• Heritage Action
  • Heritage Libertad
  • More

Expanding Federal Food Programs: Means-Tested Aid for Families with Children


My name is Robert Rector. I am a Senior Research Fellow at The Heritage Foundation. The views I express in this testimony are my own, and should not be construed as representing any official position of The Heritage Foundation.

This hearing is to examine proposals to expand spending on school nutrition programs. However, it is misleading to examine spending in one or two government programs in isolation. Most families receiving subsidized school meals also receive benefits from many other programs. Proposals to expand spending in a single program must be examined holistically, in the context of overall growth of government spending.

It is therefore important to consider school nutrition spending in the context of overall means-tested assistance to low-income families with children. In FY 2011, such means-tested aid will reach around $475 billion, or roughly $33,000 for each family with children in the lowest income third of population.

At the same time, the federal budget deficit in FY2011 will be $1.2 trillion, or 8.3 percent of the gross domestic product. As the national debt rises rapidly toward 100 percent of GDP, it is clear that the current growth of government spending is unsustainable. In that context, calls for long-term increases in spending on school meal programs are irresponsible.

Understanding the Means-Tested Welfare System

Since the beginning of the War on Poverty, government has spent vast sums on welfare or aid to the poor; however, the aggregate cost of this assistance is largely unknown because the spending is fragmented into over 70 separate programs. (See the table at the end of this testimony for a list of these programs.)

Even before the present recession, means-tested welfare or aid to poor and low-income persons was the third most expensive government function. Its cost ranked below support for the elderly through Social Security and Medicare and below government expenditures on education, but above spending on national defense. Prior to the current recession, one dollar in seven in total federal, state, and local government spending went to means-tested welfare.

Means-tested welfare spending or aid to the poor consists of government programs that provide assistance deliberately and exclusively to poor and lower-income people. By contrast, non-welfare programs provide benefits and services for the general population. For example, food stamps, public housing, Medicaid, and Temporary Assistance for Needy Families (TANF), the Women Infants and Children food program (WIC), the Child and Adult Care Food Program (CACFP) and the Summer Food Program are means-tested aid programs that provide benefits only to poor and lower-income persons. The free meals and reduced price components of the National School Lunch Program (NSLP) and the School Breakfast Program (SBP) are also means-tested. On the other hand, Social Security, Medicare, police protection, and public education are not means-tested; they provide services and benefits to persons at all income levels.

In the typical year, around 71 percent of means-tested spending comes from federal funds and 29 percent from state funds. Nearly all state means-tested welfare expenditures are matching contributions to federal welfare programs. Ignoring these matching state payments into the federal welfare system results is a serious underestimation of spending on behalf of the poor.

In FY 2008, 52 percent of total means-tested spending went to medical care for poor and lower-income persons, and 37 percent was spent on cash, food, and housing aid. The remaining 11 percent was spent on social services, training, child development, targeted federal education aid, and community development for lower-income persons and communities. Roughly half of means-tested spending goes to disabled or elderly persons. The other half goes to lower-income families with children, most of which are headed by single parents.

Growth of the Welfare State

Welfare spending has grown enormously since President Lyndon B. Johnson launched the War on Poverty. Welfare spending was 13 times greater in FY 2008, after adjusting for inflation, than it was when the War on Poverty started in 1964. (See chart 1.) Means-tested welfare spending was 1.2 percent of the gross domestic product (GDP) when President Johnson began the War on Poverty. In 2008, it reached 5 percent of GDP. Over the next decade, total means-tested spending is likely to average roughly 6 percent of GDP.

Annual means-tested welfare spending is more than sufficient to eliminate poverty in the United States. The U.S. Census Bureau, which is in charge of measuring poverty and inequality in the nation, defines a family as poor if its annual income falls below official poverty income thresholds. If total means-tested welfare spending were simply converted into cash benefits, the sum would be nearly four times the amount needed to raise the income of all poor families above the official poverty line.

Since the beginning of the War on Poverty, government has spent $15.9 trillion (in inflation-adjusted 2008 dollars) on means-tested welfare. In comparison, the cost of all other wars in U.S. history was $6.4 trillion (in inflation-adjusted 2008 dollars).

