7

Evaluation of Cost

In this chapter, the methods used to generate cost estimates for both the current and revised food packages and to ensure the cost neutrality of the revised packages are described. As explained in Chapter 1, ensuring cost neutrality was at the crux of the committee’s final decisions for food package revisions. As part of the committee’s strategy to emphasize consideration of the value of the food packages for the mother–infant dyad (described in Chapter 6), this chapter also presents comparisons of the market value of the current and revised food packages for the three types of mother–infant pairs: fully breastfeeding, partially breastfeeding, and fully formula-feeding. The sensitivity of the results to several of the assumptions used to estimate cost are tested and described in Chapter 8. In Chapter 10 (an abridged version of Appendix U), the projected effects of the food package changes on overall costs to the program are evaluated.

METHODS USED TO ESTIMATE COSTS

Overview

As described in Chapter 1 (see Figure 1-1), the framework for revising the food packages required the committee to consider several sets of food packages in an iterative fashion. In each iteration, adjustments were made to ensure that the revised set of packages met the criteria outlined in Chapter 1, Box 1-4, while also being cost-neutral.

To evaluate cost-neutrality, the committee created spreadsheets for each food package that detailed the costs of each food in the package as

well as costs of the total package for both actual redemption (i.e., based on available redemption data) and full redemption. These spreadsheets also detailed the nutrient composition of each food in each package.¹ The spreadsheets were linked to participation numbers so the weighted average per-participant food package costs could be estimated for both the revised and current food packages. The difference in the weighted-average, per-participant cost of a single food package between the committee’s final, proposed set of revised food packages and the current set of food packages was required to be cost neutral, specifically no greater or less than $0.10.

As explained in more detail in the sections that follow, the weighted-average, per-participant costs of the food packages were based on the number and distribution of the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) participants across packages and estimated cost of each food package. The methods used to evaluate cost-neutrality were similar to those described in the previous WIC report (IOM, 2006). However, this committee benefited from the availability of cost data specific to WIC foods as well as redemption data. Estimated food package costs were calculated using these prices and redemption rates.

General Data Considerations

The base year for the comparison of cost neutrality between the sets of revised and current food packages was fiscal year (FY) 2015, the year for which the most complete set of participant data was available. The quantities for food items were based on the maximum allowances specified for the current and revised packages (for current food packages, see Chapter 1, for the revised food packages, see Chapter 6). There was no single data source for any component used in the committee’s cost-neutrality analyses (i.e., participation, prices, composition of the WIC food categories for which there are multiple options, and redemption). When possible, the committee used data that were considered the most representative of WIC national trends (i.e., the U.S. Department of Agriculture’s Food and Nutrition Service [USDA-FNS] price and redemption dataset [further described below] or data from the WIC Participant and Program Characteristics Series), filling in gaps with supplementary sources (i.e., state-specific data) as needed and as available to the committee.

__________________

¹ The nutrient composition of the current packages is presented in Chapter 3, Tables 3-11 through 3-13. The food group and nutrient content of the current compared to the revised packages are presented in Appendix T.

Estimating the WIC Participation Distribution Across Food Packages

To evaluate the cost-neutrality of the proposed set of revised food packages, the committee estimated the food costs to the WIC program based on the estimated costs of each food package and the number and distribution of participants. WIC participant distributions across food packages are presented in Table 7-1. These distributions were based on two sources: (1) WIC Participant and Program Characteristics 2014: Food Package Report (USDA/FNS, 2016a) representative of participation in April 2014, and (2) average participation by category for FY2015 from administrative data (USDA/FNS, 2016b). Briefly, the Food Package Report provides the distribution of the 27 food package options across six participant categories for all participants certified to receive benefits in the month of April of the assessment year (i.e., for this analysis, 2014). This data source includes all participants certified to receive benefits in April 2014, regardless of whether they were issued or claimed their benefits. As a result, total participation in the Food Package Report is higher than actual participation. The distributions reported in the Food Package Report were applied to the administrative data,² a report of average monthly participation for the entire WIC program for FY2015 (USDA/FNS, 2016c). The regulatory impact analysis (see Appendix U) includes a detailed explanation of how participation for each food package was determined.

