On Thursday, December 18, the United States Conference of Mayors will release its annual report on hunger and homelessness in America. The mayors have released a similar report each year since 1987. The report measures "hunger" by the number of persons using food banks or soup kitchens. The Conference of Mayors has reported that the number of persons using food banks or soup kitchens in major cities has increased substantially in each of the past 16 years, and it is expected that a similar increase will be reported this year.
The mayors' hunger reports, however, are vague. They do not give the number of persons using food banks or soup kitchens. Instead, they merely report the rate of increase in use compared to the prior year. In 2002, for example, the report stated that emergency food use had increased by 19 percent relative to 2001.
According to the mayors' data, the number of persons receiving emergency food aid is 12 times higher today than in 1986. This seems implausible. The U.S. Department of Agriculture (USDA) reports that, at present, between 18 million and 24 million persons receive emergency food aid each year. If the mayors' data were accurate and representative of the nation, it would mean that fewer than 2 million persons received food aid in 1986 (one-twelfth of the present number).
- Are contradicted by U.S. Census Bureau surveys. Census data show no increase in the use of food pantries and soup kitchens in either central cities or the nation as a whole between 1995 and 2001. By contrast, the mayors' reports claim an increase in use of 150 percent during the same period.
- Are contradicted by Second Harvest reports. The mayors' data are contradicted by detailed surveys conducted by Second Harvest, the major supplier to food banks. Second Harvest reports that emergency food use increased by 9 percent between 1997 and 2001. The mayors' reports claim that emergency food use increased by nearly 100 percent during the same period.
The mayors' annual reports show an implausible growth in food bank use throughout the past 16 years. Their data are contradicted by other more reliable surveys. One possible explanation is that the 25 major cities surveyed in the mayors' reports are very unrepresentative of the nation as a whole. This seems unlikely. It seems more likely that the mayors' survey methods, which rely on self reports of food bank operators and employ vague counting standards, are flawed and their conclusions inaccurate.
Each year, the U.S. Conference of Mayors releases A Status Report on Hunger and Homelessness in America's Cities, which presents an estimate of housing and homeless data from the previous year in major U.S. cities.1 Year after year, these documents consistently report that hunger and homelessness are "on the rise in major U.S. cities."2
The Conference of Mayors began to collect data on hunger and homelessness in 1983 and has been releasing these data in their current form since 1987. The most recent report, from 2002, is compiled from a survey of the 25 U.S. cities whose mayors were members of the Conference's Task Force on Hunger and Homelessness.3 It is expected that the 2003 version will be released on Thursday, December 18.
In compiling the report, the Conference distributes questionnaires to each city. These questionnaires ask for estimates of (1) the demand for emergency food assistance and emergency shelter and the capacity of local agencies to meet that demand, (2) the causes of hunger and homelessness and the demographics of the populations experiencing these problems, (3) exemplary programs of efforts in the cities to respond to hunger and homelessness, (4) the availability of affordable housing for low-income people, and (5) the outlook for the future and the impact of the economy on hunger and homelessness.
The 2002 report finds that requests for emergency food assistance--provided through a variety of resources including food banks, food pantries, and soup kitchens--increased by an average of 19 percent over the past year. Requests increased by an average of 17 percent among families with children and 19 percent among elderly persons.4 Similar findings are expected in the upcoming 2003 Conference of Mayors report.
As noted above, the mayors' report does not give figures on the number of persons receiving emergency food aid. Instead, it simply reports the percentage of growth in emergency food use from year to year. Over the past 16 years, the mayors have reported that food bank use has grown at an average rate of 17 percent per year. (See Appendix, Table A.) According to the mayors' reports, food bank use roughly doubles every four years.
When the mayors' annual growth figures are seen in aggregate, the picture is quite alarming. Chart 1 shows the growth in emergency food use from 1986 to 2001 according the mayors' data.5 (Since the mayors' reports do not specify actual numbers of users, 1986 is set as base year equaling 100.) If the mayors' figures are accurate, there were 12 times more people receiving food aid in 2001 than in 1986.
The United States Census Bureau also tracks utilization of food pantries, food banks, and soup kitchens in its annual Current Population Survey. In 1995, the Census Bureau began asking Americans two questions: "In the last 12 months, did you ever get emergency food from a church, a food pantry, or a food bank?" and "In the last 12 months did you ever eat any meals at a soup kitchen?"6 These questions were repeated in 1997, 1999, and 2001.
The results are shown in Table 1. In the seven-year period between 1995 and 2001, the Census survey showed that the number of persons using food pantries or soup kitchens dropped slightly from 9.2 million to 8.7 million. The number of persons receiving emergency food in central cities remained unchanged at 2.8 million.
It is important to note that the number of persons using emergency food each year does appear to be significantly underreported in the Census surveys. However, since the Census methodology is exactly the same each year, there is no reason to believe that the undercounting would have increased or de-creased between 1995 and 2001. Thus, while the absolute numbers of persons reported to receive emergency food in Table 1 may be too low, the trend (increase or decrease) over time reported by the Census Bureau is likely to be fairly accurate.