Welfare Spending Increases under the Obama Administration

Table 1 shows the growth in means-tested spending over recent years. In FY 2007, total government spending on means-tested welfare or aid to the poor was a record high $657 billion. By FY 2011, total government spending on means-tested aid will rise to $953 billion, nearly a 50 percent increase.

Table 1. Growth in Means-Tested Spending

Federal Spending

(in billions)

State Spending

(in billions)

Total Spending

(in billions)

FY 2007




FY 2008




FY 2009




FY 2010




FY 2011




President Obama’s increase in federal means-tested welfare spending during his first two years in office is two and a half times greater than any previous increase in federal welfare spending in U.S. history, after adjusting for inflation.

Supporters of the President’s spending might counter that these spending increases are merely temporary responses to the current recession. But that is not the case; most of Obama’s spending increases are permanent expansions of the welfare state. According to the long-term spending plans set forth in Obama’s FY 2010 budget, combined federal and state spending will not drop significantly after the recession ends. In fact, by 2014, welfare spending is likely to equal $1 trillion per year.

According to President Obama’s budget projections, federal and state welfare spending will total $10.3 trillion over the next 10 years (FY 2009 to FY 2018). This spending will equal over $100,000 for each taxpaying household in the U.S.

Means-Tested Welfare Spending on Lower-Income Persons

With more than 70 overlapping means-tested programs serving different low-income populations, it is difficult to determine the average level of benefits received by low-income persons. One way of estimating average welfare benefits per recipient would be to divide total means-tested spending by the total number of poor persons in the United States. According to the Census Bureau, there were 39.8 million poor persons in the U.S. in 2008, the most recent year for which data are available. An additional 1.5 million persons lived in nursing homes. (These individuals, though mostly poor, are not included in the annual Census poverty and population survey.) Total means-tested spending in 2008 was $708 billion. If this sum is divided by 41.3 million poor persons (including residents in nursing homes), the result is $17,100 in means-tested spending for each poor American.

However, this simple calculation can be misleading because many persons with incomes above the official poverty levels also receive means-tested aid. Although programs vary, most means-tested aid is targeted to persons with incomes below 200 percent of poverty. Thus, a more a accurate sense of average total welfare spending per recipient can be obtained, if total welfare aid is divided among all persons within this larger group. Dividing total means-tested aid by all persons with incomes below 200 percent of poverty results in average welfare spending of $7,700 per person, or around $30,000 for a family of four.

Means-Tested Spending on Families with Children

Another way of examining spending levels is to look at welfare spending on families with children. In FY 2011, total means-tested spending will be $950 billion. About half of this spending ($475 billion) will go to families with children. (Around one-third of this spending will go to medical care.)

If the $475 billion in welfare spending were divided equally among the lowest income one third of families with children (around 14 million families), the result would be around $33,000 per low income family with children.

In addition, most of these lower-income families have earned income. Average earnings within the whole group are typically about $16,000 per year per family (though in the midst of a recession, earnings will be lower). If average welfare aid and average earnings are combined, the total resources is likely to come to between $40,000 and $46,000 for each lower-income family with children in the U.S. It is very difficult to reconcile this level of spending with conventional claims that millions of lower-income families are chronically hungry, malnourished, or ill-housed.

Food Insecurity in America

Last November, the U.S. Department of Agriculture (USDA) released its annual report on household food security in the United States. According to USDA, some 17 million households, or 14.6 percent of all households, experienced “household food insecurity” at some point in 2008 and some 49 million people lived in households with some form of food insecurity.[1] Most of these households were low income.