For the revised set of packages, it was assumed that neither the revisions nor allowed substitutions had any effect on participation, with one exception. As a result of its proposed option for an infant to be partially breastfed in the first 30 days, the committee anticipates that 5 percent of the formula-fed infants and 5 percent of women who were postpartum for 6 months or less but not breastfeeding were shifted to their respective partially breastfeeding categories.³ Additionally, in the current food packages, postpartum women who are not breastfeeding no longer receive WIC benefits after 6 months (food package VI). To account for a predicted shift of the latter group to partially breastfeeding (for whom benefits extend for 1 year postpartum), 5 percent of these nonbreastfeeding women were shifted back into the program to the partially breastfeeding package (V-B). The 5-percent shift was selected based on data presented in USDA/FNS (2011) that indicated that the 2009 food package (in which women to choose between formula feeding or fully breastfeeding) resulted in an approximately 7- to 11-percent shift of women out of the partially breastfeeding food package. The shifts were largely to the fully formula-fed package (as discussed

__________________

² Administrative data includes all state agencies, Indian Tribal Organizations, and territories.

³ An explanation of the committee’s recommendation to allow up to 364 fluid ounces of formula in the first month (i.e., the partial breastfeeding option in the infant’s first 30 days) is provided in Chapter 6.

TABLE 7-1 Participant Distributions Used to Estimate Costs for Current and Revised Food Packages

NOTES: Participation information derived from the WIC Participant and Program Characteristics 2014: Food Package Report (USDA/FNS, 2016a) and WIC administrative data (USDA/FNS, 2016b). A detailed description of the participation distributions is available in Chapter 10.

^a The participant distribution for the current and revised food packages are the same, with one exception. For the revised food packages, 5 percent of formula-fed infants and postpartum, nonbreastfeeding women were shifted to the partially breastfeeding packages. Five percent of postpartum, nonbreastfeeding women who were no longer receiving a package because their infants were 6 months or older were shifted back onto the program, to receive a partially breastfeeding food package.

^b This group includes both nonbreastfeeding, postpartum women and minimally breastfeeding women who are less than 6 months postpartum.

^c Food package V in the current set of food packages is issued to both pregnant and partially breastfeeding women.

^d In the revised set of packages, food package V is separated into V-A for pregnant women and V-B for partially breastfeeding women.

SOURCES: USDA/FNS, 2016a,b.

further in Chapter 2). The 5-percent value was therefore considered a conservative estimate for the number of participants who would shift back to partial breastfeeding.

Estimating WIC Food Costs

In contrast to the IOM (2006) report, this committee had the benefit of cost data that were specific to WIC foods. USDA-FNS provided the committee with 12 months (August 2013 through July 2014) of price and redemption data from a convenience sample of six WIC state agencies, representing five of the seven regions of the country (hereafter referenced as “FNS redemption dataset”). The states were diverse in terms of size and did not include Indian Tribal Organizations or territories.^4,5 (A discussion of other sources of redemption data used by the committee is provided in the “Use of Redemption Rates” section.) The identity of the agencies was not known to the committee. Data from each state agency included the main category of food (e.g., “legume”); subcategory of food (e.g., “canned beans”); size and measure of the container (e.g., 16 ounces); number of containers redeemed in the month; average price for the container; and the total amount paid by WIC for that specific item in the given month. Some states provided specific food subcategories (e.g., “soft corn tortillas”) while others provided broad categorizations (e.g., “whole grains, all types”). USDA-FNS summed all available data across the six states and all months. These composite data were used to calculate average price per unit (e.g., price per ounce), as redeemed by WIC participants.

In three cases (i.e., infant formula, yogurt, and new grain sizes), prices were not available in the FNS redemption dataset. For these, marketing data from the 2014 Information Resources, Inc. (IRI) Consumer Network Database were used instead. Prices from all sources were adjusted to represent FY2015.⁶ Adjusted prices are presented in Appendix R, Table R-1. Additional details about infant formula cost estimates are provided below.

The committee had no information on either the costs of medical foods or portions of these costs paid by WIC. Thus, the food package III costs presented here account only for foods in the corresponding age and physiological state food package. They do not account for medical products.

__________________

⁴ These data are available in the public access file for this study (Email: paro@nas.edu).

⁵ As characterized by USDA-FNS.

