Elsevier

Journal of Health Economics

Volume 43, September 2015, Pages 244-268
Journal of Health Economics

Review Article
An economy of scales: A selective review of obesity's economic causes, consequences, and solutions

https://doi.org/10.1016/j.jhealeco.2015.03.001Get rights and content

Abstract

This paper reviews the economic research on obesity, covering topics such as the measurement of, and trends in, obesity, the economic causes of obesity (e.g. the monetary price and time cost of food, food assistance programs, income, education, macroeconomic conditions, and peer effects), and the economic consequences of obesity (e.g. lower wages, a lower probability of employment, and higher medical care costs). It also examines the extent to which obesity imposes negative externalities, and economic interventions that could potentially internalize such externalities, such as food taxes, subsidies for school-based physical activity programs, and financial rewards for weight loss. It discusses other economic rationales for government intervention with respect to obesity, such as imperfect information, time inconsistent preferences, and irrational behavior. It concludes by proposing a research agenda for the field.

Overall, the evidence suggests that there is no single dominant economic cause of obesity; a wide variety of factors may contribute a modest amount to the risk. There is consistent evidence regarding the economic consequences of obesity, which are lower wages and higher medical care costs that impose negative externalities through health insurance. Studies of economic approaches to preventing obesity, such as menu labeling, taxes on energy-dense foods, and financial rewards for weight loss find only modest effects on weight and thus a range of policies may be necessary to have a substantial effect on the prevalence of obesity.

Introduction

The prevalence of obesity has risen dramatically in the United States and most other economically developed countries in the past several decades (National Center for Health Statistics, 2014, OECD, 2014). Concern about the health risks, medical care costs, and other consequences of obesity has led to a host of policies and programs implemented by companies, school districts, and governments to prevent and reduce obesity. It has also resulted in a large body of economic research on the causes and consequences of obesity and on approaches to prevent and reduce obesity. The purpose of this article is to organize and synthesize what is known from this literature and to identify the most pressing needs and promising opportunities for future research.

The Centers for Disease Control and Prevention and the World Health Organization define obesity in adults as a body mass index or BMI (which is calculated as weight in kilograms divided by height in meters squared) of 30 or higher, and define obesity in youth as a weight-for-height that exceeds the historic 95th percentile (Institute of Medicine, 2012, U.S. DHHS, 2010, WHO, 2000) (see Table 1 for the definitions of each clinical weight classification). BMI does not measure fatness, merely weight for height, so individuals with above-average lean mass (e.g. muscular individuals) may be incorrectly classified as obese (Burkhauser and Cawley, 2008). More accurate measures of obesity are based on fat mass rather than total body mass, and these measures are in some cases more strongly correlated with outcomes than BMI.1 Although a few data sets, such as the National Health and Nutrition Examination Survey (NHANES) contain measurements of body composition (such as fat mass and fat free mass), most social science datasets continue to include only weight and height. As a result, most of the research discussed in this review defines obesity using BMI.

Trends in the prevalence of obesity in the U.S. have been tracked for more than 50 years using data on BMI calculated using measured weight and height in the series of NHANES surveys. Americans began gaining weight in the 1920s but this did not result in large increases in obesity until decades later: the 1960s or 1970s if one defines obesity using percent body fat or the 1980s if one defines obesity using BMI (Komlos and Brabec, 2010, Burkhauser et al., 2009). Specifically, the prevalence of obesity (defined using BMI) among U.S. adults rose slowly from 13.3% in 1960–1962 to 15.1% in 1976–1980. From that point on, it rose dramatically, from 15.1% in 1976–1980 to 23.3% in 1988–1994 to 35.3% in 2007–2010 (National Center for Health Statistics, 2014); see Fig. 1.2 The prevalence of obesity in youths follows a similar pattern; see Fig. 2. The increase in obesity is particularly striking because it happened at the same time as many healthier trends such as decreased smoking and reduced consumption of alcohol, as well as decreased drug use among youths (e.g. Stewart and Cutler, 2014, Institute of Medicine, 2012, Centers for Disease Control and Prevention, 2010).

Obesity also rose in many other countries of the world during the same period, but the prevalence in the U.S. remains one of the highest. Among the large economically developed countries that are members of The Organisation for Economic Co-operation and Development (OECD), the U.S. has the highest prevalence of adult obesity (35.3%), followed by Mexico (32.4%), New Zealand (31.3%), Hungary (28.5%), Australia (28.3%), Canada (25.4%), Chile (25.1%), the United Kingdom (24.7%) and Ireland (23.0%); see OECD (2014).3 East Asian countries such as Japan and Korea have the lowest prevalence (3.6% and 4.6%, respectively); see OECD (2014). The dozen or so countries with higher rates of obesity than the U.S. include small Pacific island nations like Nauru, Tonga, and Samoa, where the prevalence exceeds 50% (World Health Organization, 2011). The Institute of Medicine and the World Health Organization describe current rates of obesity as epidemic (Institute of Medicine, 2012, World Health Organization, 2011).