While these numbers sound ominous, it is important to understand what “food insecurity” means. According to the USDA, “food insecurity” is usually a recurring and episodic problem rather than a chronic condition.[2] In 2008, around two-thirds of food insecure households experienced “low food security,” meaning that these households managed to avoid any disruption or reduction in food intake throughout the year but were forced by financial pressures to reduce “variety in their diets” or rely on a “few basic foods” at various times in the year.[3]

According to the USDA, the remaining one-third of food insecure households (around 6 percent of all households) experienced “very low food security,” meaning that at least once in the year their actual intake of food was temporarily reduced due to a lack of funds for food purchase.[4] At the extreme, 1.5 percent of all adults in the U.S. went an entire day without eating at least once during 2008 due to lack of funds for food.[5]

Poor children are generally shielded from food insecurity. Around one million children, or 1.5 percent of all children experienced “very low food security” and reduced food intake at least one time during 2008.[6] Around one child in 150 missed at least one meal in the preceding month due to food shortages in the household.[7] One child in a thousand went a whole day without eating at least once during the year because the family lacked funds for food.[8]

Political advocates proclaim that the USDA reports suggest there is widespread chronic hunger in the U.S.[9] But the USDA clearly and specifically does not identify food insecurity with the more intense condition of “hunger,” which it defines as “discomfort, illness, weakness, or pain…caused by prolonged involuntary lack of food.”[10]

Food Insecurity and Obesity

While temporary food shortages are a concern, what is rarely discussed is that the government’s own data show, paradoxically, that the overwhelming majority of food insecure adults are, like most adult Americans, overweight or obese. Among adult males experiencing food insecurity, fully 70 percent are overweight or obese.[11] Nearly three-quarters of adult women experiencing food insecurity are either overweight or obese, and nearly half (45 percent) are obese. Virtually no food insecure adults are underweight.

Food insecure men are slightly less likely to be overweight or obese than men who are food secure (70 percent compared to 75 percent). But food insecure women are actually more likely to be obese or overweight than are women who are food secure (73 percent compared to 64 percent).

Thus, the government’s own data show that, even though they may have brief episodes of reduced food intake, most adults in food insecure households actually consume too much, not too little, food over the long term. To improve health, policies must be devised to encourage these individuals to avoid chronic over-consumption of calories and to spread their food intake more evenly over the course of each month to avoid episodic shortfalls.

Eating Too Much, Not Too Little

Yet most proposed policy responses to food insecurity call for giving low-income persons more money to purchase food despite the fact that most low-income persons, like most Americans, already eat too much. Such policies are likely to make the current situation worse, not better. One commonly proposed policy, for example, is to expand participation in the Food Stamp program. Participation in the Food Stamp program, however, does not appear to reduce food insecurity. Households receiving food stamps do not have improved food security compared to similar households with the same non-food stamp income who do not participate in the program.[12] Moreover, participation in the Food Stamp program does not appear to increase diet quality. Compared to similar households who do not receive food stamps but have the same non-food stamp income, households receiving food stamps do not consume more fruits and vegetables but do, unfortunately, consume more added sugars and fats.[13]

While the Food Stamp program has little positive effect on food quality, considerable evidence indicates that the program has the counter-productive effect of increasing obesity. For example, a recent study funded by USDA found that low-income women who participate in the Food Stamp program are substantially more likely to be obese than women in households with the same non-food stamp income who did not receive food stamps. Over the long term, food stamp receipt was found to increase obesity in men as well.[14] While other research has failed to confirm this link between food stamps and obesity, the possibility that this program has harmful effects remains quite real.[15]

Similarly, the research on the relationship between school meal programs and obesity is mixed and cautionary. Some research indicates that participation in the National School Lunch Program (NSLP) leads to higher obesity among young students in kindergarten and first grade.[16] Other studies have found this effect for the School Breakfast Program (SBP) but not for the NSLP.[17]

Dispelling Misconceptions

Developing a rational policy on nutrition and poor Americans will require dispelling common misconceptions concerning poverty and obesity. For example, one common misconception is that poor people become obese because they are forced, due to a lack of financial resources, to eat too many junk foods that are high in fat and added sugar. According to this theory, poor persons struggle to obtain sufficient calories to maintain themselves and are forced to rely on junk foods as the cheapest source of calories, but because junk foods have high “energy density” (more calories per ounce of food content), these foods paradoxically induce a tendency to overeat and thereby cause weight gain.[18]

