Given that the Great Recession of 2008, despite its official termination in 2009, is clearly continuing on its gloomy way around the globe, it can only be cheering to learn that economic downturns are good for public health. This idea is so counter-intuitive that it may be hard to take seriously. (José Tapia’s Ph.D. dissertation adviser, when shown the results of years of research demonstrating this effect, demanded he redo the calculations because it seemed impossible.) Yet this is what the statisticians call a robust, even if little-known, result of statistical investigation of economic and health data for many countries. Since even official economic gurus like Lawrence Summers and Stanley Fischer are now predicting a long-term period of “secular stagnation” (something readers of the Rail could have learned from my articles on the recession in 2008 – 09) it’s good to know that there are immediate benefits to be expected from this, in addition to the possibilities raised by growing disenchantment with the workings of the “free market.”

—Paul Mattick

data.euro.who.int/hfadb/) ">

Rail: You have done research on the effects of economic crises on health. You are an economist, right?

José Tapia: Well, I have a degree in economics, but I also have a background in public health and medicine.

Rail: Do you agree with the idea that the Great Recession has had major harmful effects on health?

Tapia: No, I do not, but let me explain. What has been baptized the “Great Recession” is the economic downturn that started in late 2007, which became manifest with the global financial crisis of 2008 and affected in greater or lesser extent almost all national economies of the world. This, which has been the most important economic crisis since the 1930s, has had a relatively minor impact on the so-called emerging economies of Latin America and Asia, or on Australia and the countries of the former U.S.S.R. It was, however, a major economic downturn in the U.S., though it is in Europe that its impact has been the greatest. The semiofficial chronology of the National Bureau of Economic Research dated the end of the recession in the U.S. in mid-2009, when the unemployment rate reached about 10 percent. Since then, the unemployment rate has slowly declined. However, in many European countries economic growth lingered at negative levels in 2009 and 2010 and the rate of unemployment has remained at two-digit levels and even above 20 percent in a number of countries. The recession was very severe in Estonia, Greece, Iceland, Ireland, Latvia, Lithuania, Slovenia, and Spain, where the unemployment rate increased by several percentage points between 2007 and 2010.

Your question is probably motivated by the fact that in newspaper articles it has been claimed that, as a consequence of the recession and austerity policies, suicides have soared and population health has deteriorated, particularly in countries such as Greece and Spain, where the effect of the crisis has been compounded with the wrong economic policies. Some authors have claimed that the health consequences of crises depend on the policies applied by the government, so that when the government decides to expend more on health and other social services to fight the crisis the recession is, first, rapidly overcome and, second, has no bad consequences for health. In a recent paper in the British medical journal the Lancet this claim has been repeated in relation to Greece, where supposedly health has been quickly deteriorating because of austerity, while in Iceland and Finland, which have avoided austerity policies, the harmful consequences for health of the economic crisis have supposedly been avoided. Authors who maintain these views claim that health has deteriorated in the countries hit first by the Great Recession and then by austerity policies, in a similar way as it deteriorated in the countries of the old Soviet Union during the economic crisis of the early 1990s, during the transition to a market economy.

All these claims look plausible; some of them are true, others are plainly wrong. But to decide which are right and which are wrong, the proper data have to be examined.

The first question is whether health is damaged by economic crises. Answering this question involves deciding how to define and measure “health,” as well as defining what an economic crisis is.

Rail: So, what is “health” and how can we measure it?

Tapia: The World Health Organization defines health as “a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity.” But to measure such a broadly defined concept, researchers have focused on something much more concrete: deaths. Since deciding to what extent “disease or infirmity” is present in a society at a given time is a very difficult thing, the best way to measure health is to measure how frequent death is. The basic idea is that the healthier a society is, the less frequent death is. Of course, aging is a basic cause increasing the risk of death, and thus to compare the health of two populations one of which is younger (say North Dakota compared to Florida) it is required to compensate, “to adjust for,” the effect of age. Thus demographers and epidemiologists compute age-adjusted mortality rates to evaluate how the health of a population evolves along time (the age-adjusting compensates for the fact that, in general, populations are becoming older in recent decades), or to compare different populations at a given point in time. The figure on this page presents the evolution of health—as measured by age-adjusted total mortality—in seven European countries between 1980 and 2011.