⁶ The two sources of price data encompassed different timeframes. The FNS redemption dataset represented August 2013 through July 2014 and the IRI prices represented calendar year 2014. To create a common base year, prices were inflated to FY2015 prices using Bureau of Labor Statistics Consumer Price Index (CPI) adjustments (see BLS, 2016). To accomplish this, the average item-specific CPI for FY2015 was divided by the average CPI for the timeframe encompassed by the available unit price. Additional detail is available in Appendix R, Table R-1.

Costs for the revised food package III were estimated based on 2015 prices and the numbers and categories of participants who were issued that package in 2015.

Infant Formula Cost and Rebate Assumptions

The retail price of each form of formula (powdered, ready-to-feed, and concentrate) was determined by using a 2014 IRI price, adjusted to reflect a FY2015 price. Per-unit prices for each form of formula were then weighted by the volume of each form issued in the WIC program, as reported by USDA-FNS (USDA/FNS, 2013a), to determine a composite price per unit. Starting with this composite price per unit, the post-rebate cost of formula was determined.

To maintain lower costs, states are required to negotiate rebate contracts with infant formula companies. According to the 2010 WIC Cost of Foods report (USDA/FNS, 2013b), the post-rebate cost for formula was $927 million of $2,615 million (retail). This is equivalent to a rebate of 65 percent. Because this number was from 2010, the committee evaluated the stability of rebates between 2010 and 2015.⁷ With the exception of 2011, rebates between 2010 and 2015 were relatively stable. Therefore, the committee multiplied the composite price per unit by a factor of approximately 0.35. All assumptions applied to estimate the costs of formula are presented in Appendix R, Table R-2.

Creating Composites for the WIC Food Categories

Similar to the approach used in the IOM (2006) report, the estimated costs of (and nutrients provided by) the current and revised food packages were based on the unit prices for each food (or food category, i.e., “whole grains”), adjusted to FY2015, and the amounts of each food in each food package. For each of the food categories, the price was a weighted average of several food items, estimated using a series of assumptions. The specific assumptions applied for cost are presented in Appendix R, Tables R-1 and R-2. As mentioned previously, the information used to weight the proportions of particular foods for each WIC food category came from different sources, including the USDA-FNS redemption dataset, the Food Package Report series, and/or data from individual states. In this way, available information was used to create food package cost and nutrient profiles that represented WIC redemptions as closely as possible (see Appendix R, Tables R-3 and R-4). As an example, the USDA-FNS redemption dataset

__________________

⁷ Rebates were reported as “rebates billed” before 2013 and “rebates received” after 2013. These were considered equivalent for the purpose of estimating rebate changes over time.

included prices for both canned and dried beans, as well as peanut butter. To estimate the cost of the “legume” WIC food category, the average cost per unit for each of these options was calculated. For each food package, a composite was developed depending on whether participants are issued both legumes and peanut butter each month or only one of these options. In this example, the ratio of canned to dry legumes typically redeemed was ascertained from the USDA-FNS redemption dataset. For food packages that required participants to choose between beans or peanut butter, a proportion of 50 percent of each was assumed to be issued.

For all foods, except those acquired with the cash value voucher (CVV), the committee’s cost (and nutrient) composites accounted only for foods for which the USDA-FNS redemption dataset indicated that redemption was at least 1 percent. For example, tofu was not included in the “milk” WIC food category composite because it represented less than 1 percent of redemptions. Therefore, any price (or nutrient) difference of tofu was considered to have no substantive effect on the composite.

For vegetables and fruits, only the five most commonly redeemed vegetables and six most commonly redeemed fruits were considered in the composite. Redemption data for vegetables and fruit were obtained from Massachusetts, Texas, and Wyoming.⁸ They were combined with USDA’s Economic Research Service (USDA-ERS) price data to determine the cost of a cup-equivalent serving of a composite vegetable and composite fruit (USDA/ERS, 2013). These composite vegetable and composite fruit values were then weighted by the ratio of vegetables to fruits typically purchased with CVV dollars, as indicated by redemption data. This process resulted in one price per cup-equivalent serving representative of vegetables and fruits commonly redeemed by WIC participants.⁹

Use of Redemption Rates

In contrast to the IOM (2006) report, the committee also benefited from data on redemption rates.¹⁰ The USDA-FNS redemption dataset was

__________________

⁸ Information provided from states is available in the public access file for this study (Email: paro@nas.edu).

