Click for next page ( 19


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 18
Overview of WIC and the Current Estimation Methodology WIC started as a pilot program in 1972 and became a permanent program in fiscal year (FY) 1974, at which time it grew rapidly. The num- ber of women, infants, and young children served per month increased from 0.2 million in FY 1974, to 3.6 million in FY 1988, to 7.2 million in FY 2000. WIC provides three types of benefits: supplemental food, usually in the form of vouchers or checks that can be exchanged for specific foods from participating retail grocers; nutrition education; and referrals to health care and to other social services. The federal government gives grants to states, territories, and Indian tribes to provide the supplemental food, nutrition education, and health and social service referrals and to administer the program. State grant allo- cations are based on the amount the state received in the previous year and on the estimated number of income-eligible infants and children in that state.1 States then fund local agencies that actually provide the services to participants. Since WIC is not an entitlement program, allocated amounts of funding may not be sufficient to serve all the eligible persons who wish to participate. If local agencies do not receive enough funds to serve all eligible applicants, they establish a prioritized waiting list. Federal regula- tions specify a seven-point priority system (7CFR Subpart C, Section These estimates of state levels of income eligibility are separate from the national-level estimates, but the sum of allocations to all states equals the national estimates. 18

OCR for page 18
OVERVIEW OF WIC AND THE CURRENTESTIMATIONMETHODOLOGY 19 246.7(d)4; Federal Register, April 19, 1995, 60 (75)19, 487-489, 491), in which priority is based on the type of nutritional risk and the eligibility category. In general, pregnant and breastEeeding women and infants have higher priority than children and nonbreastEeeding postpartum women. Within these groups, those who are nutritionally at risk based on anthropo- metric, hematologic, or other nutrition-related medical conditions have higher priority than those who qualify as nutritionally at risk based on inadequate diet (see U.S. Department of Agriculture, 2002, for further information on the WIC priority system). The last year a state had to imple- ment a priority waiting list was 2002. States that experienced shortages of funds to serve all eligible applicants in 2002 obtained supplemental fund- ing from the federal government. This chapter explains how the estimates of eligibility and participation are used by the U.S. Department of Agriculture (USDA) and why the esti- mates are made. It also includes information on eligibility for WIC and a description of current USDA methods used to estimate eligibility and par- . . . . . tlclpatlon natlonwl( ~e. PURPOSES OF ESTIMATING ELIGIBILITY AND PARTICIPATION The USDA estimates of eligibility and participation in the WIC pro- gram are used to guide budget requests for the program each year. The estimates of eligibility and the estimates of participation among eligibles are used for additional purposes for example, to evaluate the effectiveness of the program in serving the eligible population. Budgetary Planning In the recent past, it has been the goal of both Congress and the ad- ministration to fully fund WICthat is, to provide enough funds so that everyone who is eligible and wanted to participate in the program could. USDA has used estimates of the number of full-funding participants to inform budgetary requests. In the mid- to late 1990s, many observers came to the conclusion that the full-funding level of participants had been reached. This was because states had some unspent funds for FY 1996 and because coverage rates for the program (the number of participants divided by the number of people estimated to be eligible for the program) were well over 100 percent for infants and pregnant women throughout the late