Rail: What about economic crisis?

Tapia: A crisis is any event that is, or is expected to lead to, an unstable and dangerous situation affecting an individual, community, or society. Loosely, the term means a “testing time” or an “emergency event.” Thus an economic crisis would be a situation in which the economy is going through an unstable situation or an emergency. Since “the economy” refers to the functional aspects of the social system related to the production and distribution of “wealth”—goods and services used to cover the needs and wants of the society and mostly of those who have power— it is proper to consider as economic crisis any situation in which the normal production and distribution of goods and services is disrupted.

Rail: So it seems logical that when economic activities are more or less disrupted or disorganized in a crisis the health of the people deteriorates.

Tapia: Well, not necessarily. In pre-capitalist societies, where most of the economy was linked with food production and agriculture, and most people were undernourished, economic crises were most of the time caused by natural causes damaging harvests: climatic events such as droughts or flooding, extreme temperatures, hurricanes, pests. During the Middle Ages, and even as recently as the 19th century in agricultural economies, bad harvests due to extreme temperatures or droughts were followed by lack of food and epidemics that killed significant numbers. Even in the 20th century, famines have occurred—for environmental or political reasons—in which mortality rates have skyrocketed. The importance of some of these famines that killed millions—e.g. the one in the Western part of the U.S.S.R. in the early 1930s, that in Bengal during World War II, or the one in China during the so-called Great Leap Forward in 1959 – 1961—are hotly disputed by historians and demographers. But in general, with the passing of time methods to keep food and have reserves for the years of bad harvest have made it possible to disconnect weather events and harvest failures from the economy and from changes in mortality.

Of course, the causes of economic crises in modern market economies—a.k.a. capitalism—are also hotly disputed among economists and economic commentators. But very few today would dare to blame economic crisis on climate and sunspots, as Stanley Jevons famously did in the 1890s. What is generally agreed is that the so-called business cycle—also known as trade cycle, industrial cycle, or crisis cycle—is an observable fluctuation of the market economy between periods of several years of expansion, upturn, or prosperity, and more or less sudden periods of contraction, downturn, or slump—often named recessions, depressions, or crises. The recent Nobel prize-winner in economics Eugene Fama has stated, unashamed, that economists don’t know and have never known what causes recessions. Other economists would disagree: Keynesians and monetarists believe crises are the consequence of wrong active or passive government policies or so-called “exogenous shocks” (like increases in price of raw materials), while Marxist authors and some economists of the institutionalist school consider crises as consubstantial to the working of the capitalist economic system—and thus unavoidable and unamenable to policy.

Rail: It seems to me you are losing track. What about the relation between economic crises and health?

Tapia: Oh yes, sorry! Let me say first that during the 19th century it was a prevalent view that mortality, economic crises, and lack of food were phenomena occurring together. Thomas Malthus became famous by ascribing mortality crises to limits on the production of food in the face of excess population, which he basically attributed to the tendency of poor people to reproduce like rabbits. Of course he was criticized and his views were despised by many—for instance, by Friederich Engels. Malthus maintained that population is always pressing on the means of subsistence; that as soon as production increases, population increases even more; and that the inherent tendency of the population to multiply in excess of the available mean of subsistence is the root of all misery and vice. Engels denied all these ideas and claimed that it was misery and poverty caused by exploitation that caused disease and death. In The Condition of the Working Class in England, a remarkable book written in 1843, when he was just 23 years old, Engels suggested that commercial crises, in which great numbers became jobless and destitute, were often associated with epidemics. He mentioned how need and suffering prevailed among the unemployed during such crises, and how it was only because mutual help that deaths by starvation were prevented and most unemployed emerged from crises with their lives after dire privations. But for Engels every crisis caused indirectly, by disease, a multitude of victims. He cited medical reports that attributed typhus to the wretched condition of the poor and asserted that the insufficient satisfaction of vital needs promoted contagion and widespread epidemics, so that, for instance, a commercial crisis or a bad harvest had always occurred at the time of a typhus epidemic.