⁹ The committee assumed that the composite of vegetables and fruit would not change when the CVV is expanded, as the committee had no information to assess quantitatively how the composite might change at higher CVV amounts. To the extent that families shift their choices of vegetables and fruits with a higher CVV as their preference for fruits is progressively satisfied, this might lead to different ratios of fruits to vegetables and might affect nutrients and food group delivery of the food packages.

¹⁰ The overall redemption rate refers to the percentage of the maximum allowance prescribed to WIC participants that is actually obtained by recipients. A full redemption means the person redeemed their entire prescription. A partial redemption means they redeemed some but not all. Some participants redeem none of their prescribed benefits.

the primary source of these data.¹¹ For food categories without redemption data in the USDA-FNS redemption dataset, the committee applied redemption values provided by a few individual states over the course of the study. Redemption values published in the Altarum electronic benefit transfer (EBT) report were also considered (USDA/ERS, 2014a). For the revised food packages, the committee calculated a revised redemption rate based on equations that considered changes to food amounts, the distribution of redemption (e.g., the degree of nonredemption, partial redemption, or full redemption) for specific foods from the Altarum report, and predicted behavioral changes based on reduced or increased options. A detailed description of the method for estimating redemption rates used in this report is provided in Appendix R.

Cost of Substitutions

The revised set of food packages include the same substitutions that are permitted in the current food packages, plus some additional options. In addition to covering all the foods in the basic WIC food categories (formula, infant foods, milk, cheese, peanut butter, beans, whole wheat bread [grains], eggs, fish, and fruits and vegetables), the committee’s cost evaluations also covered substitutions.

Substitution costs were based on data on the substitution rate, when available (see Appendix R, Tables R-3 and R-4). For example, generation of cost (and nutrient) information for the low-fat milk composite was based on ratios of nonfat milk, 1% milk, and soymilk that reflect redemption patterns. The same is true for beans: The composite cost (and nutrient) values account for the proportions of canned to dry beans based on available redemption data. When such data were not available, the substitution rate was based on a likely preference. For example, because there was no information on the amount of milk that is redeemed as cheese or yogurt, the committee assumed full substitution.¹² Inasmuch as this full substitution is of greater monetary value to the recipient (that is, 1 pound of cheese or 1 quart of yogurt are more expensive than the milk that is replaced) it resulted in a conservative cost estimate (the highest possible cost) for both

__________________

¹¹ To keep the state agencies anonymous, USDA-FNS inputted average monthly participation by participant category for each state into a spreadsheet containing the redemption equations created by the committee, and returned the overall unweighted average redemption across the state agencies per food package item. Through this process, USDA-FNS identified one of the six states as a clear outlier, and removed it from the averages (personal communication, K. Castellanos-Brown, USDA-FNS, June 22, 2016). As such, redemption estimates represent five of the six state agencies included in the FNS redemption dataset.

¹² The substitution options differ by food package. A specific ratio of the possible full substitution options was assigned to participants for each food package, as detailed in Appendix R.

the current and revised food packages. For example, as shown in Table 7-2, redeeming 1 quart of whole milk yogurt in place of 1 quart of whole milk costs an additional $2.36.

The effects of other selected substitutions are also presented in Table 7-2. For example, redeeming four 16-ounce cans of beans instead of 1 pound of dry beans costs an additional $2.55. Teff and buckwheat both cost substantially more than whole-wheat bread. Although these two new substitutions may not align with cost-containment strategies of all states, states with WIC-participating populations that use these grains as a staple food have the option to make other modifications to their state food lists to accommodate these choices. New whole grain options were not included in the revised food package grain composite because there is no available information upon which to base an assumption about redemption.

Estimating Per-Participant Program Costs for Food

To estimate weighted-average, per-participant food package costs for the current and revised food packages, the estimated number of participants who received each package in 2015 (see Table 7-1) was multiplied by the estimated cost of the respective package (see Table 7-3) and then divided by the total number of participants. Values for the final, cost-neutral set of food packages are presented in the bottom row of Table 7-3.

