Health as Investment: Economic Costs of Disease and Disability

This article, “Health as an Investment,” by Selma J. Mushkin, published in the Journal of Political Economy in 1962, serves as a foundational text in the emerging theory of human capital. It critically examines the economic contribution of improvements in the quality of people to overall economic growth. This central question arose as researchers observed that production in developed economies was increasing at a rate inexplicable by solely considering traditional inputs like physical capital and labor force additions. Beyond academic interest, the widespread attention reflects a global desire to demonstrate that efforts against disease and illiteracy are not only humanitarian but also significant drivers of economic growth. Preliminary findings suggesting a high rate of return on human investment have been readily embraced by countries seeking to integrate such investments into their economic development programs.

Mushkin emphasizes that health services improve people as productive agents, yielding a continuous economic return in the future. The paper meticulously draws comparisons between health and education, highlighting their roles as crucial forms of human investment.

Similarities between Health and Education as Investments

Mushkin identifies several key similarities:

• Improvement of Individuals: Both services become integral to an individual’s effectiveness in their work and daily life.
• Quantifiable Returns: The future increases in labor product that result from both health and education programs can be quantified to aid in planning and programming.
• Complementary Nature: Health and education are often joint investments made in the same individual, enhancing their effectiveness as both producers and consumers. Good health is essential for formal schooling, as demonstrated by the average of 8.4 days of schooling lost per year in the United States in 1958 due to ill health, which reduces the effectiveness of educational investment. Conversely, improved education enhances the return on life-saving health investments. Education is also responsible for training health personnel.
• Dual Function as Investment and Consumption: It is often difficult to separate the two aspects. Individuals seek health services to feel better (consumption) but also to perform more effectively as producers (investment).
• Essential for Welfare: Beyond satisfying basic wants, both are considered fundamental ingredients of human welfare and integral components of a standard of living. Without vitality, other forms of consumption lose their meaning, and without education, unique human qualities are diminished.
• Community Benefits: Returns from both health and education investments accrue not only to the individual but also to the broader community. For example, the prevention of contagious diseases like smallpox or poliomyelitis, or the treatment of tuberculosis, benefits the entire community by preventing disease spread and increasing overall economic productivity.
• Financing from Current Consumption Funds: Both are largely financed from current consumption, which Denison suggests makes a positive net contribution to economic growth. In the U.S. around 1960, expenditures for both health and education were approximately $25 billion. However, public outlays accounted for less than 25% of total health spending compared to about 80% for education.

Differences between Health and Education as Investments

Despite these similarities, Mushkin also highlights important differences that necessitate distinct approaches to measuring human capital:

1. Workforce Expansion vs. Quality Focus: Health programs directly increase the number of individuals in the working force by reducing deaths and disability, in addition to improving the quality of labor’s product. Education, on the other hand, primarily affects the quality of producers. The direct addition of workers through health programs provides a measurable unit of labor gained. This is particularly significant in non-industrial nations, where relatively small outlays (e.g., DDT spraying, BCG immunization, penicillin treatment for malaria, tuberculosis, syphilis, and yaws) can lead to rapid and substantial increases in life expectancy.
2. Lack of Standardized Quality Units: Unlike education, where “years of schooling” serves as a measure of educational stock, there is no comparable “quality unit” for health. Most common health status indexes are negative (e.g., death rates and morbidity rates), reflecting changes in numbers rather than quality. While some positive indicators exist (e.g., children being taller, earlier puberty, or physical fitness tests), their application for measuring work capacity is still limited.
3. Difficulty in Attributing Effects: It is challenging to isolate the effects of specific health programs from other influencing factors such as improved nutrition, better housing, enhanced working conditions, and higher incomes. Sickness is often both a cause and a consequence of poverty, making it difficult to fully disentangle the contributions of various factors to overall health improvements.
4. Absence of Earning Differentials Indexes: Unlike education, where income differentials based on years of schooling are readily available, there are no similar indexes for income differences associated with gradations in health. However, data from the National Health Survey shows an inverse relationship between family income and time lost from work due to illness, with lower-income individuals losing significantly more workdays. Negative measures like workmen’s compensation rates and “hazard pay” provide some market-based evaluations of the risk of sickness, death, and potential earning loss.
5. Fundamental Approach: Education is described as a developmental process that cultivates talent and builds on existing knowledge. Health programs, conversely, fundamentally aim to prevent a hostile environment from causing death and crippling, effectively striving to counteract natural biological selection. While historical societies practiced biological selection (survival of the fittest), modern economies reject this approach due to humanitarian principles and the prohibitive cost of foregoing the productive contributions of those who would die early. In modern economies, brainpower and other human capabilities are far more crucial than mere physical stamina.

Measuring Capital Formation and Labor Product through Health Care

Mushkin details how health care can be viewed as an investment, outlining two primary measurement methods:

• Cost Basis: This involves measuring the cost of environmental and curative health services embodied in the labor force over their lifespans. For example, producing a labor force member aged 18 in 1960 involved over $1,000 in health resources alone (at 1957-58 prices). For the 73 million people in the 1960 labor force, this represented a $73 billion stock of health care up to age 18, based on replacement costs. Total health-care expenditures in the U.S. rose from $1.091 billion in 1914 to $25.196 billion in 1958-59, increasing from 2.7% to 5.4% of the Gross National Product.
• Present Value of Added Labor Product: This method values the future earnings generated through health programs, accounting for depreciation through retirement and death. Some programs, like the complete eradication of malaria or typhoid from an area, can yield returns into perpetuity because the asset (health of the population) does not depreciate, benefiting succeeding generations.

