The reversal of the relation between economic growth and health progress: Sweden in the 19th and 20th centuries

https://doi.org/10.1016/j.jhealeco.2007.09.006Get rights and content

Abstract

Health progress, as measured by the decline in mortality rates and the increase in life expectancy, is usually conceived as related to economic growth, especially in the long run. In this investigation it is shown that economic growth is positively associated with health progress in Sweden throughout the 19th century. However, the relation becomes weaker as time passes and is completely reversed in the second half of the 20th century, when economic growth negatively affects health progress. The effect of the economy on health occurs mostly at lag 0 in the 19th century and is lagged up to 2 years in the 20th century. No evidence is found for economic effects on mortality at greater lags. These findings are shown to be robustly consistent across a variety of statistical procedures, including linear regression, spectral analysis, cross-correlation, and lag regression models. Models using inflation and unemployment as economic indicators reveal similar results. Evidence for reverse effects of health progress on economic growth is weak, and unobservable in the second half of the 20th century.

Introduction

One of the unarguable proofs of social progress in recent centuries is the reduction of mortality rates and the associated increase in life expectancy. Thirty years ago, only 36 out of 178 countries had a life expectancy at birth of over 70 years, while there are now 87 nations exceeding this figure (UNDP, 2005). In a large number of countries today most humans are able to reach an advanced age, a privilege that was enjoyed worldwide by small minorities only one century ago (Riley, 2001).

The historical decline of mortality has been attributed to various factors associated with economic and social advancement, including the rising availability of material goods, urbanization and improvement of physical infrastructures and housing, increasing levels of education, improvement in personal and social hygienic behaviors, medical advances, the disappearance of slavery, and other significant reductions in discrimination for gender, religious, or ethnic reasons (Cutler et al., 2006, Fogel, 1994, Kunitz, 1986, Riley, 2001, Sen, 1990). It is generally agreed that the dramatic reduction of mortality due to infectious disease during the past two centuries has been the major determinant of the transition from high to low levels of death rates in every country. Consequently, among the potential causes explaining the drop in mortality, a prominent one is the improvement in nutrition, leading to a strengthening of the immune resistance to infection (Fogel, 1991, Fogel, 1994, Harris, 2004, Kunitz, 1986, McKeown, 1985, McKeown, 1988). Public policies—improving the hygienic quality of drinking water, milk, and various other foods; sanitizing the urban environment; cleaning up housing—have also been claimed as major factors for the dramatic reductions in mortality rates (Szreter, 1988, Szreter, 1999). Though there are extant controversies and lacunae in knowledge (Fogel, 1997), because the microbiological knowledge and the pharmacological or biological tools to fight infections (antibiotics, chemotherapeutic drugs, and vaccines) became available only many years after the accelerated drop in infectious disease mortality had started, it is usually accepted that progress in medical technology may have had a quite limited role in the historical decline of mortality (Fogel, 1997, Grundy, 2005, Kunitz, 1987, Preston, 1996). Views on the impact of medical technologies and medical care on mortality rates in recent decades cover a wide spectrum (Korda and Butler, 2006, McKinlay et al., 1989). At any rate, if the improvement in nutrition were the basic link in the chain leading to the secular decline in death rates; or if it were the public policies such as supply of clean drinking water, building of sewage networks, removal of garbage, and widespread vaccinations; or if several of these factors played important roles, what is undeniable is that the historical decline in mortality must be somewhat related to the process of economic development.

Assuming that particular aspects of the process of economic development were associated with the secular decline in mortality rates, it would be expected that the faster the process of economic development, providing health-improving goods, services, and infrastructures, the faster the progress in health as measured by declines in mortality rates. Indeed, research on preindustrial societies has shown links between harvest yields, grain prices, real wages, and changes in mortality. However, mortality responses to agricultural failures, grain price inflation, or changes in real wages become more muted as the level of development increases (Thomas, 1941, Lee, 1981, Bengtsson and Ohlsson, 1985, Galloway, 1988). Moreover, in the United States and Britain, some historical periods of rapid economic growth during the early years of industrialization have been shown to coincide with increasing mortality (Easterlin, 1999, Haines et al., 2003, Higgs, 1979, Szreter, 1998). In modern India and China during recent decades of strong economic growth, the declines in mortality rates have been small compared with large drops in mortality during the slow-growth decades before economic liberalization (Cutler et al., 2006).

