States Can Improve Child Care Assistance Programs Through Cost Modeling

The child care industry in the United States has been rocky, at best, since long before the COVID-19 pandemic. Scarcity of resources has led to too few child care providers and too few staff, making child care far too expensive and inaccessible for American families. Throughout the pandemic, new funding helped many states provide greater reimbursements to providers participating in the subsidy program. However, the state-determined subsidy rates for administering the Child Care and Development Block Grant (CCDBG) program are too low and are predominantly determined through a method—market rate surveys—that will never meet the real financial needs of child care providers. These surveys measure what providers charge at a single moment in time, which states then use as the basis for their subsidy reimbursement rates for up to three years.

The reliance on market rate surveys builds low wages, low benefits, and low capacity into the child care system going forward. In fact, data analysis has shown that the true cost of providing child care exceeds not only the reimbursement rates provided by subsidy programs but also the price charged to families.¹

As of fiscal year 2022, 49 states were using a methodology that sets future maximum payment rates based on past market rates charged, making them unresponsive to market or economic changes. By the final year the payment rates are used, the underlying data could represent the child care market of up to five years prior.

There is, however, an alternative to market rate surveys: cost estimation modeling. Cost models use regulatory requirements and program characteristics to quantify the typical expenses child care providers face and develop an estimate of what it truly costs to provide care.² Moving to cost modeling is a change that states could make already with the approval of the Administration for Children and Families (ACF); in fact, District of Columbia and New Mexico have already done so, and Virginia will soon join them. Public agencies and legislators should pursue using cost modeling analyses—basing payment rates on the true cost of providing high-quality child care.

State agencies responsible for executing the child care subsidy program must submit their Child Care and Development Fund (CCDF) plans every three years. These plans cover a wide range of information, including who is eligible for the program and how much a state will reimburse a participating child care provider. The plans comprise comprehensive documents that are hundreds of pages long, with numerous questions ranging from basic information such as which agency administers the program to very niche and specific questions such as where states list searchable data to find providers.

The reimbursement rates defined in CCDF plans are a critical choice that states make when administering their subsidy program. The payment rates affect how many and which providers accept the subsidy program, which in turn affects whether the low-income families receiving a subsidy have an equitable opportunity to find care.

ACF recommends that reimbursement rates be set at the 75th percentile of the market rate for child care³ to ensure that they are “sufficient to ensure equal access for eligible children”—a requirement set out in the Child Care and Development Block Grant Act of 2014, the federal program that provides funds to states in order to subsidize child care for low-income families.⁴ Finding the 75th percentile of the market rate, though, is not a simple exercise. Both the authorizing legislation and ACF require that states conduct a statistically valid and reliable market rate survey, or use another approved methodology, no more than two years prior to the submission of their CCDF plan. These statistical analyses play a crucial role in determining how each state will compensate providers who serve low-income families for a period of up to three years.

With CCDBG being the only federal program dedicated to supporting accessible, affordable, and quality child care, the reimbursement rates that states set using their CCDBG funding affect the overall child care landscape significantly. About 2 million children receive a subsidy funded by CCDBG in a given year.⁵ Providers must opt in to participating and accepting subsidies, but this does not guarantee they will receive any payment from the program. In most states, reimbursements are paid based on the child’s attendance, not enrollment—meaning the provider only receives income for a subsidy child if they provide care on a specific day.⁶ Since providers cannot anticipate whether each child will be present on a given day, they plan their staffing and other resources around enrollment and are often not able to adjust day-of based on attendance. This leads to providers incurring greater costs than the CCDBG program will reimburse—a huge problem for child care businesses that already operate with less than 1 percent profit margins.⁷

These processes make it difficult for providers to rely on CCDBG for operating costs. Because child care programs are not required to accept public subsidies, low subsidy rates that do not actually cover the cost of providing care may deter providers from participating at all. This results in few options for low-income families who do receive subsidies.

The CCDBG law requires state agencies to perform a market rate assessment as part of their effort to provide subsidy families with equal access to high-quality child care. Though states are permitted to create an alternative methodology, nearly all conduct a market rate survey, which is intended to capture what the existing price of child care is across the state so that the lead agency can determine an appropriate reimbursement rate.