The daily calorie surplus that explains the increase in weights in the U.S. is relatively small: 220 calories per day for adults (Hall et al., 2011) and 41 calories per day for youth (Wang et al., 2012).4 Unfortunately, there are very little direct data on calories consumed and calories burned over time, and what does exist is measured with substantial error. For example, one source of data on calories consumed is from 24-h dietary recall; these were collected as part of the Continuing Survey of Food Intakes by Individuals (CSFII) and continue to be collected as part of the National Health and Nutrition Examination Surveys (NHANES). However, these reports are plagued by significant reporting error. Archer et al. (2013) estimate that food intake in the NHANES is underreported by 281 calories per day for men and 365 calories per day for women (i.e. greater than the calorie surplus that explains the rise in weight over time in the U.S.), and describe the food intake data for a majority of NHANES respondents as “not physiologically plausible” (p. 8), i.e. incompatible with survival. Changes in measurement protocols and potential changes in reporting error over time further complicate efforts to measure trends in energy intake (Archer et al., 2013). Data on calories expended are limited to occasional time use surveys that record only activities, not actual energy expenditure.

Data on weight and obesity have their own challenges and limitations. Most datasets used by economists contain only self-reports (not measurements) of weight and height. The NHANES data, which contain both self-reports and measurements of weight and height, indicate that the reporting error in weight is non-classical; specifically, individuals tend to underreport their weight, with heavier individuals underreporting to a greater extent (see, e.g. Cawley et al., 2015).5 Such non-classical reporting error can bias regression coefficients in both models in which weight is the dependent variable and models in which weight is a regressor, and it can be difficult to sign the bias (Bound et al., 2002).

One way to address reporting error in weight is to use the NHANES as validation data (Cawley, 2004, Cawley and Burkhauser, 2006, Bound et al., 2002). Specifically, one can regress measured weight on self-reported weight in the NHANES and then “transport” the NHANES coefficients to the dataset of interest and use them to correct self-reported weight for reporting error.6 This assumes “transportability” (that the nature of the reporting error is the same in both datasets) and does not completely eliminate reporting error (O’Neill and Sweetman, 2013; Cawley and Burkhauser, 2006). Estimating models of instrumental variables can also address the problem of reporting error in weight, but when the error is non-classical the correction will be incomplete (Bound et al., 2002; O’Neill and Sweetman, 2013).

The rise in obesity has serious implications for public health. Excess fat is harmful because fat cells are collectively an endocrine organ, releasing hormones such as resistin, which causes insulin resistance and diabetes, and leptin, which damages the cardiovascular system (Trayhurn and Beattie, 2001, Hu, 2008).7 The pancreas compensates for insulin resistance by releasing even more insulin, which in turn increases the risk of various forms of cancer (Calle and Kaaks, 2004). It is estimated that obesity is responsible for 17.3% of coronary heart disease, 61% of Type II diabetes, 24% of osteoarthritis, 20.8–35.4% of colorectal cancers, 26.9% of pancreatic cancer, 35.5% of gallbladder cancer, and 42.5% of kidney cancer (Eckel, 2003, Calle and Kaaks, 2004). The greatest health impact results from morbid obesity (BMI  35); those just over the threshold of obesity (30  BMI < 35), which is called class 1 obesity, do not tend to fare worse than those of healthy weight in terms of medical care costs (Cawley and Meyerhoefer, 2012) or mortality risk (Mehta and Chang, 2009, Flegal et al., 2013). The health impact of morbid obesity is substantial, however, resulting in a roughly one-third higher risk of mortality (Flegal et al., 2013).8 In light of these health consequences, a White House Task Force described obesity as a “national health crisis” for the U.S. (White House Task Force on Childhood Obesity, 2010).

Economists took note of the rise in obesity and its implications, and research on the subject increased dramatically. According to the EconLit database of journal articles, dissertations, and working papers, the number of annual economics publications with the keyword “obesity” rose from zero in the year 1995 to four in 2000, 48 in 2005, and 152 in 2010. The economics of obesity, on which there was virtually no research 20 years ago, has become such an active topic that it is challenging to stay abreast of the literature. Given the size of the literature, this review is by necessity selective; for example, it focuses on the studies with the most credible and interesting identification strategies and excludes from consideration unpublished working papers because their results may change.

