Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
10 Summary This report has reviewed the current methodology used by USDA to estimate, on a national basis, eligibility and participation for the WIC pro- gram. Our review first described the two basic purposes for which the esti- mates are being made: to develop budget estimates for the upcoming fiscal years and to gauge how well the program is reaching the population it intends to reach, that is, program coverage. To a lesser extent, the estimates are also used to estimate how program changes affect eligibility and partici- pation and how external influences, such as economic conditions, affect eligibility and participation. The panel has argued that if the purpose of the estimates is to understand program coverage and evaluate the effectiveness of program characteristics, then it is essential to estimate the number of people eligible for WIC and the Percentage of those who may narticinate in WIC. 1 ~ ESTIMATING ELIGIBILITY , . 1 The panel concludes that current estimation methods result in a sub- stantial underestimate of eligibility because monthly income and adjunc- tive eligibility are not adequately addressed. Panel estimates show that a significantly greater number of people would be determined eligible for WIC if a monthly income measure were used instead of an annual income measure. Using Survey of Income and Program Participation (SIPP) data, 159
160 ESTIMATING ELIGIBILI>~D PAR TICIPATION FOR THE ~CPROGR~ the use ot monthly income (and accounting for WIC certification periods and adjunctive eligibility through reported enrollment in food stamps, Medicaid, or Temporary Assistance for Needy Families TANF) resulted in a 46 and 54 percent increase in the number of income-eligible infants in 1997 and 1998, respectively, and a 34 and 36 percent increase in 1997 and 1998 for children, compared with the current USDA estimates. The panel also determined that current methods used to estimate eli- gibility do not adequately account for adjunctive eligibility. With expan- sions in the Medicaid program that raised the income limit for eligibility well over 185 percent of federal poverty guidelines in many states, some people with annual incomes over 185 percent of poverty could be eligible for WIC because they were enrolled in Medicaid, but they would not be counted as such in the eligibility estimates. Using SIPP data and reported participation in Medicaid and other public assistance programs that con- fer adjunctive eligibility (TANF and food stamps), the panel estimates that an additional 18 percent of infants are eligible for WIC and an additional 10 percent of children are eligible compared with estimates based on the current USDA methodology, which uses an annual income measure. The panel concludes that current estimation methods result in an underesti- mate of eligibility because monthly income and adjunctive eligibility are not addressed. Options for Estimating Eligibility The panel proposes two options for USDA to consider using in order to estimate income eligibility for WIC. One option maintains the March Current Population Survey (CPS) as the base survey from which eligibility estimates are derived, but it uses multipliers to make appropriate adjust- ments. The second option relies on SIPP, which has a longitudinal design and collects monthly data on income and program participation. An out- line of these two options is given below: CPS with multipliers: Use annual income to estimate the number of in- fants and children eligible for WIC. Count those who report receiving Medicaid, food stamps, or TANF as adjunctively eligible. Use a con- stant multiplier to increase the estimates to account for monthly in- come and certification practice and underreporting of participation in means-tested programs in the CPS. This multiplier could be based on the estimates from the Transfer Income Microsimulation (TRIM)
SUMMARY 161 model (a multiplier of 1.20 for infants and 1.05 for children), or on a SIPP-based multiplier. These core estimates for infants and children would then be used to estimate the number of eligible pregnant and postpartum women. SIPP option: Use monthly income and account for WIC certification prac- tice to estimate the number of income-eligible infants and children. Count those who report receiving Medicaid, food stamps, or TANF as adjunctively eligible. Directly estimate the number of eligible pregnant and postpartum women. Both of these options have benefits and limitations. The major limitation of the SIPP data is that their public release is not as regular or as quick as the CPS. The major limitation of the CPS-based option is that month-to- month income variability that differs from the panel's estimate would in- troduce error in the constant multiplier proposed to correct for the use of annual income. Either option is better than the current method because they both account for income variation across the year and adjunctive eligi- bility. Accounting for monthly income and adjunctive eligibility are high pri- orities for improving CPS-based estimates of eligibility. The panel also makes recommendations about the methods used to infer the number of income-eligible pregnant and postpartum women from the number of in- come-eligible infants, to estimate breastEeeding rates among postpartum women, and to estimate the prevalence of nutritional risk. These recom- mendations can be summarized as follows: 1. To correct for CPS undercounts of infants and overcounts of chil- dren, use adjusted weights (Chapter 41. 2. To estimate the number of income-eligible postpartum women from CPS-based estimates (both breastEeeding and nonbreast- feeding), continue to use the current adjustment factor of 0.9844 to account for multiple births and infant and fetal deaths (Chapter 61. 3. To obtain the number of income-eligible pregnant women, apply an adjustment of 0.533 (instead of the 0.75 factor) to the number of income-eligible infants to account for income variability during pregnancy (Chapter 61. 4. Use more recent data to estimate breastEeeding rates and duration among income-eligible postpartum women less than 12 months
162 ESTIMATING ELIGIBILI>~D PAR TICIPATION FOR THE ~CPROGR~ postpartum. Apply them to the estimates of income-eligible post- partum women to determine the number breastfeeding and nonbreastSeeding (Chapter 61. 5. The percentage of the income-eligible population also at nutri- tional risk is very close to 100 percent. Methods used to screen for nutritional risk are not accurate enough to precisely identify the small percentage of those who are income eligible but not at nutri- tional risk. As a result, the panel concludes that a nutritional risk screen is not useful for determining eligibility. If the USDA drops this aspect of eligibility determination, no adjustment for the prevalence of nutritional risk is needed to estimate eligibility (Chapter 71. The first three recommendations apply to the CPS-based option for estimating eligibility. The last two recommendations apply to both the CPS- based and SIPP-based options for estimating eligibility. ESTIMATING FULL-FUNDING PARTICIPATION The full-funding participation rate (FFPR) is the percentage of indi- viduals eligible for WIC who choose to participate, if funds are sufficient to serve them. The panel makes the claim and illustrates that changes in pro- gram administrative practice or changes in program outreach can increase or decrease the number of WIC participants. The FFPR is a level of partici- pation that policy makers could set a goal to achieve. Based on the premise that the full-funding level of WIC participation is a policy goal, the panel recommends a strategy to predict the number of participants each year for the purpose of making budget estimates. The strategy the panel recom- mends depends on whether the goal FFPR has been achieved or not. If the FFPR has been achieved, then the method to estimate participation levels is simply to use last year's participation levels. However, if the FFPR has not been achieved, then the method multiplies the desired FFPR by the esti- mated number of eligible persons in the eligibility category. The assessment of whether each category of eligible persons has met the desired rate of participation should be made each year. Furthermore, because it is likely that not all eligibility categories will meet the full funding level, separate assessments and estimation strategies should be made for each eligibility category.