Market rate surveys capture a moment in time; they tell policymakers what the price was when the provider was surveyed. Yet this method has two key flaws: First, it does not account for inevitable future changes in price; and second, in an industry as undercompensated as child care, market rate surveys capture prices that are only attainable by continuing to pay child care staff poverty wages without necessary benefits or advancement opportunities.

Additionally, survey research is complex and requires specific expertise to do well. Only a few states that conducted their new survey without a waiver produced the market rate survey through their agency, while nearly half (45 percent) contracted with academic researchers at an institution of higher education and 34 percent worked with a private consulting firm.

For the FY 2022–24 CCDF plan period, all states except New Mexico and the District of Columbia set their reimbursement rates based on a market rate survey. Due to the disruptions of the COVID-19 pandemic, ACF allowed lead agencies to request a waiver for updating their market rates prior to the submission of their plans. According to a Center for American Progress review of each state plan, 20 states received such a waiver at the time they submitted their CCDF plan. (see Methodology)

Basing reimbursement rates on market rate surveys does not adequately serve the needs of the child care sector. Market rate surveys look backward, at what costs were during a specific period of time, and lock future payments at a level that makes them unresponsive to market or economic changes. While ACF and the CCDBG Act allow for states to update their reimbursement rates more frequently than the CCDF plan length, there is no requirement that they do so. States that produced a new market rate survey for the FY 2022–24 plan period collected their data as early as 2019. So, by the third year of a CCDF plan—in this case, 2024—the market data could represent the child care market of up to five years prior.

This is a problem, as the costs for child care providers rise over time. For example, Arizona’s 2022 market rate survey stated that the existing subsidy rates based on the 2018 survey data were below the 33rd percentile of the 2022 market prices.⁸ Although Arizona’s reimbursements may have allowed equitable access in 2018 when they were initially set, the increases in the cost of providing care over time decreased the strength of the subsidy.

Kansas, as part of its analysis, found that providers “do not charge what it costs to cover the cost of child care and provide child care professionals with a living wage, as based on wages/benefits and other costs reported.”⁹ The discrepancies between the analyses of what it costs to provide care and the reimbursement rates that result from the market rate survey process indicate that it is extremely challenging to meet the financial needs of providers when continually basing future payments on the broken system of the past. To address this issue, public agencies and legislators should begin integrating forward-looking analyses, basing payment rates on the expected price of care in the future, rather than what it once was in the past.

The Center for American Progress completed a comprehensive review of the approved CCDF plans from each state and the District of Columbia for the FY 2022–24 period.¹⁰ (see Methodology) This review revealed how states currently operate within the market rate survey system, which includes vastly different approaches to market rate survey methodology and how it is translated into reimbursement rates, as well as a wide range of rates relative to the ACF recommendation.

States provide base rates that indicate the reimbursement provided prior to assessing any adjustments for quality or other special circumstances. Agencies are allowed to vary their reimbursement rates according to a variety of factors that the CCDF plan covers. All 51 plans reviewed provided different reimbursement rates depending on the age of the child being served and the setting or type of care provided (e.g., family day home or center-based). Thirty-four states adjusted rates based on geographic criteria—which included a variety of approaches, such as county-level rates, tiers or regions, and urban/rural—but 17 states provided one rate for all areas. Additionally, 35 states provided additional payments for meeting higher-quality standards, such as a quality rating and improvement system (QRIS) level, accreditation standard, or specific licensing criteria. Some state plans allowed for other adjustments to reimbursement rates, such as additional payments for nontraditional-hour care or serving a child with special needs.

Agencies are asked to provide the base payment rates for the area that serves the most children within the CCDF plan, and they must provide both the dollar amount of the reimbursement and the percentile that amount represents. Across infants, toddlers, and preschool-aged children, reimbursement rates were set between the 5th and 90th percentile of market rates, depending on the state, type of care provided, and age of child. (see Figure 1) When considering all categories of care for children under the age of 5, 15 states set base rates in their FY 2022–24 CCDF plans equal to or greater than the ACF recommendation for all ages and types of care. An additional eight states met the ACF-recommended 75th percentile in at least one combination of age and type of care. Because infant care is the most expensive care to provide due to the higher staff-to-child ratios required for the care of babies, this section provides the most detail on rates for infant care.

