Obesity: Causes, Consequences and Public Policy Solutions

94Obesity: Causes,Consequences and PublicPolicy Solutions

Obesity:Causes, Consequences and Public Policy SolutionsBy Robert Kaestner(Editor’s Note: Previous editions of The IllinoisReport have focused on the broad scope of thehealth care issue in Illinois. But within the challengesof providing a program that benefits all thosein the state who need assistance lie policy issuesspecific to areas of health that often do not figureinto the larger debate. This is one of those issues.)Obesity is one of the most pressing publichealth challenges facing Illinois andthe nation. Nationally, the prevalence ofobesity among children and adults has increasedsignificantly over the last 30 years. 1Among adults, obesity has increased nearly150 percent, from 14 percent of the populationin 1971-75 to 34 percent of the populationin 2005-2006. Even larger relativeincreases have been observed for children;the prevalence of obesity among childrenhas grown by 300 percent, from 5 percent in1971-1975 to 16 percent in 2003-2006. Andthe United States is not alone. Many developedcountries have witnessed an increasein obesity over the past 20 to 30 years.Rates of obesity in Illinois mirror those in thecountry, as shown in Figures 1 through 4.In 2007, approximately 27 percent ofwomen age 18 to 64 and 29 percent of similarlyaged men in Illinois and the U.S.were obese. 2 Also evident from Figures 3Figure 1Prevalence of Obesity Among Women in US and IllinoisPercent45%40%35%30%25%20%15%10%5%0%1995 1997 1999 2001 2003 2005 2007Females-USFemales-IllinoisFigure 2Prevalence of Obesity Among Men in US and IllinoisPercent45%40%35%30%25%20%1The medical community uses the Body MassIndex (BMI), defined as a person’s weight (inkilograms) divided by the square of the person’sheight (in meters), to classify individuals asobese whose BMI equals or exceeds 30 kg / m 2 .The BMI can also be calculated as [Weight inPounds/(Height in inches) 2 x 703.]2Figures differ from those cited above becausethey were calculated using different datasources, different ages, and are based on self-reportedweight and height. The trends in obesityreflected here are consistent with those reportedother data sources.15%10%5%0%1995 1997 1999 2001 2003 2005 2007Males-USMales-Illinois95

The Illinois Report 2009Currently, 25percent of theU.S. populationis completelysedentary and60 percentdoes notexercisesufficiently tomeetrecommendedlevels.98void. The greater availability of fast foodrelative to supermarket-purchased foodslowers the price of fast foods in these communities.Causes of Reduced Energy ExpenditureEvidence on changes in energy expenditure(e.g., exercise, walking, physical labor,active leisure activities) is less availablethan for energy intake, but what evidencethere is suggests that there has been a substantialdecline. The percentage of personsworking in physically demanding jobs remainedsteady between 1950 and 1970, butdecreased markedly over the next 30 years.Similarly, more people drive and fewerpeople walk. In 1960, only 67 percent oftravel to work was by automobile, but by2000 that figure had risen to 88 percent.Much of the change in commuting comesfrom change in where people live. Asurban sprawl has gradually increased, automobiletravel has increased. Finally,there has been a change in recreational activities.Television viewing has increasedby approximately 30 minutes per dayevery 10 years since 1950. While decreasesin energy expenditure alone cannot explainrising obesity, when it is combinedwith increasing calorie consumption theresults could be significant, as evidencedby the growing rates of obesity.Technological change is the likely cause ofthese changes in energy expenditure. Innovationhas made work, travel, and leisureactivity less strenuous. Most notably, technologicalchange has increased productivityin manufacturing and agriculture andsignificantly reduced the number of physicallydemanding jobs in the economy.Technology also has increased the valueand reduced the price of sedentary recreationalactivities such as watching televisionand playing computer games.Technology has increased the variety andsophistication of these leisure activities,while at the same time improvements inmanufacturing productivity has led to adrastic reduction in the cost of televisions,computers, and goods used with these activities(e.g., software, cable programming).In contrast, technological changemay lower the cost of a tennis racquet, butthe change in the price of engaging in tennishas probably increased more over timethan the price of television viewing becausethe time costs are the same in eachactivity, but the costs of complementarygoods have decreased more for television(e.g., expansion of cable TV) than tennis. Ingeneral, the costs of time-intensive activitieslike exercise have risen because of risingwages and increased participation ofwomen in the work force. Therefore, relativelytime-intensive recreational activitieshave declined.Another factor affecting energy expenditureis urban sprawl. The interstate highwaysystem and the rise in incomes havemade suburban living more affordable(lower travel costs) and desired (desire forlarger houses on cheaper land) and hascaused more persons to live in areas thatrequire more travel, leaving less time forexercise. A recent study reported that approximately12 percent of the increase inobesity between 1970 and 2000 can be explainedby suburbanization.Overall, physical activity levels are relativelylow and, while hard to measure, exercisehas likely declined due to changes inthe nature of work, travel, and leisure timeactivities. Currently, 25 percent of the U.S.population is completely sedentary and 60percent does not exercise sufficiently tomeet recommended levels.Public Policy SolutionsObesity is a growing and costly problemand many experts say rising costs of medicalcare justify government intervention.However, the obesity-related cost of healthcare is not necessarily the best argumentfor government intervention. First, somerecent studies have shown that obesity, as