The paper is structured as follows. Section 2 discusses evidence on the economic causes of obesity, such as the money price and time cost of food, food assistance programs, income, education, the macroeconomy, and peer effects. Section 3 discusses the economic consequences of obesity, such as wages, employment, and medical care costs. Section 4 discusses possible mechanisms by which obesity may impose negative externalities and decrease social welfare. Section 5 examines the research on possible mechanisms for internalizing such externalities, and Section 6 discusses the evidence for other economic rationales for government intervention, such as imperfect information and consumers having time inconsistent preferences or behaving irrationally. Section 7 offers an economic research agenda on obesity, noting the greatest needs and most promising opportunities for future research.

Section snippets

The economic causes of obesity

The first law of thermodynamics states that in a closed system energy cannot be created or destroyed; it can only be transformed. In the context of obesity, this implies that calories consumed must be either expended, excreted, or stored by the body as fat. Rearranging this energy balance equation indicates that the recent increase in obesity must be due to either an increase in calories consumed or a decrease in calories burned (see, e.g. Hill et al., 2012).9

The economic consequences of obesity

The economic consequences of obesity include worse labor market outcomes, such as lower wages and a lower probability of employment, and higher medical care costs. When studying the causes of obesity, there is often the potential to conduct randomized experiments: e.g. to randomize children into a preschool program in order to measure the impact of education on obesity or to randomize college roommates in order to test for peer effects in weight. That is more difficult when obesity is the

The negative externalities of obesity

Obesity imposes large external costs, which likely decrease social welfare. Third-party payers cover most of the costs of obesity related illness; e.g. in 2005, 88.2% of the total medical care costs of obesity in the U.S. were paid by health insurance companies, Medicare, and Medicaid (Cawley and Meyerhoefer, 2012). Ultimately, many of these costs of obesity were paid by non-obese individuals. In group health insurance, the premium paid by each individual does not depend upon his or her weight;

Policies to internalize the external costs of obesity

There are several possible approaches to internalizing the external costs of obesity. The most direct would be to tax excess body fat by an amount equal to its marginal external cost. No country has yet implemented such a tax, perhaps because of concerns about discriminating against those with a genetic predisposition to fatness. However, Singapore implemented a program called “Trim and Fit” from 1992 to 2008 in which schoolchildren's BMIs were measured and youths deemed overweight were

Policies to correct other market failures

Internalizing external costs is not the only economic rationale for government intervention. Economists also recognize a role for government policies to address the market failure of imperfect information, help consumers who have time-inconsistent preferences, or protect those who act irrationally.

The economics of obesity: a research agenda

This section proposes a research agenda for the economics of obesity.

Conclusion

The large literature on the economics of obesity indicates that numerous economic variables, such as food prices, the time cost of food acquisition, income, education, the macroeconomy, and peer behaviors, may have modest effects on weight and obesity. The magnitude, and sometimes the sign, of these effects vary by subgroup (e.g. gender, race, age, and income). There appears to be no single dominant economic cause of obesity; instead, a wide variety of factors may each contribute a small

References (237)

  • J. Cawley et al.

    The impact of physical education on obesity among elementary school children

    Journal of Health Economics

    (2013)
  • J. Cawley et al.

    Lighting up and slimming down: the effects of body weight and cigarette prices on adolescent smoking initiation

    Journal of Health Economics

    (2004)
  • J. Cawley et al.

    The medical care costs of obesity: an instrumental variables approach

    Journal of Health Economics

    (2012)
  • J. Cawley et al.

    A case study of a workplace wellness program that offers financial incentives for weight loss

    Journal of Health Economics

    (2013)
  • K. Charles et al.

    Labor market fluctuations and health: is there a connection and for whom?

    Journal of Health Economics

    (2008)
  • S. Chou et al.

    An economic analysis of adult obesity: results from the behavioral risk factor surveillance system

    Journal of Health Economics

    (2004)
  • T. Christian et al.

    Trends in US food prices, 1950–2007

    Economics and Human Biology

    (2009)
  • C. Cotti et al.

    Fast food prices, obesity, and the minimum wage

    Economics & Human Biology

    (2013)
  • C. Courtemanche et al.

    Supersizing supercenters? The impact of Walmart Supercenters on body mass index and obesity

    Journal of Urban Economics

    (2011)
  • T. Coutinho et al.

    Combining body mass index with measures of central obesity in the assessment of mortality in subjects with coronary disease role of normal weight central obesity

    Journal of the American College of Cardiology

    (2013)
  • D.M. Cutler et al.

    Understanding differences in health behaviors by education

    Journal of Health Economics

    (2010)
  • D.M. Dave et al.