One problem with this theory is that junk foods are not a particularly cheap source of calories. For example, soft drinks are high in added sugar and are generally associated with weight gain, but as a source of calories, brand name soft drinks such as Coca-Cola and Pepsi are often more expensive (in terms of calories per dollar) than milk. Snack foods such as potato chips and donuts cost two to five times more per calorie than healthier staples such as beans, rice, and pasta. Financially strained families truly seeking to maximize calories per dollar of food expenditure would focus not on junk and snack foods but on traditional low-cost staples such as beans, rice, flour, pasta, and milk. These foods are not only less expensive but actually have below-average energy density and therefore a lower potential to promote weight gain.[19]

In reality, poor people are increasingly becoming overweight for the same reason that most Americans are becoming overweight: They eat too much and exercise too little. Like the rest of America, the poor appear to eat too many high-fat foods and foods with added sugars, but they do this for the same reason the average American over-consumes these foods: They are highly palatable. While it would be desirable for poor people (like all Americans) to drink fewer soft drinks and eat more broccoli, simply expanding the Food Stamp program and other nutrition programs would not accomplish that goal.

Child Nutrition Programs and Childhood Obesity

As noted, research on the effects of school meal and child nutrition programs on children’s weight is mixed, with some studies showing harmful effects. The most positive study of the effects of school meal programs on children’s weight was conducted by Mathematica Policy Research.[20] This analysis found that participation in the National School Lunch Program had no overall effect on children’s weight, but participation in the School Breakfast Program (SBP) did have positive effects.

The study found that participation in the school breakfast program had no impact on obesity per se, but did reduce the average body mass index (BMI) of students. The research concluded that the BMI of full-time participants in the School Breakfast Program (SBP) was 0.75 lower than the BMI for similar non-participating students. This BMI reduction is equivalent to 3 to 4 pounds for a middle school student.

Unfortunately, the cost of the SBP (around $325 per student per school year) is quite large when compared to the weight loss achieved. This means it costs over $300 per student to produce a weight reduction of three to four pounds. Moreover, this weight reduction is neither permanent nor cumulative. A student must participate in the SBP in each subsequent year in order to maintain the small effect.

The full cost for a student to participate in the SBP each year through primary and secondary school would be over $4,000. While the Mathematica study suggests that participating students may weigh a few pounds less in each year, $4,000 is a high price to pay for that modest impact. One wonders how many middle-class parents would pay more than $4,000 so that their child could weigh a few pounds less during primary and secondary school. One wonders, as well, whether there are more cost effective means to achieve this same result.

Limiting School Distribution of Low Nutrient Energy Dense Foods

One promising alternative is simply to limit the amount of low-nutrient- energy-dense (LNED) foods, such as soft drinks, candy, chips, and french fries that schools provide or make available to students. There is accumulating evidence that the consumption of LNED foods may lead to weight gain among children and youth. A logical response is for local schools to limit the amount of LNED food offered to students. (There should be no limit on the choices parents make in providing food for their children.) Changing the composition of foods offered by schools may have positive results on children’s weight and would not impose added costs on the taxpayer.

A great many schools are already adopting this sort of policy. What is needed here is flexibility and experimentation. There is, no need for mandatory national standards, nor for the U.S Congress to assume the role of national “cookie czar,” dictating food policies for local schools. Such a usurpation of power would be unwise and unwarranted.


Fiscal policy with respect to the poor must be viewed holistically. It is misleading to examine a few nutrition programs in isolation as if no other aid were given to low-income children. This is particularly important since financial resources are fungible within each household. One extra dollar in government spending on food and child nutrition programs for a family will rarely result in one extra dollar of food expenditure by the family. Instead, the main effect may be to displace cash spending on food within the household.

The federal government operates 71 different means-tested aid programs, providing cash, food, housing, medical care, and social services to poor and low-income families. In FY 2011, government will spend around $475 billion on means-tested aid for families with children. This amounts to over $30,000 for each low-income family with children. At the same time, the federal budget deficit in FY 2011 will be $1.2 trillion, or 8.3 percent of gross domestic product. The nation simply cannot afford the current level of spending. In this context, the call for even more funding for school nutrition programs is unsupportable.