For center-based infant care, only 18 states indicated that they set base rates at the 75th percentile of their market rate survey, leaving 31 states below the ACF recommendation—excluding the two states who rely on cost modeling and therefore do not provide percentiles. For family child care, 21 states met the 75th percentile or higher for infants. The median percentiles used by states were 66th for center-based care and 70th for family child care.

By percentile, the highest reimbursement rate for center-based infant care was in Washington state and Louisiana, both setting rates at the 85th percentile. The lowest percentile for center-based infant care was found in Colorado, Ohio, Alaska, and Georgia—all providing reimbursement rates at the 25th percentile of their market rate. For infant family child care, three states (Washington, Louisiana, and West Virginia) all set their rates at the 85th percentile, the highest in the category. The lowest rate for infants in family child care was in Vermont, where the base rate was set at the 21st percentile.

The highest reimbursement rate for center-based infant care, by dollar amount, was $478.65 per week in Washington state—the highest dollar base rate of any category in any state plan. Louisiana, also using the 85th percentile of its state market rates, pays $178.25 per week. The difference in the dollar amount of the payment reflects the major differences in the overall market costs. For instance, Tennessee provides a similar payment amount—$178 per week for center-based infant care—but that amount reflects the 59th percentile of the state market.

The lowest base rate for infants in a center-based setting, by dollar amount, was $110.77 per week in Mississippi, which the state indicated was the 75th percentile of its 2017 market rates, as Mississippi received a waiver for its updated survey. The four states at the lowest percentile (Colorado, 25th percentile or $319.40; Ohio, 25th percentile or $229.32; Alaska, 25th percentile or $209.30; and Georgia, 25th percentile or $150.00) all set their reimbursement rates at a higher dollar amount than Mississippi, but because their markets are much more expensive overall, their rates provide fewer options for families.

The impact of the pandemic on the child care market is also reflected in the FY 2022–24 plans: 20 states received waivers on their market rate survey and 26 received a waiver for their narrow cost analysis. The CCDF plan asked, in Section 4.2.3(b), whether the state believed the market rate data they had gathered adequately reflected the child care market at the time they were submitting the plan, given the impact of the pandemic on the market; 21 states responded, “no.” The majority of these states (17) were those that had a waiver and were not able to use an updated market rate survey; however, there were also four states that had conducted new surveys.

On top of the ways COVID-19 affected the market and CCDF process, approximately half of all states used a portion of their American Rescue Plan (ARP) child care discretionary funds to increase or supplement their reimbursement rates.¹¹ Given that the majority of states are not yet meeting the ACF recommendation for reimbursement rates, this represents a critical investment for subsidy programs that will run out when the one-time emergency funding is exhausted. Even as the cost of child care continues to increase, many states will likely end up reducing their reimbursement rates if they do not replace the ARP funds.

Cost modeling is an implementation choice that all states could make without the need for Congress to act. Cost modeling does not address the underlying shortage of funding or the need for greater public investment, but it does provide a path for improving the distribution of funds and quantifying the shortfall between existing and needed funding. As congressional leaders indicate a desire to make progress on child care and as some states consider their own investments, it is important that the system for using CCDF funding is as strong as possible. That means ending the reliance on a moment-in-time analysis that does not consider wages, benefits, or prevailing economic factors.

States can also become more responsive to immediate needs or shocks that the sector may face. For example, New Mexico’s CCDF plan states: “A key benefit of the alternative methodology approach with the cost estimation model is that the cost model is able to be modified to account for additional costs that may arise due to the ongoing COVID-19 pandemic, such as the cost of substitute time and paid sick leave due to quarantine and testing requirements, personal protective equipment and cleaning costs, or lower group size requirements.”¹² These types of adjustments are difficult to account for in the existing market rate survey environment without undertaking a new survey.