OCR for page 18
20 ESTIMATING ELIGIBILITYANDPARTICIPATIONFOR THE WICPROGRAM 1990s. However, some observers claimed that the number of people eli- gible for the program was underestimated, that there was still unmet need, and that more funding should be allocated to meet these needs. As dis- cussed in Chapter 1, these two views were the impetus for the USDA re- quest for this study. Throughout this debate concerning the estimates of eligibility and par- ticipation, USDA has continued to make estimates ofthe number of people eligible and likely to participate in WIC. However, it has not made any changes to its methodology to address these criticisms because it is waiting for this panel's report. Although estimates of eligibility and participation continue to be made, they have not been used to inform budgetary re- quests. Instead, budgets have been set to serve a particular number of people believed to be close to the number of fully funded participants (about 7.5 million). Is WIC Fully Funded' As we discuss in Chapter 3, the number of participants served in a given year is very close to the number the budget was intended to support. A naive assessment might conclude that the methods used to estimate par- ticipation to inform budgetary requests are reasonably accurate. But this apparent accuracy is, in reality, an artifact of a circuitous process. Fund allocations based on participation estimates are made to states, and then the states use the funds to serve as many eligible applicants as they can with the funds. It is not surprising, then, that the number of participants served is very close to the number for which the budget was set. In the event that more eligible people apply for WIC than were originally budgeted for, states may implement a priority waiting system, or it is possible for states to receive supplementary funding to serve all those who are eligible, but in the recent past, waiting lists and supplementary funds have very rarely been needed. It is not correct, however, to conclude that, since waiting lists and supplementary funds have not been needed, WIC has been fully funded. The number of participants under full funding is a policy choice that is, the number can be altered by changes in program rules or administrative practice. Furthermore, there may be many people who are eligible for WIC and who could benefit from the program, but who are unaware of it or cannot easily take part. Local outreach efforts or changes in the administra- tion of the program (e.g., more conveniently located offices, evening office

OCR for page 18
OVERVIEW OF WIC AND THE CURRENTESTIMATIONMETHODOLOGY 21 hours) may mean that a greater percentage of eligible people apply for the program. States receive funds that can be used for program outreach, but they have an incentive to increase program participation only to the level that their allocated funds allow them to serve additional participants. It is inherently more difficult to estimate the number of likely participants be- cause participation in WIC is a behavioral choice.2 Some individuals may decide that the benefits of WIC are too low to offset the time spent apply- ing for the program, or that the stigma of participation is too high, or some eligible people may not be aware of the program. The panel focuses much more on the estimation of eligibility than on participation because eligibil- ity is less of a behavioral choice than is the decision to participate. Coverage Rate Estimates Estimates of the number of eligible persons are also important for pro- grammatic reasons. Each year the Food and Nutrition Service (ENS) of USDA publishes coverage rates for the WIC program. These rates estimate the percentage of eligible persons by category who participate in WIC. The estimates are computed by dividing the number of WIC participants in a year (based on administrative records) by the number of people estimated to be eligible for that year. USDA's estimated coverage rates, by category, for the past eight years are given in Table 2-1. These coverage rate estimates have been used by ENS and policy makers to gauge the span of the pro- gram over the eligible population. Notably, coverage rates for infants and postpartum women have exceeded 100 percent since 1994. In 2000, cover- age rates for infants were 131 percent and coverage rates for postpartum women were 134 percent, indicating that over 30 percent more infants and women participated in WIC than were estimated to be eligible. Mean- while, coverage rates for children and pregnant women were much lower, generally between 60 and 70 percent. Coverage rates in excess of 100 per- cent have prompted some members of Congress to raise concerns that in- eligible people are participating in the program. 2There may be a behavioral component to eligibility also. An individual who is aware of the WIC program may change her behavior to become eligible for the program (e.g., WIC's benefits could offset some of the time a mother may have to work in order to gain the same level of consumption).