RESULTS AND DISCUSSION: PROGRAM COSTS

Estimates of monthly, postrebate, per-participant costs for both the current and revised food packages are presented in Table 7-3. Although prerebate costs are an indication of the market value of the food packages to the recipients (e.g., prerebate costs for mother–infant dyad packages only are presented in Table 7-5), post-rebate costs reflect USDA-FNS’s costs for the food packages. There are two important features of Table 7-3. First, for the infant packages, subtotals are presented for each of the various food package I and II subpackages (formula-fed, partially breastfed, and fully breastfed). These values were then weighted by the share of participants who received each subpackage to calculate weighted average monthly cost for both food packages I and II. The second feature of note is that the values presented in columns 1 and 2 are based on calculated redemption rates. The rates used for column 1 were calculated from average amounts redeemed in the current packages; the rates used in column 2 represent the amounts that are projected to be redeemed in the revised packages. Columns 3 and 4 present costs for the fully redeemed packages.

Average weighted per-participant costs of the food packages are provided in the bottom row. The committee’s estimate of the per-participant

cost of the current food packages is $37.27 (i.e., column 1). The USDA-FNS preliminary estimate for 2015 based on reimbursement to states is $43.37¹³ (USDA/FNS, 2016b). USDA-FNS calculated the value based on state-reported, per-participant costs. In contrast, the committee’s evaluation of WIC food package costs was based primarily on the USDA-FNS redemption dataset, which was the best available for this purpose at the time of this analysis. Potential reasons for differences between the USDA-FNS value and the committee’s value are discussed below.

Possible Reasons for Differences in Average Package Cost Compared to the USDA National Average

Possible reasons for the difference between the USDA-FNS weighted average per-participant cost and the committee’s estimate are listed here:

• Inasmuch as price and redemption data used were based primarily on information from six and five unidentified states, respectively, the committee’s analysis is not nationally representative. USDA-FNS data indicate that average 2015 food package costs vary among states, ranging between $28.89 (Texas) and $78.62 (Guam), depending on costs of food and distribution of participants by package (USDA/FNS, 2016b). This very wide range suggests that estimates generated using the USDA-FNS redemption dataset could vary widely depending on which six states were included.
• The USDA-FNS redemption dataset may have been sourced from EBT states. The weighted average food package cost for EBT states (not including Indian Tribal Organizations) in 2015 was $34.98, compared to $46.23 across all other (non-EBT) states, suggesting that EBT states have considerably lower food package costs compared to other states.
• The USDA-FNS redemption dataset could have come from states with lower cost of WIC foods compared to average national costs of WIC foods.
• Redemption rates in the five states may be lower than the national average generally.¹⁴

__________________

¹³ The average monthly benefit equals total annual food cost divided by average monthly participation, divided by 12.

¹⁴ Only five states were used to generate average redemption rates because the data from one state resulted in very low redemption values that fell outside of what was considered the normal range.

TABLE 7-2 Estimated Costs of Basic Foods, Selected Substitutions, and Net Cost Changes Resulting from Selected Substitutions in WIC Food Packages

NOTES: CCV = cash value voucher; veg/fruit = vegetables and fruits.

* Whole wheat bread and whole wheat bread substitutions that are expressed as a range were calculated by multiplying the average 2014 IRI price per ounce for all sizes listed by 24 ounces, assuming 24 ounces will be the most commonly redeemed size. This may be an overestimation of cost for some substitutions, for instance corn tortillas, for which 16 ounces is a commonly available size, however, the average IRI price is automatically weighted by the most common size purchased. Brown rice, barley, and wheat tortillas were not evaluated as substitutions for whole wheat bread as they comprise less than 1 percent of substitutions redeemed, based on redemption data made available to the committee. Costs were from FNS redemption data, CPI-adjusted to 2015, with the exception of yogurt and whole wheat bread, which were from IRI 2014.

SOURCES: Personal communication, K. Castellanos-Brown, USDA-FNS, April 7, 2016, and June 30, 2016; BLS, 2016; 2014 IRI Consumer Network Database.

TABLE 7-3 Comparison of Estimated Costs of Current and Revised Food Packages

NOTES: BF = fully breastfed; BF/FF = partially breastfed; FF = formula fed. See Appendix R, Table R-5, for determination of redemption percentages. See Appendix R, Tables R-1 through R-3, for assumptions applied to develop the current and revised food package cost profiles. Prices were from USDA-FNS redemption data, CPI-adjusted to 2015, with the exception of infant formula, revised package whole wheat bread and alternatives, and yogurt, which were from the 2014 IRI Consumer Network Database.