Moreover, there is little or no evidence suggesting that government spending on child nutrition programs can be a cost effective means of reducing overweight and obesity. Instead, reducing consumption of low nutrient energy dense foods may be a promising means to limit weight gain among children. Schools can accomplish this by limiting the amount of such food they provide to students. This can be accomplished without added costs to taxpayers. The implementation of such policies should be determined by local schools and should not be mandated by the federal government.

Chart3 sr67

Show references in this report

[1]Mark Nord, Margaret Andrews, and Steven Carlson, Household Food Security in the United States, 2008, ERR-83 U.S Department of Agriculture, Economic Research Service, November 2009.

[2] Ibid., p. 9.

[3] Ibid., p. 4.

[4] Ibid.

[5] Ibid., p. 45.

[6] Ibid., p. 7.

[7] Ibid., p. 47.

[8] Ibid., p. 42.

[9]See Food Research Action Council, “Hunger in the United States,” January 17, 2007, at www.frac.org/html/hunger_in_the_us.

[10]Nord et al., “Household Food Security in the United States, 2008,” p. 52.

[11]The shares of food secure and food insecure individuals who are underweight, overweight, and obese was calculated using body mass index (BMI) data and food security data from the 2003–2004 National Health and Nutrition Examination Survey (NHANES). The BMI cutoff points for underweight, normal weight, overweight, and obese were calculated using the BMI ranges for adults as reported by the Centers for Disease Control. Specifically, an adult with a BMI of less than 18.5 is underweight; between 18.5 and 24.9 is within the normal weight range; between 25 to 29.9 is considered overweight; and at or above 30 is obese. See Centers for Disease Control, “National Health and Nutrition Examination Survey,” November 2007, at www.cdc.gov/nchs/nhanes.htm, and Centers for Disease Control, “About BMI for Adults,” May 22, 2007, at www.cdc.gov/nccdphp/dnpa/bmi/adult_BMI/about_adult_BMI.htm.

[12]Craig Gunderson and Victor Oliveira, “The Food Stamp Program and Food Insufficiency,” American Journal of Agricultural Economics, November 2001.

[13]Parke E. Wilde, Paul E. McNamara, and Christine K. Ranney, “The Effect of Income and Food Programs on Dietary Quality: A Seemingly Unrelated Regression Analysis with Error Components,” American Journal of Agricultural Economics, November 1999.

[14]Charles Baum, “The Effects of Food Stamps on Obesity,” U.S. Department of Agriculture, Contractor and Cooperator Report No. 34, September 2007.

[15]Michele Ver Ploeg, Lisa Macino, Biing-Hwan Lin, Food and Nutrition Assistance Programs and Obesity:1976-2002, ERR-48, U.S Department of Agriculture, Economic Research Service, September 2007.

[16]Diane Whitmore Schanzenbach, “Do School Lunches Contribute to Childhood Obesity?” Chicago Illinois, Harris School Working Paper 5-13, October 2005.

[17]D.T. Millimet et al., “School Nutrition Programs and the Incidence of Childhood Obesity,” IZA DP:3664, Bonn, Germany, August 2008

[18]Adam Drewnowski and S.E. Spencer, “Poverty and Obesity: The Role of Energy Density and Energy Costs,” American Journal of Clinical Nutrition, January 2004, pp. 6–16.

[19]Barbara Rolls and Robert A. Barnett, The Volumetrics Weight-Control Plan (New York: HarperCollins, 2000), pp. 124–25.

[20]Phillip Gleason, et al., School Meal Program Participation and Its Association with Dietary Patterns and Childhood Obesity, Final Report, Mathematica Policy Research Inc. July 2009.

Heritage's daily Morning Bell e-mail keeps you updated on the ongoing policy battles in Washington and around the country.

The subscription is free and delivers you the latest conservative policy perspectives on the news each weekday--straight from Heritage experts.

The Morning Bell is your daily wake-up call offering a fresh, conservative analysis of the news.

More than 200,000 Americans rely on Heritage's Morning Bell to stay up to date on the policy battles that affect them.

Rush Limbaugh says "The Heritage Foundation's Morning Bell is just terrific!"

Rep. Peter Roskam (R-IL) says it's "a great way to start the day for any conservative who wants to get America back on track."

Sign up to start your free subscription today!