OCR for page 18
22 ESTIMATING ELIGIBILII~YANDPARTICIPATIONFOR THE WICPROGRAM TABLE 2-1 USDA Estimates of Coverage Rates of Infants, Postpartum and Pregnant Women, and Young Children Postpartum Pregnant Year Infants Women Women Children 1993 97.8 78.9 52.3 48.0 1994 111.0 101.2 59.0 56.9 1995 109.4 105.3 58.0 64.4 1996 113.8 117.2 62.0 69.5 1997 121.7 121.7 69.1 74.5 1998a 127.7 127.4 72.9 74.4 1999a 130.4 130.1 72.4 76.0 2000 130.9 134.1 72.9 78.8 aThe coverage rate estimates for 1998 and 1999 are unofficial USDA estimates pro- vided to the panel to show what the estimates would be using the existing methodology for those two years. NOTE: Coverage rates are defined as the average monthly number of WIC participants (from administrative data) divided by the estimated number of eligible people (from the Current Population Survey) for each category for a given year. Estimating the Effects of Changes in Program Policies Estimates of eligibility and of participation among eligible people are also important to understanding how changes in the rules or administrative policies of WIC or of other programs affect eligibility and participation. WIC program rules and administrative policies change. Such changes may affect how many people are eligible for WIC and what percentage of those who are eligible are likely to participate. Furthermore, changes in other assistance programs can affect the number of people eligible and likely to participate in WIC because those who are enrolled in Medicaid, Temporary Assistance for Needy Families (TANF), and food stamps are adjunctively eligible for WIC. The only way to really understand how changes in these factors affect participation levels (and thus budget needs) is to use a model- based approach to estimating WIC participation, in which participation decisions of individuals are modeled. We return to this issue in Chapter 8. Understanding External Influences on WIC Eligibility and Participation External influences may affect the number of people eligible and the likelihood that they will participate. For example, a downturn in the

OCR for page 18
OVERVIEW OF WIC AND THE CURRENTESTIMATIONMETHODOLOGY 23 economy may mean more people are eligible for the program and more may participate. Changes in the birth rates of certain subgroups with a higher propensity to participate in WIC may affect program participation as well. Lower participation by some subgroups points to areas in which outreach may need to be more effective. WIC ELIGIBILITY Types of Eligibility and Certification Practices To receive WIC benefits, an applicant must be categorically eligible, income eligible, and nutritionally at risk. Applicants must also be residents of the state in which they apply. Box 2-1 outlines the eligibility require- ments for WIC. To be categorically eligible, an applicant must be an infant or child under the age of 5 years, a pregnant woman, a nonbreastEeeding postpartum woman less than 6 months postpartum, or a breastSeeding post- partum woman less than 12 months postpartum. The supplemental foods provided differ for each of the five eligibility categories. For example, the

OCR for page 18
24 ESTIMATING ELIGIBILITYAND PAR TICIPATION FOR THE ~CPROGR~ food provided to a nonbreastSed infant includes infant formula, while the food provided to a child includes milk, vitamin C-rich juice, iron-fortified cereal, and eggs. To be income eligible, an applicant's income must be less than or equal to 185 percent of the federal poverty guidelines, defined by the U.S. De- partment of Health and Human Services (DHHS) poverty guidelines ac- cording to family size. Those who are enrolled in the federal Medicaid, Food Stamp, or TANF programs are adjunctively eligible for WIC even if their income exceeds 185 percent of poverty. States also may use enroll- ment in other means-tested programs, such as the National School Lunch Program and the Supplemental Security Income (SSI) program, to qualify an applicant as automatically income eligible for WIC. To be considered nutritionally at risk, an applicant must meet at least one of the many approved nutritional risk criteria. These risk criteria fall under five broad categories: anthropometric risk (e.g., underweight, obe- sity); biochemical risk (e.g., low hematocrit); medical risk (e.g., diabetes mellitus); dietary risk (e.g., inappropriate dietary patterns); and predispos- ing factors (e.g., homelessness). Once it is determined that an applicant is fully eligible for WIC, she is then certified to receive WIC benefits for a period of time. The certification period over which an eligible applicant may receive monthly WIC benefits varies by the category of eligibility. Pregnant women can be certified from the time they become pregnant through 6 weeks postpartum. Postpartum women are certified for up to 6 months after giving birth if they are not breastSeeding and up to a year after giving birth if they breastSeed for more than 6 months. Infants are certified for 6 months or until they reach age 1 year most often for their entire first year. Children are certified every 6 months, but not beyond their fifth birthday. If the eligibility status of a participant changes during the certification period, it should be reported to WIC staff. Except for the case in which a breastSeeding mother changes her status to nonbreastSeeding, which results in an increase in the amount of infant formula received, there is little incentive to declare changes in status. Flexibility in Determining Eligibility Regulations governing the administration of the WIC program give considerable flexibility to local WIC agencies to determine whether an applicant is eligible. For example, WIC rules allow state and local agencies flexibility in the documentation used to verify an applicant's income and