A whole body of research, mostly published in medical journals, claiming short-term and long-term effects of periods of economic slowdown in rising mortality rates in 20th century industrial economies (Brenner, 1971, Brenner, 1977, Brenner, 1979, Brenner, 1981, Brenner, 1983, Bunn, 1979) has been discredited by later and more solid studies showing short-term oscillations of mortality fluctuating up in expansions and down in recessions, with the death rate sometimes even reversing its declining long-term trend during periods of accelerated economic growth (Chay and Greenstone, 2003, Dehejia and Lleras-Muney, 2004, Gerdtham and Ruhm, 2006, Graham et al., 1992, Laporte, 2004, Neumayer, 2004, Ruhm, 2000, Ruhm, 2003, Ruhm, 2005a, Ruhm, 2005b, Tapia Granados, 2005a, Tapia Granados, 2005b). Recent research seems therefore to suggest an inverse relation between the rate of improvement in health conditions and the rate of economic growth, at least in the short run and in advanced economies in recent decades. Moreover, in modern industrialized nations, it is not hunger but harmful caloric overconsumption and its pathologic effects—overweight, diabetes, hypertension, cardiovascular disease, and cancer—that are the scourge of the poor (Isaacs and Schroeder, 2005). A little-known historical natural experiment was the one occurring in the Nordic countries during World War II, when major drops in mortality due to cardiovascular disease and diabetes took place, apparently as a consequence of food shortages and cuts in the consumption of dairy products (Malmros, 1950). Something similar may have happened in the 1990s in Cuba (Franco et al., 2007).

Though it is increasingly accepted that in the short run economic growth may have harmful effects on health, in the long term a beneficial impact of economic growth on health improvement is usually accepted. For instance, in a recent commentary (Ruhm, 2005a), Christopher Ruhm—probably the author who has most forcefully shown in recent years the association between economic expansions and mortality increases—has stated,

Higher mortality during temporary expansions need not imply negative effects of permanent growth. The key distinction is that transitory increases in output usually require more intensive use of labour and health inputs with existing technologies, whereas lasting changes result from technological innovations or expansions in the capital stock that have the potential to ameliorate any costs to health. Individuals are also more likely to defer health investments during temporary than permanent increases in work hours and sustained growth permits the purchase of consumption goods (like safer cars) that benefit health.

Since it has been proved that the poorest countries also have the worst health indicators, and in these countries income growth has a direct translation into improved health conditions (Pritchett and Summers, 1996), it is tempting to apply the same reasoning to high- or medium-income countries, assuming for instance that the capacity to generate higher earnings “facilitates an increase in the consumption of health-related goods such as adequate food or medicine” and healthy changes in lifestyle (López-Casasnovas et al., 2005).

In the field of historical demography, however, expectations about the impact of economic growth in the long-run decline of mortality are generally modest. For instance, according to already classical estimates by Samuel Preston, only between 10% and 25% of the massive international declines in mortality between the 1930s and the 1960s could be attributed to improved standards of living measured in terms of income per capita (Preston, 1976, Preston, 1996), an estimate that has not been seriously challenged to date.

In each country, large increases in the output and availability of goods and services took place during the transition from an agrarian economy, mainly producing for self-consumption, to an industrial monetary economy in which markets and commerce play a much larger role. That is precisely the process through which Sweden passed during the past two centuries. The population involved in agriculture and subsidiary occupations was about 80% of all Swedes in 1800, still over 50% in 1900, but below 5% at the end of the 20th century. The share of “agriculture and ancillaries” in the Swedish gross domestic product (GDP) ranged between 35% and 40% from 1800 up to the 1870s, decreased to about 25% in the 1910s, and dropped to only 2.5% in 1990 (Krantz, 2002; Swedish Board of Agriculture, 2005; Thomas, 1941).