    How does the business cycle affect eating habits?

    Social Science & Medicine

    (2012)
  • P. Deb et al.

    The effect of job loss on overweight and drinking

    Journal of Health Economics

    (2011)
  • T. De Vriendt et al.

    Chronic stress and obesity in adolescents: scientific evidence and methodological issues for epidemiological research

    Nutrition, Metabolism, and Cardiovascular Diseases

    (2009)
  • L.H. Epstein et al.

    Experimental research on the relation between food price changes and food-purchasing patterns: a targeted review

    American Journal of Clinical Nutrition

    (2012)
  • L.C.H. Fernald et al.

    Cash component of conditional cash transfer program is associated with higher body mass index and blood pressure in adults

    Journal of Nutrition

    (2008)
  • L.C.H. Fernald et al.

    Role of cash in conditional cash transfer programmes for child health, growth, and development: an analysis of Mexico's opportunities

    Lancet

    (2008)
  • E.A. Finkelstein et al.

    Mandatory menu labeling in one fast-food chain in King County, Washington

    American Journal of Preventive Medicine

    (2011)
  • E.A. Finkelstein et al.

    Implications of a sugar-sweetened beverage (SSB) tax when substitutions to non-beverage items are considered

    Journal of Health Economics

    (2013)
  • G. Ainslie

    Specious reward: a behavioral theory of impulsiveness and impulse control

    Psychological Bulletin

    (1975)
  • R. Akee et al.

    Young adult obesity and household income: effects of unconditional cash transfers

    American Economic Journal: Applied Economics

    (2013)
  • M. Anderson et al.

    Are restaurants really supersizing America?

    American Economic Journal: Applied Economics

    (2011)
  • E. Archer et al.

    Validity of U.S. nutritional surveillance: national health and nutrition examination survey caloric energy intake data, 1971–2010

    PLoS ONE

    (2013)
  • M. Ashwell et al.

    Six reasons why the waist-to-height ratio is a rapid and effective global indicator for health risks of obesity and how its use could simplify the international public health message on obesity

    International Journal of Food Sciences and Nutrition

    (2005)
  • S.L. Averett

    Labor market consequences: employment, wages, disability and absenteeism

  • K. Baicker et al.

    The Oregon experiment – effects of Medicaid on clinical outcomes

    New England Journal of Medicine

    (2013)
  • G. Becker et al.

    A theory of rational addiction

    Journal of Political Economy

    (1988)
  • J. Bhattacharya et al.

    Does health insurance make you fat?

  • J. Bhattacharya et al.

    Breakfast of champions? The School Breakfast Program and the nutrition of children and families

    Journal of Human Resources

    (2006)
  • J. Bhattacharya et al.

    Health insurance and the obesity externality

    Advances in Health Economics and Health Services Research

    (2007)
  • J. Bhattacharya et al.

    Who pays for obesity?

    Journal of Economic Perspectives

    (2011)
  • G. Bishop et al.

    Personal responsibility and physician responsibility – West Virginia's Medicaid Plan

    New England Journal of Medicine

    (2006)
  • S. Bleich et al.

    Why is the developed world obese?

    Annual Review of Public Health

    (2008)
  • S.N. Bleich et al.

    Reducing sugar-sweetened beverage consumption by providing caloric information: how black adolescents alter their purchases and whether the effects persist

    American Journal of Public Health

    (2014)
  • K.J. Blumenthal et al.

    Medicaid incentive programs to encourage healthy behavior show mixed results to date and should be studied and improved

    Health Affairs

    (2013)
  • B. Bollinger et al.

    Calorie posting in chain restaurants

    American Economic Journal: Economic Policy

    (2011)
  • J. Bound et al.

    Measurement error in survey data

  • K.D. Brownell et al.

    The public health and economic benefits of taxing sugar-sweetened beverages

    New England Journal of Medicine

    (2009)
  • K.D. Brownell et al.

    Ounces of prevention – the public policy case for taxes on sugared beverages

    New England Journal of Medicine

    (2009)
  • G. Brunello et al.

    The causal effect of education on body mass: evidence from Europe

    Journal of Labor Economics

    (2013)
  • Cited by (0)

    I gratefully acknowledge financial support from an Investigator Award in Health Policy Research from the Robert Wood Johnson Foundation, and support from the Cornell Institute for Health Economics, Health Behaviors, and Disparities. Barton Willage provided expert research assistance. For their helpful comments and suggestions, I thank Adriana Lleras-Muney, David Cutler, and the other editors of the Journal of Health Economics, David Frisvold, Chad Meyerhoefer, and Nathan Tefft.

    View full text