OCR for page 18
OVERVIEW OF WIC AND THE CURRENTESTIMATIONMETHODOLOGY 25 in the time period (weekly, monthly, annually) covered by the documenta- tion. Such variation among agencies creates uncertainty in selecting the most appropriate methods to estimate the number of people eligible for WIC on a national level. In this section, we discuss areas in which some flexibility is allowed namely, the accounting period for measuring in- come and nutritional risk criteria. The accounting period for measuring income is likely to have the largest impact on estimates of eligibility made using nationally representative surveys. In Chapter 5 we provide estimates of eligibility using different accounting periods across multiple years of data. Accounting Periodtfor Income State and local agencies have wide discretion over what accounting period is used to determine whether an applicant's income meets eligibility guidelines. Agencies may consider annual income or current income (ei- therweekly or monthly). In 1998, 87.7 percent of local agencies accepted a pay stub, 80.2 percent accepted a W-2 form, 72.1 percent accepted a letter from an employer, and 81.1 percent accepted a tax form as documentation of income (United States Department of Agriculture, 2000b). We do not have information on what percentage of applicants actually present the different forms of income documentation.3 USDA specifies only that the agency should consider whichever period is a more accurate indicator of the family's status. For example, an unemployed applicant may report in- come in the period during which she is unemployed, but a teacher who is paid on a 10-month basis should be asked to report income during the period for which she is employed. Variation in the application of rules about the accounting period for income means that definitions of income used to estimate eligibility may not exactly match definitions of income as they are applied by local WIC offices in assessing eligibility. For instance, for those whose income varies over the course of a year, monthly incomes may be below the eligibility threshold but annual incomes may be above the eligibility threshold. National estimates of eligibility based on a monthly measure are likely to be greater than eligibility estimates based on an annual measure. Chapter 5 addresses this topic in more detail. 3Given that a pay stub is probably the most readily accessible form of documentation, it is likely to be the most widely used.

OCR for page 18
26 ESTI~TING ELIGIBILI>~D PAR TICIPATION FOR THE ~CPROG~ Nutritional Risk Criteria Prior to 1998, the nutritional risk criteria used by states were unstandardized, and states had wide latitude in determining which criteria to use and the cutoff values that would separate persons at risk from those not at risk. States now have adopted standardized anthropometric, medical, predisposing, and certain dietary risk criteria. However, they still are al- lowed to use cutoff values that are more stringent than those in the stan- dardized list of criteria, and they may omit the use of some of the standard- 1ze( ~ criteria. Since two widely used types of dietary risk criteria, failure to meet dli- etary guidlelines and inadlequate diet, still are unstandardized across states, local WIC offices use different methods for determining if a person is at dietary risk. Based on a recent Institute of Medicine report (2002) that reviewed the scientific basis for these two types of dietary risk, we anticipate that states soon will work to adopt standardized criteria for those two types of dietary risk as well. In the meantime, variation in local methods to deter- mine nutritional risk could contribute to inaccuracies in national estimates of the numbers of persons eligible for WIC. CURRENT METHODS FOR ESTIMATING ELIGIBILITY AND FULL-FUNDING PARTICIPATION This section provides an overview of the methods currently used by USDA to make national estimates of the numbers of women, infants, and young children who are eligible for WIC and likely to participate if the program is fully funded. We first describe the methods USDA uses to esti- mate the number of persons fully eligible for WIC, that is, the number who meet categorical, income or adjunctive, and nutritional eligibility criteria. Boxes 2-2 through 2-6 provide an overview of the estimation steps for each eligibility category. We then describe the method used to estimate how many fully eligible persons will participate if funds are available. Greater detail on the currently used methods can be found in U.S. Department of Agriculture (1999a).