This view that commercial crises, recessions in modern parlance, were linked to epidemics and mortality outbreaks was common throughout the 19th century. Indeed, when statistics were available, as in Sweden in the early decades of the 19th century, they showed that years of good harvests were years of general prosperity, high marriage rates and birth rates, and low mortality rates, while bad harvests were associated with the opposite of all these. A key element in this link was the lack of effective technologies for storing food, so that a bad harvest was directly followed by food scarcity, famine, and epidemics.

But times were changing, and if any thinker of the 19th century was looking at the future rather than at the past, it was Karl Marx. Like Engels, he rejected Malthusian ideas that linked misery to overpopulation. At the time the writings of Louis-René Villermé, Johann Peter Frank, Rudolf Virchow, William Farr, and other authors had popularized the idea that living and working conditions were the cause of ailments and death. For Marx as for many of his learned contemporaries—except most economists—it was the social and economic system that caused misery and disease. Commerce—epidemics were often disseminated by merchant ships—and industries were generating malaise, and regulations to defend workers’ health and public health had to be imposed, usually against the will of merchants and manufacturers. In Capital, Marx emphasized that working conditions were often the cause of bad health. It is well enough known, he wrote, “how much economy of space, and therefore of buildings, crowds workers together in cramped conditions. A further factor is economy on means of ventilation.” These, together with long working hours, were responsible for a great increase in respiratory diseases and consequently increased mortality, as shown by statistics on higher death rates and excess of lung disease in districts with large indoor industry. It must be remembered these were the times when working days in factories extended for 12, 14 or even more hours.

Somewhat contradicting the prevalent view that economic crisis necessarily had a harmful impact on health, because of the lack of income and food scarcity, Marx cited medical opinions asserting that the interruption of work in Lancashire textile districts during the cotton crisis of the American Civil War had had advantages from the point of view of health. In spite of being subjected to a poor alimentation, their exclusion from the factories prevented workers from exposure to the harmful factory atmosphere and the damaging overexertion of work, so that their health actually improved. Mortality of children was decreasing because their mothers were now allowed to give them their own breasts, and because lack of money prevented the administration of “Godfrey’s cordial” (a “pediatric syrup” made with opiates) to infants to keep them quiet. Marx thus referred to the beneficial effects of breast feeding a century before the World Health Organization launched international campaigns to discourage the nursing bottle and promote mother’s milk—a remarkable feat indeed. But he was also suggesting the potential healthy effects of economic slumps decades before social scientists began to look at statistics that, astonishingly, show drops in general mortality coinciding with economic depressions.

Rail: Are you saying that statistics show drops of general mortality coinciding with economic depressions?

Tapia: Yes, indeed. To my knowledge, it was William Ogburn and Dorothy Thomas who discovered this, when in 1922 they examined the relationship between changes in death rates and economic ups and downs in the United States. They noticed that mortality increased in times of prosperity, with reductions in mortality coinciding with economic downturns. As this seemed illogical to them, they pursued the issue further. Dorothy Thomas investigated this in her Ph.D. dissertation, which was published in 1925 as Social Aspects of the Business Cycle. Reexamining the U.S. data and analyzing British data on mortality, she concluded that contrary to expectation, deaths, both in the United States and England, showed a strong tendency to increase with prosperity and diminish with depression. She agreed that it was difficult to find a satisfactory explanation for this, but insisted that the data showed clearly that high death rates were associated with periods of prosperity and low death rates with periods of depression.