As Chart 2 shows, the trend reported by the Conference of Ma-yors over the same period is radically different. While the Census shows no growth in emergency food use between 1995 and 2001, the mayors' reports show that emergency food use in major cities increased 150 percent.
Mayors' Reports Contradicted by Second Harvest Surveys
America's Second Harvest, which is the largest food distributor in the nation, consisting of a network of nearly 200 banks, also provides data on the incidence of hunger and food insecurity in the United States. In its Hunger 1997: The Facts and Faces report, Second Harvest estimated that 21.4 million people sought emergency food assistance from a food pantry, food kitchen, or shelter at least once during 1997.7 In the next version of the report, Hunger in America 2001, Second Harvest reported that 23.3 million clients sought emergency food assistance at least once during 2001.8 Thus, the two Second Harvest surveys show a 9 percent increase in emergency food use in the four years between 1997 and 2001.
By contrast, the mayors' hunger reports show that emergency food use increased by almost 100 percent between 1997 and 2001. This rate of increase is 10 times greater than that shown by Second Harvest. If the mayors' growth figures were accurate and representative of the nation, the number of persons receiving emergency food aid would have grown from 21.4 million in 1997 (as reported by Second Harvest) to around 40 million in 2001. That figure would be 17 million higher than the figure reported by Second Harvest for 2001.
The U.S. Conference of Mayors report may not provide an accurate or complete understanding of the utilization of food pantries and food banks for several reasons. First, the data are compiled based on surveys of city officials, with very minimal apparent documentation requirements. City officials in the 25 surveyed cities are asked to report changes in demand at emergency food providers and city services by reporting whether the demand for emergency food assistance has increased, decreased, or stayed the same, and by what percentage. Respondents are simply encouraged to "include any other data which supports it."9
Another reason that the mayors' survey may overcount the growth of emergency food use may be that it fails to account properly for the food pantries that close down. According to USDA estimates, approximately 20 percent of emergency food providers go out of business each year.10 New providers generally open to replace those that close.
Scrupulous accounting of the food banks that close is necessary for an accurate assessment of changes in emergency food use. If a surveyor reports only increases in food delivery in newly opened and continuing food banks, but fails to account fully for banks that ceased operation, the result will be a considerable overcount in the growth rate of emergency food use. Since the mayors' report provides only the sketchiest account of how its numbers are collected, it is impossible to determine to what extent this problem may contribute to the mayors' very rapid growth figures.
In 2001, to determine the number of persons using food banks and soup kitchens in the United States, the USDA conducted the National Emergency Food Assistance System Client Survey. On the basis of that survey, the USDA estimates that 4.3 million persons are served by food pantries in a given week.
Using these figures, in addition to the number of times a person reports visiting a food pantry during a given month, the USDA estimates that 12.5 million people are served by a food pantry in a given month and between 18 million and 24 million persons are served at least once during the course of a year. These figures are generally consistent with the estimate made by America's Second Harvest of 23.3 million people seeking emergency food assistance from a food bank, pantry, or kitchen at least once during 2001.
About three-quarters of the individuals receiving emergency food say they prefer getting free food from a pantry, as opposed to using government aid, such as food stamps.11 Roughly three-quarters of the households using food pantries have no employed members.12
How many people in the United States experience hunger? The best answer to that question is provided by the Household Food Security Survey, conducted by the U.S. Department of Agriculture each year since 1995. The USDA defines hunger not as the use of food pantries, but as physical discomfort caused by actual food shortages due to a lack of funds to obtain food. The USDA makes clear that hunger is not the same as malnutrition and that most hunger experienced in the United States is short-term.13
According to the USDA, on a typical day, less than one American in 200 will experience hunger due to a lack of money to buy food.14 The hunger rate rises somewhat when examined over a longer time period; according to the USDA, some 6.9 million Americans (2.4 percent of the population) were hungry at least once during 2002.15 Nearly all hunger is short-term and episodic rather than continuous.
Some 92 percent of those who experienced hunger in 2002 were adults, and only 8 percent were children. Overall, some 567,000 children (or 0.8 percent of all children) were hungry at some point in 2002.16 In a typical month, roughly one child in 400 skipped one or more meals because the family lacked funds to buy food.
According to the USDA, overall hunger has declined slightly since measurement be-gan in 1995. In that year, 4.1 million households had at least one person who experienced hunger at some point during the year; by 2002, the number had fallen to 3.8 million households.17
Hunger among children, however, has declined substantially since the mid-1990s. As Chart 4 shows, the number of hungry children was cut by a third between 1995 and 2002. According to the USDA, in 1995, there were 887,000 hungry children; by 2002, the number had fallen to 567,000.18
The USDA also reports on the number of households that are "food insecure without hunger." Advocacy groups often label these families as "at risk" of hunger, although the USDA explicitly states that these households are not hungry and do not face food shortages. These families do face financial constraints in purchasing food at some point or points during the year, and they report anxiety that at some future time they may not be able to buy sufficient food. They may temporarily substitute cheaper foods for regular items in their diet.