OCR for page 18
OVERVIEW OF WIC AND THE CURRENTESTIMATIONMETHODOLOGY 27 Estimates of the Number of Fully Eligible Persons Estimates of the numbers of persons eligible for WIC are produced separately for each category of eligibility. The estimate of the number of infants who are eligible is especially important to obtaining an estimate of the total number eligible for WIC because it is used to derive the numbers of eligible pregnant and postpartum women. Estimates of income-eligible infants and children are referred to as the "core" estimates because they are

OCR for page 18
28 ESTIMATING ELIGIBILITYANDPARTICIPATIONFOR THE WICPROGRAM calculated directly from nationally representative survey data on those two groups. USDA uses the March Demographic Supplement ofthe Current Popu- lation Survey (CPS) to estimate the number of infants and children living in families with annual incomes below 185 percent offederal poverty guide-

OCR for page 18
OVERVIEW OF WIC AND THE CURRENTESTIMATIONMETHODOLOGY 29 lines.4 The March CPS gathers data on family income for the previous calendar year (e.g., the 2002 March CPS collected income for calendar year 20011. Appendix A provides more detail on the March CPS. In assessing family income in the eligibility determination process, states can adopt either the income guidelines for the Free and Reduced Price School Lunch Program or the income guidelines for free or reduced price health care programs. Box 2-7 lists income sources counted under both guidelines. If the free and reduced price health care definition is used, the following sources of income are excluded for determining eligibility: the value of in-kind housing and other in-kind benefits and payments or benefits provided under certain federal programs (e.g., some social pro- 4The DHHS poverty guideline should not be confused with the Census Bureau's pov- erty thresholds, which are used to provide annual estimates of the number of individuals and families in poverty. The DHHS poverty guidelines that are used to administer many low- income transfer programs take into account only family size. The Census Bureau's poverty thresholds account for differences in family size, number of children, and age of the head of the family. DHHS issues new poverty guidelines annually, and WIC program offices begin using the new thresholds in June. For the purposes of developing eligibility estimates, USDA averages the guideline from the previous year with the guideline for the year of the CPS survey. For example, the March 2002 survey provides income data for calendar year 2001. To estimate the number of income-eligible individuals in 2001, USDA would average the guide- lines for 2000 and 2001.

OCR for page 18
30 ESTIMATING ELIGIBILITYANDPARTICIPATIONFOR THE WICPROGRAM grams, such as the Low Income Home Energy Assistance Program, the value of benefits from the Free and Reduced Price School Lunch Program, and payments to members of various Indian tribes). For the purposes of estimating income eligibility, current USDA meth- ods use annual census money income to define income, which includes all the sources listed in Box 2-7. Families are defined using the Census Bureau's family definition that is, a group of two or more people related by birth, marriage, or adoption and residing together, including related subfamily members. To obtain the number of income-eligible pregnant women, the current USDA method multiplies the number of income-eligible infants by 0.75. This adjustment accounts for the pregnancy lasting for 9 months of a year. The number of births is assumed to be constant over the time period be- tween when the estimates of infants are made and the 9 months prior to the birth of the infant.