Rail: So what happened to that discovery, now almost a century old?

Tapia: Dorothy Thomas died in 1977, after a long and productive life in academia, but her findings were basically ignored for half a century. After the commotions of the 1930s and the 1940s there was a quarter century of prosperity in which rapid economic growth was supposedly overcoming all social ailments. At that time, any suggestion that prosperity could have harmful consequences would have gone against the grain. This probably explains why the discovery of Ogburn and Thomas was forgotten.

Facts are stubborn, though. In 1977, a biologist at the University of Pennsylvania, Joseph Eyer, examined U.S. mortality rates between the 1870s and the 1970s. What he found and published in a paper titled “Prosperity as a Cause of Death” was what he baptized “Thomas’s Effect”: peaks in the unemployment rate coincide with troughs of mortality, and troughs of unemployment coincide with peaks of mortality. Since unemployment is at its lowest levels when the economy is booming and at its highest when the economy has been depressed for a while, the clear inference was what Thomas had found five decades before: mortality tends to decrease in recessions and to increase in expansions.

data.euro.who.int/hfadb/">

However, when facts do not fit with general views about society, someone will soon appear to “demonstrate” that they do not exist. Thus in the 1980s and the 1990s a sociologist named Harvey Brenner “demonstrated” that the increase in mortality observed during a period of prosperity is “actually” the consequence of lagged harmful effects of the earlier recession. Though his investigations were soon criticized because of his tortuous statistics, lack of information about the data, and the impossibility of replicating his results, the outcome of the controversy was that the surprising relation between expansions, recessions, and mortality was again forgotten for 20 years, during the 1980s and 1990s. That changed when in 2001 and later a number of papers were published in which the “Thomas Effect” was found again. 

Rail: I see in the figure that in Russia, Belarus, and Estonia there was a huge increase in mortality in the early 1990s. Was not that the moment of the crisis of the end of the U.S.S.R.?

Tapia: Yes, the breakdown of the U.S.S.R. and the sudden transition from the centrally planned, “communist” economy of the Soviet times to a market economy under the guidance of Jeffrey Sachs and other Western economic advisors during the Boris Yeltsin era was one of the greatest health catastrophes of the 20th century. Several million extra deaths have been attributed to that transition. But that was a special crisis which had little to do, in its nature and in terms of health consequences, with business-cycle recessions.

For instance, there were recessions in Greece in 1987 and 1993, and in Ireland in 1983. But mortality (see the former figure) has troughs in those years, which means it drops faster. A peak in mortality in Ireland in the late 1990s is also observable. This coincided with the period of accelerated economic growth when the term “Celtic Tiger” was invented for an Irish economy growing at rates close to 10 percent per year. So a lot suggests that it is booms, not recessions, which increase mortality rates. Indeed, if you look at the graph for the European countries since the Great Recession started in 2007, you will notice that mortality has been falling as fast or even faster than in previous years.

Rail: It is said that suicides are increasing very fast in the countries where austerity policies are being applied. Do you agree?

Tapia: Well, what is true is that suicides often increase during periods of economic recessions when unemployment rates increase. However, the assertion that suicides are becoming a major public health problem, for instance in Greece, and that this is a consequence of austerity policies, seems quite an exaggeration. Statistics show that suicides in Greece have actually increased very slightly since 2007 and they are at very low levels (see figure on this page) compared with, say, Finland or Iceland—where, by the way, suicides also increased in 2007 – 09. But in Iceland and Finland austerity policies have not been applied, so the evidence in favor of the idea that suicides are very sensitive to austerity policies does not seem very strong, to put it mildly. At any rate, a lot of evidence suggests that it would be a good idea to reinforce the prevention of suicide during recessions. Unfortunately, what usually happens is that governments at national, regional, and local levels cut social work and mental health services because of lack of financing.