According to the USDA, 25.5 million individuals (or 9.1 percent of the population) were "food insecure without hunger" at some point during 2002. This condition is generally temporary. The number of persons who were food insecure without hunger dropped slightly between 1995 and 2002.19
Overall, some 97 percent of the U.S. population lives in families that reported they had "enough food to eat" during the entire year, although not always the kinds of foods they would prefer. Around 2.5 percent stated that their families "sometimes" did not have "enough to eat" due to money shortages, and one-half of 1 percent (0.5 percent) said they "often" did not have enough to eat due to lack of funds.20
According to the U.S. Department of Agriculture, the number of Americans who are "hungry" has dropped slightly since measurement began in 1995. The number of hungry children, however, has declined substantially.
Policymakers should be wary of the claims of "increasing hunger" that are likely to be made in the upcoming 2003 U.S. Conference of Mayors report. Year after year, the mayors' hunger report shows an alarming increase in use of food banks. The mayors have reported that use of food banks in major cities has increased at an average rate of 17 percent per year for the past decade and a half. According to the mayors, food-bank use roughly doubles every four years.
The mayors' figures, however, seem implausible; they are contradicted by other more reliable surveys. The continuing broadcast of alarming but inaccurate figures can only distract from real problems facing the nation.
Melissa G. Pardue is the Harry and Jeanette Weinberg Fellow in Social Welfare Policy, Robert E. Rector is Senior Research Fellow in Domestic Policy Studies, and Kirk A. Johnson, Ph.D., is the Harry and Jeanette Weinberg Fellow in Statistical Welfare Research in the Center for Data Analysis at The Heritage Foundation .
1. U.S. Conference of Mayors, A Status Report on Hunger and Homelessness in America's Cities 2002: A 25-City Survey, December 2002, at www.usmayors.org/uscm/hungersurvey/2002/onlinereport/
3. The mayors of the 25 cities that participated in the 2002 survey are Thomas Menino, Boston, Massachusetts; Peter Clavelle, Burlington, Vermont; Joseph Riley, Charleston, South Carolina; Patrick McCrory, Charlotte, North Carolina; Richard M. Daley, Chicago, Illinois; Jane Campbell, Cleveland, Ohio; Wellington Webb, Denver, Colorado; Kay Barnes, Kansas City, Missouri; James Hahn, Los Angeles, California; David Armstrong, Louisville, Kentucky; Manny Diaz, Miami, Florida; Bill Purcell, Nashville, Tennessee; Ray Nagin, New Orleans, Louisiana; Paul Fraim, Norfolk, Virginia; John Street, Philadelphia, Pennsylvania; Skip Rimsza, Phoenix, Arizona; Vera Katz, Portland, Oregon; John Lombardi, Providence, Rhode Island; Rocky Anderson, Salt Lake City, Utah; Ed Garza, San Antonio, Texas; Greg Nickels, Seattle, Washington; Francis Slay, St. Louis, Missouri; Randy Kelly, St. Paul, Minnesota; Doug Palmer, Trenton, New Jersey; and Anthony A. Williams, Washington, D.C.
5. Based on numbers provided in "Hunger and Homelessness in America's Cities: A Sixteen-Year Comparison of Data," in U.S. Conference of Mayors, A Status Report on Hunger and Homelessness in America's Cities.
8. See Key Findings in America's Second Harvest, Hunger in America 2001, at www.secondharvest.org/site_content.asp?s=82#.
9. See copy of survey, which is included in U.S. Conference of Mayors, A Status Report on Hunger and Homlessness in America's Cities 2002. The authors of this paper repeatedly attempted to contact the person at the Conference of Mayors who compiles the results of the survey to learn more about the methodology used to create the report, but calls were not returned.
15. Mark Nord, Margaret Andrews, and Steven Carlson, Household Food Security in the United States, 2002, U.S. Department of Agriculture, October 2003, p. 7. The numbers in the text were taken from Table 1 of the USDA publication. Many individuals reside in households where at least one family member but not all family members experienced hunger. This is particularly true among families with children where the adults are far more likely than the children to experience hunger. According to Table 1of Household Food Security in the United States, 2002, 9.3 million persons lived in households where at least one household member experienced hunger; however, not all of these persons experienced hunger themselves. The number of persons who experienced hunger individually was lower: 6.8 million people, including 6.3 million adults and 567,000 children.
17. See Household Food Security in the United States in 1995, Summary Report of the Food Security Measurement Project, U.S. Department of Agriculture, September 1997, p. 48. See also Nord, Andrews, and Carlson, Household Food Security in the United States, 2002, Figure 2, p. 4.
19. See Nord, Andrews, and Carlson, Household Food Security in the United States, 2002, Figure 2, p. 4, or trend chart at www.ers.usda.gov/Briefing/FoodSecurity/trends/.