Cost modeling is not foreign to most state lead agencies. As part of the CCDF plan, states are required to complete a narrow cost analysis on top of their market rate survey. Through this analysis, they must be able to answer questions about how quality of care relates to the cost of providing care, whether their subsidy rates cover the actual cost of high-quality care, and other detailed questions about their methodology. In the FY 2022–24 set of plans, 10 states—or 43 percent of the 23 that did not receive a waiver—used cost modeling to satisfy the requirement for a narrow cost analysis. In addition, the District of Columbia and New Mexico fully rely on cost modeling, and two more states (Delaware and Nevada) are using part of their ARP funding to create a cost model.¹³ Virginia has also said it will be transitioning to a cost model for future reimbursement rate calculations.¹⁴ In total, that is at least 17 states that have already developed a cost model; due to the large number of states that received waivers for this portion of the CCDF plan, it is unclear whether more states have or plan to implement such a model.

The use of cost modeling for CCDF has bipartisan support in Congress. The Republican CCDBG Reauthorization Act, introduced by Sen. Tim Scott (R-SC) in the 117th Congress, included a requirement for states to use a cost estimation model and garnered 14 Republican co-sponsors.¹⁵ Meanwhile, the Democratic Child Care for Working Families Act, introduced by Sen. Patty Murray (D-WA) in the 117th Congress, included a parallel requirement and garnered 36 Democratic co-sponsors.¹⁶

Developing a cost model is a way to account for all factors that affect the business of child care: rent and utilities, staffing requirements and ratios, compensation and benefits, local costs of living, and more. Importantly, cost models are adjustable, allowing administrators to understand how changes in key drivers of program quality, such as staffing, affect overall costs.¹⁷ The responsiveness of a cost model would allow state leaders to identify priorities for how additional funding could be used, if available, but would not force them to raise payments before the investment is approved. It also would allow agencies to target their changes to specific priorities. For example, given the extremely low wages in the child care sector, an agency could adjust its model to determine the needed payment rate that would support a $15 per hour minimum wage. Cost modeling can begin to address the systemic failures facing the child care system by allocating funding based on what providers need, not what they have been able to scrape by on in the past.

As leaders across the United States reckon with the reality of how critical child care is for families, children, and the economy, states have options for near-term changes. The current process of relying on a snapshot of child care prices will never enable the sector to move beyond barely making ends meet. Transitioning to cost modeling would allow for greater understanding of how short America’s current investments fall relative to the sector’s needs. It would also enable agencies to use their funding strategically in the future, investing in higher wages and better facilities while also improving the quality of child care.

Cost models are not a replacement for additional funding. In fact, universal implementation of cost modeling could help to quantify and advocate for the level of funding necessary to meet the strategic goals set by leadership. At lower funding levels, states may face trade-offs between the size of subsidy payments and the number of families they are able to serve. In the long run, public investments can be made to ensure that payments cover both the true cost of providing child care and all families who are in need of assistance.

This report uses the publicly available state-level CCDF plans for fiscal years 2022 through 2024. The plans are formatted as a questionnaire and comprise eight sections focused on different aspects of administering child care assistance.¹⁸ Each state submits its CCDF plan to the Administration for Children and Families (ACF). The plans are available through ACF and through each state’s lead child care agency. The author reviewed all 50 U.S. state and the District of Columbia’s plans to aggregate information about their child care subsidy reimbursement rates. The review primarily focused on sections 4.2 and 4.3 of the plans, titled “Assess Market Rates and Analyze the Cost of Child Care” and “Establish Adequate Payment Rates,” respectively.

Within Section 4.2, state agencies answer a series of questions about their methodology for assessing the market rates, including what type of methodology they used, how they incorporated public input on the methodology, the dates of data collection, and where the full report of their methodology and findings can be found online. Where necessary to gather complete information, the author consulted the separate, publicly available state reports on their methodology.

Section 4.3 requires states to provide their “base payment rates and percentiles … based on either the statewide rates or the most populous area of the state (area serving highest number of children accessing CCDF)” for each combination of age group and setting of care.¹⁹ They define the age groups within the questionnaire as infants (6 months old), toddlers (18 months old) and preschoolers (4 years old). The questionnaire also includes rates for a school-aged children (6 years old). The author excluded this final type of care because it is primarily offered as before- and after-school hours, and converting the rates to the equivalent of full-day, full-time care does not offer a meaningful comparison.

The complete aggregation of data from all 51 plans is on file with the author. All in-text references to states’ reimbursement rates and methodologies refer to what the states submitted to ACF for their CCDF plan and do not incorporate any updates to reimbursement rates that may have occurred since their submission.