Institute of Government & Public Affairscurrently measured by body mass indexabove 30, is not as strongly correlated withmorbidity and mortality as many have assumed.If correct, this fact would suggestthat many estimates of the health carecosts of obesity are exaggerated. Second,some argue that obese persons pay morefor health care – through lower wages onjobs that provide health insurance – thanless-obese persons. If so, then obese personsare paying their fair share, at least inthe employer-sponsored market that insuresmost non-elderly persons in theUnited States. Third, while the medicalcosts may be large there are other fiscalconsequences of obesity that are largely ignoredand that may offset the higher medicalspending associated with obesity. Forexample, if obesity does result in earlydeath, social security spending would besignificantly reduced. The upshot is thatmedical costs alone may not be sufficientto warrant government intervention.A more persuasive argument for governmentintervention is that government hasundoubtedly caused a significant part ofthe obesity problem through policies thataffect the price of food (e.g., farm subsidies),the built environment (e.g., highwayconstruction subsidies and zoning laws),and physical activity (e.g., school physicaleducation programs). Modifying or eliminatingthese myriad programs is not feasibleand may not be warranted by programeffects on obesity, and so direct action onobesity is justified.The problem that policymakers face is figuringout how best to intervene to reduceobesity. On this issue, there are many moreproposals than good policies. Also, there isa distinct naiveté that ignores personal responsibilityand the rational tradeoff thatpeople make between health and consumption.There are many examples ofpeople trading off consumption for health:working in risky jobs (e.g., law enforcement)for higher pay, buying home healthand safety equipment to prevent accidentalinjury or death, and paying for morescreening exams to detect illness. Obesitymay largely be a result of a similar tradeoff– eating more and consuming more sedentaryleisure activities, which are sources ofgreat satisfaction, for a less healthy andsomewhat shorter life.No force is compelling people to eat fastfood. People clearly enjoy McDonald’s anddecide freely to eat at McDonald’s restaurants.The fact that more people worklonger hours and family meals at home aremore infrequent, generally, does not implythat eating at McDonald’s is not a choice.Moreover, the fact that there are more fastfoodrestaurants today than 20 or 30 yearsago likely reflects the fact that home-preparedfood is more costly now than in thepast because of demands outside thehome. Given the voluntary nature of theactivity, two incomes almost certainly providegreater benefits than one. Therefore,limiting such consumption would substantiallyreduce consumer welfare and be metwith (appropriate) consumer resistance.Policies need to consider this basic fact.Consider another example. While governmentsubsidies to corn growers may havelowered the price of high-fructose cornsyrup and lowered the price of soft drinks,most of the decline in the price of softdrinks comes from technological change(in production and distribution). Therefore,most of the increase in soft-drink consumptionis not due to governmentintervention. Moreover, every adult knowsthat soft drinks are high in calories, butconsumption among adults and their childrencontinues unabated. These examplesillustrate the point that much of obesity isvoluntary, and government intervention toreduce obesity may face significant consumerresistance.TaxesGiven that the government has implementedprograms that have lowered the99