OCR for page 18
OVERVIEW OF WIC AND THE CURRENTESTIMATIONMETHODOLOGY 31 To obtain estimates of the numbers of breastEeeding and nonbreast- feeding postpartum women, the estimate of income-eligible infants is first adjusted downward slightly to account for multiple births and infant deaths. (The number of income-eligible infants is multiplied by 0.98441.5 An ad- justment is then made for the percentage of women who breastEeed and the duration of breastEeeding. The adjustment rate used to obtain the number of women who do not breastEeed and are less than 6 months postpartum is 0.374 (37.4 percent). The adjustment rate is 0.171 (17.1 percent) for the number who breastEeed and are less than 12 months postpartum. These adjustment factors are based on data from the 1988 National Maternal and Infant Health Survey (NMIHS) and were derived using life table methods that controlled for the age and income of mothers. To estimate adjunctive eligibility for each of the categories, the USDA method makes a modest adjustment for infants and children who may be eligible for WIC because they are adjunctively eligible through participa- tion in the Medicaid, Food Stamp, or TANF programs.6 The method adds 14,000 infants and 76,000 children to the core estimates of the number of income-eligible infants and children. No adjustment is made for pregnant or postpartum women who may be adjunctively eligible. Once the estimates of income-eligible people for each eligibility cat- egory are made, the USDA method adjusts for the percentage of people in each category who are at nutritional risk. The result is the number of fully eligible people. These adjustment factors are: 95 percent for infants, 75.2 percent for children, 91.3 percent for pregnant women, 93.3 percent for nonbreastEeeding postpartum women, and 88.9 percent of breastEeeding postpartum women. The nutritional risk adjustment factors used are based on estimates of the percentages of income-eligible people in each category who are at nutritional risk. All but one of these adjustment factors were obtained from the first WIC Eligibility Study (U.S. Department of Agri- culture, 19871. The infant adjustment factor was increased from 72 to 95 percent in 1991 on the basis of more recent independent estimates of nutri- tional risk among infants, which showed that 95 percent of infants met a . . . . . . . nutritional risk criterion. 5This adjustment is based on data from the second WIC Eligibility Study (called WES II). 6This accounts for only 0.9 percent of all estimated eligible infants and 1.1 percent of all estimated eligible children in 1998.

OCR for page 18
32 ESTIMATING ELIGIBILII~YANDPARTICIPATIONFOR THE WICPROGRAM Estimation of Full-Funding Participation Not all of those who are eligible for WIC will participate in the pro- gram. Once the estimates of the numbers of fully eligible people in each category are made, the numbers are adjusted downward to account for this. Past practice in making this adjustment has assumed that participation rates for WIC will mirror participation rates for the Food Stamp Program for children ages 0 through 4 years. Until recently, the food stamp participa- tion rates from the late 1980s were used to adjust the eligibility estimates, meaning that roughly 80 percent of eligible people were estimated to par- ticipate. Beginning with the 1995 estimates, USDA has not used the food stamp participation rate assumption, but rather has made budget requests with a goal of serving 7.5 million participants. To assess the reasonableness of this approach, the panel, in Chapter 8, estimates post-hoc participation rates for the WIC program but does not try to model participation deci- sions. Forecasting WIC Eligibility and Participation WIC eligibility and participation estimates based on a given year's sur- vey data are used to forecast eligibility for the budget for a future fiscal year. The length of time between the gathering of these survey data and the period for which the budget is being developed the forecasting period- can span as many as four years. For example, development of the budget for FY 2003 began in spring 2001. At that time, the latest March CPS data available were derived from the March 2000 CPS. This survey collects in- come data for the previous calendar year, which was 1999. Thus, 1999 data are used to inform the budgetary request for FY 2003. The estimates of eligibility and participation are used to inform the president's budget, which is submitted to Congress in the fall or winter. March CPS data are typically released in the fall of the year of the survey. Thus, it is possible that by the time the president's FY 2003 budget was finalized, the March 2001 CPS data may have become available and used to update the estimates made with the 2000 data. But even then, these data would be used to predict eligibility and participation three years in advance. Current methods for estimating eligibility and participation for WIC do not make adjustments for changes in the population or economy that might cause errors in these forecasts. This is equivalent to assuming no change in any of these factors.

OCR for page 18
OVERVIEW OF WIC AND THE CURRENTESTIMATIONMETHODOLOGY 33 SUMMARY Each year USDA develops estimates of the number of people eligible and likely to participate in the WIC program. These estimates are used to inform budget requests and to evaluate the effectiveness of the program. Current methods for making these estimates use the March CPS to esti- mate the number of income-eligible infants and children. The estimates of infants, along with several adjustment factors, are used to estimate the num- ber of income-eligible pregnant and postpartum women. To get the num- ber fully eligible in each category, adjustments for the prevalence of nutri- tional risk among the income-eligible population are made. Finally, USDA assumes that participation in WIC among those who are eligible is similar to the participation rate for the food stamp program.