Because of early development of a statistical registration system, Sweden has historical statistics that are probably the best in the world. Using these statistics, it is possible to analyze the long-term relation between economic growth and health progress. The results of the analysis that will be presented herein provide substantial evidence that the relation between economic growth and health progress reversed in Sweden during the past two centuries, from strongly positive in the first half of the 19th century to moderately negative in the late 20th century.

The data and methods used in the study are explained in the next two sections. Section 4 presents the results of the statistical analysis, and Section 5 discusses the findings and presents the conclusions of the study. In Appendix A we present and discuss (i) potential pathways connecting harvests, inflation, economic growth and mortality; (ii) some issues related to the GDP estimates; (iii) Granger-causality tests.

Section snippets

Data

Economic growth for the years 1800–1998 was indexed by the annual rate of growth of GDP per capita.1 To gauge inflation, the annual rate of change of the GDP deflator was used as an index of the year-to-year change in the general level of prices. The general crop

Methods

We studied the coincidental or lagged covariations of “the economy” and health by using a variety of statistical models. Though there are many perspectives on causality in economics (Hicks, 1979, Darnell, 1993), not to mention social science (MacIver, 1973), epidemiology (Rothman and Lanes, 1988, Susser, 1973), statistics (Eells, 1991, Pearl, 2000), or philosophy (Mackie, 1974), a common view among empirical researchers is that statistical results alone cannot demonstrate causality, which is

Regression modeling of coincidental effects

The historical relation between economic growth and health progress can be modeled with an equation such asΔht=b0+b1t+b2gt+b3tgt+εtwhere the progress in health indexed as a longevity gain (or alternatively, as a relative decline in a mortality-based variable) is regressed on a constant, time t, economic growth gt, and the interaction of time and economic growth, t × gt. If economic growth has an impact on health progress, we will expect significant estimates of the parameter b2. Moreover, if

Discussion and conclusions

The consistency of results using different statistical models and different economic and health indicators makes it very unlikely that the described reversal of the relation between economic growth and health progress is a spurious finding. All the models are consistent with a weakening of the strong positive association between economic growth and health progress found in the first half of the 19th century, and with a reversal of the relation between economic growth and health progress, which

Acknowledgements

The excellent advice from Professor Christopher Jencks, as well as the guidance and suggestions from the editor and an anonymous referee are greatly appreciated.

References (84)

  • D.K. Backus et al.

    International evidence on the historical properties of business cycles

    American Economic Review

    (1992)
  • T. Bengtsson et al.

    Age-specific mortality and short-term changes in the standard of living: Sweden, 1751–1859

    European Journal of Population

    (1985)
  • I.T. Berend

    An Economic History of Twentieth Century Europe

    (2006)
  • M.H. Brenner

    Economic changes and heart disease mortality

    American Journal of Public Health

    (1971)
  • M.H. Brenner

    Health costs and benefits of economic policy

    International Journal of Health Services

    (1977)
  • M.H. Brenner

    Mortality and the national economy. a review, and the experience of England and Wales, 1936–76

    Lancet

    (1979)
  • M.H. Brenner

    Economic indicators as predictors of ill-health

    Lancet

    (1981)
  • M.H. Brenner

    Political economy and health

  • M.H. Brenner

    Commentary: economic growth is the basis of mortality rate decline in the 20th century—experience of the United States 1901–2000

    International Journal of Epidemiology

    (2005)
  • A.R. Bunn

    Ischaemic heart disease mortality and the business cycle in Australia

    American Journal of Public Health

    (1979)
  • G. Casella et al.

    Statistical Inference

    (1990)
  • K.Y. Chay et al.