In general, looking at population health in general, which involves examining total mortality as well as mortality for specific causes and ages, we see that these countries have done quite well after the recession started in 2007, though Greece did a little better. During the recession all-cause mortality declined in Greece as fast as in Finland (see the previous figure); even for cardiovascular disease mortality (the leading cause of death), or mortality at ages 75 and over, the Greek rates have declined faster than the Finnish rates, so that the gap in these mortality rates in 2011 was almost closed, although Finland had a clear advantage over Greece before 2007. In terms of infant mortality, Greece has done worse that Finland, but the difference is quite small.

Rail: What could be the reasons why there are more deaths when the economy is expanding faster, and less when there is a downturn? Is not true that the unemployed have higher rates of mortality?

Tapia: Yes, compared to employed workers, jobless workers have higher mortality rates because of cardiovascular disease, suicides, and other causes. This can be because there is causality both ways—that is, bad health increases the likelihood of unemployment, but also being unemployed increases the probability of disease and death. Using data from an epidemiological study of young adults I have found that unemployed individuals have extremely high levels of depression and do little physical exercise—though, interestingly, they drink less compared with employed individuals. All this can explain why during recessions and expansions increases and decreases in unemployment correlate with rises and drops in suicides.

However, even in a big economic crisis, say the Great Depression of 1930, the unemployed are a minority of the population. For instance an unemployment rate of 25 percent (a fourth of the economically active population) would mean that the unemployed may be between 10 percent and 15 percent of the population. During recessions both workers with jobs and the population at large are exposed to changes that depend on the condition of the economy, and many of these changes are beneficial for health. For instance, in general, overtime and working hours diminish in recessions and that allows for more sleep and more exercise. Tobacco consumption also declines. Less atmospheric pollution due to less industrial activity and transportation (both industrial and recreational driving decline) imply fewer cardiovascular and respiratory deaths and fewer lesions and deaths because of industrial or traffic injuries.

Rail: You have not said anything about U.S. Has the Great Recession been bad for health in the United States? Has it been good?

Tapia: Suicide rates have significantly increased, that is true. While prior to the recession the age-adjusted rate of suicide decreased 0.9 percent in 2005 and remained stable in 2006, it increased in 2007 by 3.7 percent, and respectively by 2.6 percent, 0.9 percent, and 3.4 percent in the next three years (C.D.C. data). However, suicides are a minor part of all annual deaths, about 1 percent in the U.S., and they have a very small influence on the evolution of total mortality, which has been falling quickly: in 2008 by 0.1 percent, in 2009 by 3.3 percent, and in 2010 by 0.4 percent. What is interesting is that during the last two decades (the 1990s and the 2000s) mortality decreased on average 1.1 percent per year, with only two years in which it increased, 1993 (when it grew by 2.3 percent) and 2005 (when it grew 0.2 percent). Both were years of strong economic growth.

Christopher Ruhm, who in 2001 published a paper titled “Are Recessions Good for Your Health?”—to which his basic answer was “Yes”—has recently published another paper, “Recessions, Healthy No More?” in which he asserts that total mortality has shifted over the 1976 – 2009 period from being strongly procyclical—that is, growing in expansions and falling in recessions—to being essentially unrelated to the condition of the economy. But he also acknowledges that the relationship shows instability over time and is likely to be poorly measured when using less than 15- or 20-year periods. Ruhm claims that deaths due to cardiovascular disease and transport accidents continue to be procyclical—although possibly less so than in the past—whereas some countercyclical patterns have emerged, the most interesting of them deaths caused by overdose of prescription drugs. For Ruhm, this has probably occurred because declines in mental health during recessions are increasingly associated with the use of prescribed or illicitly obtained medications that carry risks of fatal overdoses. I would add that if procyclical mortality is to a large extent linked to industry-related effects—more atmospheric pollution because of industrial emissions and transportation, more work-related stress—the decrease of manufacturing is a potential explanation for the decreasing link between the business cycle and health. This phenomenon—the weakening of the link between business cycles and mortality fluctuations—has also been observed in Japan and Britain.