The Illinois Report 2009100price of energy intake and raised the priceof energy expenditure, it is reasonable forthe government to use tax policy to addressthese price distortions. Externalities associatedwith financing Medicaid andMedicare may also justify government intervention.However, taxing food is regressiveand hurts low-income consumers themost. Moreover, it is very difficult to designa targeted tax, say at “junk food,” thatwill be effective. Is Chinese take-out “fastfood” or “junk food”? Is pizza “junk food”?If we tax food purchases at some types ofestablishments more heavily, consumerswill switch to other types of establishmentsthat are less costly – for example, buyingprepared foods at supermarkets. Any tax ofthis type will be messy (arbitrary inclusionand exclusion of foods and establishments)and potentially ineffective, as the possibilitiesfor consumers to find substitutes thatare equally convenient and equally “unhealthy”are numerous. This conclusionstems from the basic fact that consumersare making optimal and voluntary choicesand are rationally trading off convenienceand weight gain for time.Another popular proposal is to tax softdrinks, or to tax soft drinks at rates higherthan other food items. Many states do thisnow to varying degrees. What effect willthis have on obesity? It will likely decreaseconsumption of soft drinks (perhaps evendiet soft drinks), but it will do so for bothobese and non-obese persons – the lattergroup will be clearly made worse off bysuch a policy. Note that only 30 to 35 percentof the adult population is obese and amuch larger fraction than this consumesoft drinks. Next, a soft-drink tax will increaseconsumption of other drinks includinghigh-calorie fruit juice and whole milk.A cup of orange juice has 112 calories; acup of whole milk has 160 calories; and acup of Coke – only 105 calories. Finally,soft drinks account for only 7 percent ofcalories consumed. Therefore, taxing softdrinks is unlikely to have any noticeableeffect on obesity.Taxes on food are focused on increasing thecost of energy intake, but taxes also can beused to lower the price of energy expenditure.However, designing such taxes maybe difficult, costly, and largely ineffective.For example, how could tax policy be usedto spur physical activity? The governmentcould raise taxes on recreational activitiesthat are sedentary, such as television viewing(televisions, cable subscriptions, etc.) orcomputer games. However, such taxeswould harm many consumers who usethese items and who are not obese, andharm consumers who use these products(e.g., computers) for other purposes (e.g.,school and work). And raising the price ofone class of products would create an incentivefor the market to develop substituteproducts that are not taxed. Here is yet anotherexample of the unintended consequencesof government intervention into amarket where transactions are largely voluntary.Consumers love television andcomputer games. Making these activitiesmore costly does nothing to change thisfact. Undoubtedly, there are many substitutetechnologies that could be developedthat would thwart the tax code and serveconsumer demand.Subsidies (tax cuts) for activities that increaseenergy expenditures may be moreeffective at changing behavior, but thesewill be very costly. For example, providinga tax break for gym membership will becostly because much of the break will go topersons who already purchase gym memberships.Few people will be encouragedto join gyms (and use them) by providinga tax credit because much of the cost of thegym is a time and psychological cost.In sum, tax policy has little likelihood ofbeing used to significantly reduce obesity.Public health advocates who call for taxeson soft drinks and fast foods ignore thevoluntary nature of obesity and exaggeratethe potential effectiveness of these taxes toreduce obesity. Taxes on food are also regressivehurting the poor the most.