^a Participants of these ages receiving food package III were folded into the appropriate age-and physiological state category to capture issued foods.

^b Food package V in the current set of packages is issued to both pregnant and partially breastfeeding women.

^c In the revised set of packages, food package V is separated into V-A for pregnant women, and V-B for partially breastfeeding women.

SOURCES: Personal communication, K. Castellanos-Brown, USDA-FNS, April 7, 2016, and June 30, 2016; USDA/FNS, 2014a; BLS, 2016; USDA/FNS, 2016a; 2014 IRI Consumer Network Database.

• Redemption rates in the five states may be lower for more expensive foods or for more expensive packages compared to national average redemption rates.
• The committee’s analysis assumes that all substitutions in a category are redeemed at the same rate as the default WIC food (e.g., that cheese, when substituted for milk, is redeemed at the same rate as milk). It may be that more expensive substitution options are redeemed at higher rates.
• As explained earlier, because there were no available data, the committee’s analysis does not include costs for medical foods. Although recipients of food package III comprise approximately only 6 percent of participants (USDA/FNS, 2016a), excluding costs of exempt formula or WIC-eligible nutritionals (formerly “WIC medical foods”) may result in lower average per-participant costs. For example, if the weighted-average cost of medical formula provided to the 6 percent of WIC participants who receive food package III is $300 per month (no rebate), inclusion in the cost analysis would increase the estimated weighted-average, per-participant cost of the food package to approximately $46.
• IRI data indicate that the average costs for foods range widely. The cost for each food depends on many factors (e.g., brand, size, store where purchased, how long the individual has shopped for them) (USDA/ERS, 2005, 2014b, 2015, 2016).

Given the current data limitations, it is impossible to speculate about any possible bias in the committee’s estimate of weighted-average, per-participant costs of the food packages. As the EBT system is introduced nationally, future committees should have much more nationally representative data. Although its final estimate of the weighted-average, per-participant cost for the current food package is different than that of USDA-FNS, the committee’s methods for estimating costs were applied consistently and in the same manner across all food packages to determine cost-neutrality.

Ensuring Cost Neutrality

The committee was tasked with ensuring that the revised food packages are cost-neutral. For the purposes of this review, cost-neutral means that the weighted-average, per-participant cost of each of the food packages I through VII in the revised set of food packages falls within $0.10 of the weighted-average, per-participant cost of the current set of food packages. As presented in Table 7-3, the weighted-average, per-participant cost of the revised set of food packages is $37.32, which is $0.05 higher than the committee’s estimate of the weighted-average, per-participant cost of the

current package cost. The revised set of food packages therefore meets the task requirement for cost neutrality.

Cost neutrality was achieved by offsetting increases in the costs of particular food packages with decreases in the costs of other food packages. As a result of the cost-neutral requirement, the committee was restricted in the degree to which the CVV could be increased and in the amount of fish that could be added to food packages. Inasmuch as food package cost is weighted by participation, the amount of foods added to the food packages for children had larger effects on the weighted-average food package cost compared to foods added to packages for women.

COMPARING COST INCENTIVES FOR BREASTFEEDING

In accordance with the component of the committee’s decision to view the value of the food packages from the perspective of the mother–infant dyad (as described in Chapter 6), both postrebate costs (see Table 7-4) and prerebate market values (see Table 7-5) of the three types of mother–infant pairs (formula-feeding, partially breastfeeding, fully formula feeding) were compared. The postrebate costs presented are those as actually redeemed (i.e., based on available redemption data) or projected to be redeemed. The prerebate market values, in contrast, assume full redemption of the food packages.