    The impact of air pollution on infant mortality: evidence from geographic variation in pollution shocks induced by a recession

    Quarterly Journal of Economics

    (2003)
  • Cutler, D.M., Deaton, A., Lleras-Muney, A., 2006. The Determinants of Mortality—NBER Working Paper No. W11963....
  • A.C. Darnell

    Causality and Granger causality

  • R. Dehejia et al.

    Booms, busts, and babies’ health

    Quarterly Journal of Economics

    (2004)
  • P.J. Diggle

    Time Series: A Biostatistical Introduction

    (1989)
  • R.A. Easterlin

    How beneficent is the market? A look at the modern history of mortality

    European Review of Economic History

    (1999)
  • R.A. Easterlin

    The Reluctant Economist: Perspectives on Economics, Economic History, and Demography

    (2004)
  • E. Eells

    Probabilistic Causality

    (1991)
  • J. Eyer

    Review of [Brenner's] Mental illness and the economy

    International Journal of Health Services

    (1976)
  • J. Eyer

    Review of [Brenner's] Mental illness and economy—rejoinder

    International Journal of Health Services

    (1976)
  • J. Eyer

    Prosperity as a cause of death

    International Journal of Health Services

    (1977)
  • R.W. Fogel

    The conquest of mortality and hunger in Europe and America: timing and mechanisms

  • R.W. Fogel

    The relevance of Malthus for the study of mortality today: long-run influences on health, mortality, labour force participation and population growth

  • M. Franco et al.

    Impact of energy intake, physical activity and population-wide weight loss on cardiovascular disease and diabetes mortality in Cuba, 1980–2005

    American Journal of Epidemiology

    (2007)
  • P.R. Galloway

    Basic patterns in annual variations in fertility, nuptiality, mortality, and prices in pre-industrial Europe

    Population Studies

    (1988)
  • G.V. Glass et al.

    Design and Analysis of Time-Series Experiments

    (1975)
  • J.D. Graham et al.

    Poorer is riskier

    Risk Analysis

    (1992)
  • E. Grundy

    The McKeown debate: time for burial

    International Journal of Epidemiology

    (2005)
  • D.N. Gujarati

    Basic Econometrics

    (2003)
  • M.R. Haines et al.

    The short and the dead: nutrition, mortality, and the “Antebellum Puzzle” in the United States

    Journal of Economic History

    (2003)
  • B. Harris

    Public health, nutrition, and the decline of mortality: the McKeown thesis revisited

    Social History of Medicine

    (2004)
  • Cited by (107)

    • Epidemics in pre-industrial Europe: Impacts of climate change, economic well-being, and population

      2022, Anthropocene
      Citation Excerpt :

      In fact, the conclusions outlined above should closely connect with the social background. So far, studies showed that economic growth before industrialization was positively associated with health progress (Granados and Ionides, 2008), and good economic conditions and health are mutually reinforcing in the long run (Riley, 2001). In this regard, the linkage between economy and epidemics in pre-industrial era seems different from the modern epoch.

    • Public health in restructuring coastal communities: Generational trends in self-rated health following the decline in small-scale fishing. The HUNT study, Norway

      2021, Journal of Rural Studies
      Citation Excerpt :

      On the same note, compared to previous generations, the current working age coastal population probably faces different working conditions, and the current elderly coastal population probably holds different working experiences. Although economic growth does not guarantee a progress in health (Tapia Granados and Ionides, 2008), the importance of life chances and employment opportunities for health have been thoroughly documented (McDade et al., 2011; Harris et al., 2002), which gives reason to anticipate an enhancement of public health in the surviving and restructured fishing- and coastal-involved areas. Compared to other geographical areas that have not experienced an equally sizeable decline in a local industry, one might anticipate differing trends in health over the same period, with a potentially eventual narrowing in favour of coastal areas as labour markets likely have become more similar across regions over time.

    View all citing articles on Scopus
    View full text