The gain in human labor from preventing or curing sickness can be summarized in terms of the reduction in (1) deaths (loss of workers), (2) disability (loss of working time), and (3) debility (loss of productive capacity while at work). The calculation involves estimating the gain in productive work time and then assigning a monetary value to it. A key assumption is that without the disease, affected individuals would have been working, simplifying for a concept of resource gain even in situations of unemployment. The article acknowledges that ignoring the possibility of persons saved from one disease dying from another might overestimate gains from single disease eradication. A significant challenge lies in defining the full-time equivalent of the workforce, especially when considering the economic contribution of women working in the home, whose product is typically excluded from national accounts.

For assigning a money value to added work time, using earnings as a measure of the output attributable to labor is considered more appropriate than total product per worker, as earnings directly correspond to an individual’s contribution to production.

Health’s Impact on Economic Growth: U.S. Examples

Mushkin provides compelling historical data to illustrate health’s contribution to economic growth in the United States:

• Increased Workforce: The employed population in 1960 would have been over 13 million fewer if death rates had not declined since 1900. This means the labor force in 1960 was about 25% higher than it would have been if 1900 mortality rates had continued.
• Boost to National Income: This addition of 13 million survivors contributed over $60 billion to national income in 1960 when valued at average 1960 earnings.
• Contribution to Growth Rate: The decline in mortality rates over the sixty-year period (1900-1960) alone accounts for over 10% of the overall 3% economic growth rate in the U.S. economy.
• Specific Medical Advances: The use of antibiotics and chemotherapy between 1938 and 1952 is estimated to have saved 1.1 million lives (from pneumonia and influenza), increasing the 1952 labor force by almost 0.5% and national income by over $1 billion.
• Capital Value of Improved Health: The asset value in 1960 of the labor product added by workers who survived due to reduced mortality since 1900 is estimated at $820 billion. This implies that a capital stock of $820 billion would be needed to replace the equivalent product of these additional workers.

Mushkin acknowledges that these estimates might overstate the magnitude of capital formation solely from health care, as not all gains in life expectancy are attributable to medical advances, and productivity gains also stem from non-health factors like improved education and technology. Conversely, the estimates do not account for gains from reduced disability or debility, which would suggest an understatement.

Historical Context and Earlier Studies

The article also provides a valuable historical overview of research on the economic value of health, identifying three main concepts:

1. Developmental-Cost Concept: This approach measures the lost investment in rearing a child who dies prematurely, comparing it to the investment needed for them to contribute to production. Figures like Richard Contillion (1755) and Edwin Chadwick (1842) used this concept, with Chadwick notably establishing health programs as an integral part of economic policy. While some, like Singer, used this to argue that high death rates in underdeveloped countries lead to “unproductive investment” in children, others like Coale and Hoover contested this, suggesting mortality improvements might increase the proportion of dependents. Mushkin counters that health care improves worker vitality, and many diseases are “cripplers of mankind rather than killers”.
2. Capitalized Earnings Concept: Originating in insurance and actuarial theory, this defines the capital value of a human life as the present and discounted value of future earning power, reduced by the costs of upbringing and maintenance. Proponents include Sir William Petty (17th century), William Farr (19th century), and Irving Fisher (early 20th century). Fisher notably estimated the human assets of the U.S. population in 1907 at $250 billion, exceeding the value of all other wealth. He also estimated that eliminating preventable deaths could add another $1 billion to net future earnings. This approach faced disrepute partly due to its use in political arguments for national health insurance and objections to the “crass” valuation of human life. However, cost-benefit analyses have been effectively applied in occupational health, vocational rehabilitation, and health research. The resurgence of interest in this area is linked to public expenditure theory and cost-benefit analysis. Mushkin argues that consumer preferences alone are an inadequate guide for optimal health resource allocation due to factors like the external benefits of health services (e.g., vaccination benefits the community), the “indivisibilities” of certain services (e.g., pollution control), and instances where the market system doesn’t apply (e.g., care for the medically indigent). Weisbrod and others further refined these methods, assessing the comparative costs of diseases and arguing that society implicitly values human life even when it doesn’t contribute to production, making explicit valuation preferable.
3. Contribution to Annual Output/Economic Growth: This approach focuses on the change in annual income flow rather than human assets, illustrating the yield from health expenditures in terms of annual labor product added. Studies using this method include analyses of mental illness, peptic ulcers, and murine typhus control. The murine typhus study, for instance, showed that earnings gained through “saved” working time far exceeded the cost of disease eradication, and additional taxes paid were about five times the control expenditures.

The article concludes by emphasizing the critical need for objective measurement of the economic costs of disease and the development of a sound economic-statistical framework for estimating the toll of disease and disability to guide national health resource allocation. This complex task would greatly benefit from a team approach involving biostatisticians, physicians, and economists.

Reference: Mushkin, S. J. (1962). Health as an Investment. Journal of Political Economy, 70(5, Part 2), 129–157.