The annualized postrebate, as-redeemed cost of the revised fully breastfeeding dyad packages is increased by $20 compared to the current fully breastfeeding dyad packages. For the revised partially breastfeeding dyad packages, this difference is $94. Most of these differences in the costs of the dyad packages come from revisions to the food packages for the breastfeeding mothers. The package for fully breastfeeding women is projected to be redeemed at an annual value $92 higher and for the partially breastfeeding mother $87 higher compared to the current packages, respectively. These increases in the value of the fully and partially breastfeeding mothers’ packages reflect the committee’s objective to support exclusive breastfeeding as well as breastfeeding of any intensity (see Table 7-4). It is noteworthy that, compared to the current packages, the postrebate market value of the food packages for the breastfeeding dyads are revised to be closer to the value of the packages for the fully formula-feeding dyad. Currently, the redeemed values of the packages for the fully formula-feeding dyad is $39 to $75 more than that of the breastfeeding dyads. In the revision, the redeemed values of the packages for the fully formula-feeding dyad range from $46 less to $64 more than that of the partially and fully breastfeeding dyads, respectively.

The annualized pre-rebate market value (fully redeemed) of the food packages for the fully breastfeeding dyad is $187 lower and for the partially

TABLE 7-4 Comparison of the Food Package Costs with Rebates as Redeemed, Current and Revised Food Packages for Mother–Infant Dyads

NOTES: Prices were from the USDA-FNS redemption dataset, with the exception of infant formula, yogurt, and revised package whole wheat bread and alternatives, which were from IRI 2014. All prices were adjusted to FY2015.

SOURCES: Personal communication, K. Castellanos-Brown, USDA-FNS, April 7, 2016, and June 30, 2016; BLS, 2016; 2014 IRI Consumer Network Database.

TABLE 7-5 Comparison of the Market (Prerebate) Value of Maximum Allowances (Full Redemption) for Current and Revised Food Packages for Mother–Infant Dyads

NOTES: Prices were from the USDA-FNS redemption dataset, with the exception of infant formula, yogurt, and revised package whole wheat bread and alternatives, which were from IRI 2014. All prices were adjusted to FY2015.

SOURCES: Personal communication, K. Castellanos-Brown, USDA-FNS, April 7, 2016, and June 30, 2016; BLS, 2016; 2014 IRI Consumer Network Database.

breastfeeding dyad is $42 higher in the revised compared to the current packages, respectively (see Table 7-5). However, the annualized prerebate market value of the revised food package for the fully breastfeeding mother is $55 higher and for the partially breastfeeding dyad is $67 higher than the value of the current food packages for them, respectively. Here it is striking how much higher the prerebate market values of the food packages for the formula-feeding dyads are compared to those for the breastfeeding dyads. For the current food packages, those for the fully formula-feeding dyad are valued at $843 more than those for the fully breastfeeding dyad and $476 more than those for the partially breastfeeding dyad. In the revised food packages, those for the fully formula-feeding dyad are valued at $1,008 more than those for the fully breastfeeding dyad because amounts of foods in the fully breastfed infant packages were reduced. However, the difference between value of the packages for the fully formula-feeding dyad and the partially breastfeeding dyad is smaller ($413 for the revised, $477 for the current).

As is true of all the WIC food packages, the benefit of the breastfeeding food packages extends beyond the foods provided and includes other WIC services, such as nutrition counseling, health referrals, and nutrition education. WIC-participating breastfeeding mothers may benefit from peer-counseling in particular. Communication of this benefit to participants could increase the perceived value of the breastfeeding packages.

Finally, the annualized prerebate market value of the revised formula-feeding dyad packages is slightly lower than that of the current formula-feeding dyad packages (see Table 7-5). The formula-fed packages provide approximately 100 percent of an infant’s needs, and the cost of infant formula is high, so the committee found it difficult, within cost-neutral constraints, to lower the value of the formula-feeding dyad packages without decreasing the amounts of infant formula offered.

SUMMARY

The requirement that the revised food packages be cost-neutral was at the crux of the committee’s final decisions for food package revisions. This chapter describes the methods used by the committee to ensure cost-neutrality. These methods were part of an iterative process during which the committee considered several sets of food packages. In each iteration, adjustments were made to ensure that the revised set of packages met the criteria outlined in Box 1-4 (see Chapter 1), while also being cost neutral. To evaluate cost neutrality, the committee used three steps. First, it estimated participation distributions across the packages (see Table 7-1). Next, the committee estimated food package costs for all current and revised food packages (see Table 7-3). Finally, the committee estimated

