed, in the mid-1960s some analysts argued that further improvement in life expectancy in the leading countries was unlikely, because degenerative diseases had replaced infectious diseases as the principal cause of death. In recent decades, however, the rate of improvement in the leading countries has been about the same as previously in the 20th century, reflecting the shift in focus of medical research in these countries to degenerative diseases. In the follower countries, the rate of improvement in life expectancy since World War II has been greater than that that had prevailed in the leaders, reflecting the catch-up advantage of the followers. As a result life expectancy differences worldwide have been narrowing. In this respect, the worldwide pattern of spread of the Mortality Revolution may be foreshadowing the prospective pattern with regard to MEG.

Endnotes

1. For more detailed evidence underlying the present analysis, see Richard A. Easterlin, Growth Triumphant: The Twenty-First Century in Historical Perspective, Ann Arbor, MI: University of Michigan Press, forthcoming 1996.

2. See Samuel H. Preston, "The Changing Relation between Mortality and Level of Economic Development," Population Studies, 29:2, 1975, 231-248; Robert Solow, "Technical Change and the Aggregate Production Function," Review of Economics and Statistics, 39, 1957, 312-320.

3. R. Schofield and D. Reher, "The Decline of Mortality in Europe," in R. Schofield, D. Reher, and A. Bideau, The Decline of Mortality in Europe, Oxford: Clarendon Press, 1991, p. 14.

4. Samuel H. Preston, "Causes and Consequences of Mortality Declines in Less Developed Countries in the Twentieth Century," in Richard A. Easterlin, ed., Population and Economic Change in Developing Countries, Chicago: University of Chicago Press, 1980, p. 315.

5. Nathan Rosenberg, Perspectives on Technology, New York: M.E. Sharpe, 1976, pp. 267-268.