weighted-average, per-participant food package costs for both the current and revised food packages. The weighted-average, per-participant cost of the revised set of food packages is $37.32 which is $0.05 higher than the committee’s estimate of the weighted-average, per-participant cost of the current package cost ($37.27). The revised set of food packages therefore meets the task requirement for cost neutrality. The annualized costs of the revised food packages for the breastfeeding dyads, as redeemed, are slightly higher than that of the current breastfeeding dyad packages. Similarly, the annualized costs of the revised food packages for breastfeeding women, as redeemed, are higher compared to the costs of the current food packages for these women. However, the redeemed value of the formula-feeding dyad packages remains higher than either of the breastfeeding packages because of the cost-neutral constraint and because the packages for younger formula-fed infants provide approximately 100 percent of needs.

REFERENCES

BLS (Bureau of Labor Statistics). 2016. Consumer price index—all urban consumers. Washington, DC: BLS. http://www.bls.gov/news.release/cpi.t01.htm (accessed September 30, 2016).

IOM (Institute of Medicine). 2006. WIC food packages: Time for a change. Washington, DC: The National Academies Press.

USDA/ERS (U.S. Department of Agriculture/Economic Research Service). 2005. Interstate variation in WIC food package costs: The role of food prices, caseload composition, and cost-containment practices. Washington, DC: USDA/ERS. https://www.ers.usda.gov/publications/pub-details/?pubid=46844 (accessed December 21, 2016).

USDA/ERS. 2013. Fruit and vegetable prices. Washington, DC: USDA/ERS. http://ers.usda.gov/data-products/fruit-and-vegetable-prices.aspx (accessed December 21, 2016).

USDA/ERS. 2014a. Transition to EBT in WIC: Review of impact and examination of participant redemption patterns: Final report. Washington, DC: Altarum Institute. http://altarum.org/sites/default/files/uploaded-publication-files/Altarum_Transition%20to%20WIC%20EBT_Final%20Report_071614.pdf (accessed December 21, 2016).

USDA/ERS. 2014b. Cost containment in the WIC program: Vendor peer groups and reimbursement rates. Washington, DC: USDA/ERS. https://www.ers.usda.gov/publications/pub-details/?pubid=45254 (accessed December 21, 2016).

USDA/ERS. 2015. Where do Americans usually shop for food and how do they travel to get there? Initial findings from the National Household Food Acquisition and Purchase Survey. Washington, DC: USDA/ERS. http://www.ers.usda.gov/publications/pub-details/?pubid=79791 (accessed December 21, 2016).

USDA/ERS. 2016. WIC household food purchases using WIC benefits or paying out of pocket: A case study of cold cereal purchases. Washington, DC: USDA/ERS. http://www.ers.usda.gov/publications/pub-details/?pubid=45536 (accessed August 29, 2016).

USDA/FNS (U.S. Department of Agriculture/Food and Nutrition Service). 2011. Evaluation of the birth month breastfeeding changes to the WIC food packages. Alexandria, VA: USDA/FNS. http://www.fns.usda.gov/sites/default/files/BirthMonth.pdf (accessed December 21, 2016).

USDA/FNS. 2013a. WIC participant and program characteristics 2012 final report. Alexandria, VA: USDA/FNS. http://www.fns.usda.gov/sites/default/files/WICPC2012.pdf (accessed December 21 2016).

USDA/FNS. 2013b. Fiscal year 2010: WIC food cost report. Alexandria, VA: USDA/FNS. http://www.fns.usda.gov/sites/default/files/WICFoodCost2010_0.pdf (accessed December 21, 2016).

USDA/FNS. 2014a. Special Supplemental Nutrition Program for Women, Infants and Children (WIC): Revisions in the WIC food packages. Final Rule, 7 C.F.R. § 246.

USDA/FNS. 2016a. WIC participant and program characteristics 2014 food package report. Alexandria, VA: USDA/FNS. http://www.fns.usda.gov/wic-participant-and-program-characteristics-2014-food-package-report (accessed August 30, 2016).

USDA/FNS. 2016b. WIC program. Alexandria, VA: USDA/FNS. http://www.fns.usda.gov/pd/wic-program (accessed July 13, 2016).

USDA/FNS. 2016c. Monthly data — state level participation by category and program costs. Alexandria, VA: USDA/FNS. http://www.fns.usda.gov/pd/wic-program (accessed October 3, 2016).