NCCIC. Information Products. Conducting Market Rate Surveys and Establishing Rate Policies.

Over the last 15 years, the child care field has changed in dramatic ways. In size, scope, and sophistication, child care has grown. The number of women in the labor force has continued to increase, as has the number of children in out-of-home care. Welfare reform amplified the connection between access to child care and adult employment and training activity, united funding streams, and increased the level of state and federal investment in subsidized care. Research on the importance of the birth-to-three period in a child's cognitive and emotional development, coupled with a growing awareness of the importance of early childhood education, has heightened attention to quality standards in child care. New initiatives have been launched by states involving professional development, tiered quality rating schemes, and other tools to leverage quality.

Among the changes to the child care delivery system has been a market-driven approach to the purchase of publicly funded child care services. Beginning with the Family Support Act of 1988, Federal funding requirements have stipulated that child care subsidy rates be informed by "market rates," that is, by the price of child care services in a market. The Child Care and Development Block Grant Act continued this market orientation, requiring states to permit parental choice in the marketplace. In 1996, the Personal Responsibility and Work Opportunity Reconciliation Act required states to survey their child care markets at least every two years and also both increased and streamlined funding, which gave many states the opportunity to raise the rates paid for child care. Thus, market rate surveys, and the rate-setting policies and reimbursement rules informed by them, are at the core of this market-based approach to child care. They are also central to the delicate balancing act of ensuring access to subsidized care while at the same time promoting the quality of child care.

Federal regulations outline three core components of the market-based approach to child care subsidies under the Child Care and Development Fund (CCDF). The CCDF Final Rule was promulgated in the Federal Register in August 1998 (45 CFR Parts 98 and 99), and this Rule provides the present federal regulatory framework for market rate surveys. First, with regard to rates, CCDF Lead Agencies must assure that subsidy rates are sufficient "to ensure equal access" for eligible families to child care services that are comparable to those provided to families that do not receive subsidies. In addition, child care payment systems must be structured in such a way as to permit eligible parents to choose among various types of providers, including issuing certificates or vouchers for parents to use in the private market, as well as establishing grants or contracts with providers. Third, Lead Agencies also are required "to collect and disseminate to parents and the general public consumer education information that will promote informed child care choices."

This paper is intended to provide a brief overview of the market-based approach and a discussion of planning and conducting market rate surveys and establishing rate policies for the administration of the subsidized child care system.

In 1994, in the midst of the shift toward a market-driven subsidy system, Louise Stoney, writing for the Children's Defense Fund, observed that a market-based approach "recognizes that the child care delivery system in the United States is primarily a private, fee-for-service system in which parents act as consumers by purchasing care for their children" (Stoney, 1994). Because government is not the sole or even primary consumer in the private child care market when it purchases care for low-income children, it needs to establish a rate structure that gives subsidized families the ability to purchase care at least comparable to that available to non-subsidized families. Basing reimbursement levels on the market price of child care is intended to achieve this end.

In the absence of a market-based approach, reimbursement rates might be set arbitrarily and reflect only what a state felt was a reasonable rate, or be derived backwards from a budgetary baseline without consideration of the price of care. In the past, without information about prevailing prices for child care in a market, subsidy rates tended to slip below-sometimes far below-the fees paid by non-subsidized families. And often rates remained frozen over relatively long periods. Understandably, some providers limited their enrollment of children with subsidies because the subsidy payments were too low or tardy. As explained in the CCDF Final Rule, "A system of child care payments that does not reflect the market makes it economically infeasible for many providers to serve low-income children-undermining the statutory and regulatory requirements of equal access and parental choice." These concerns, then-equal access and parental choice-are the primary goals of the market-driven subsidy system.

But how do you determine what are the prevailing conditions in the child care marketplace and assure that subsidy rate ceilings are sufficient to provide equal access for all families and freedom of choice for parent-consumers? Conducting a methodologically sound market rate survey is an essential first step in developing effective child care subsidy rate policies. Once armed with survey data, states also can assure equal access and preserve parental choice by establishing rate policies and payment system rules that reflect market realities.

The statutory requirements and regulatory framework associated with the market-based approach to child care delivery give the Lead Agencies flexibility in establishing rate ceilings and setting rate policies. The market-based approach requires that rates and rate policies be designed to achieve the goals identified above-ensure equal access, promote parental choice, and provide consumer education-but leaves room for states to demonstrate how those goals are met. For example, a Lead Agency must show that it considered three factors in determining that its child care program provides equal access for eligible families: 1) the full range of categories and types of providers; 2) adequate payment rates; and 3) affordable copayments.

In determining the adequacy of payment rates, states are required to have conducted a market rate survey within two years of the effective date of their currently approved CCDF plan, but no particular method is specified. Subsidy rate ceilings should be established taking into account the market rate; however, the preamble to the CCDF Final Rule suggests as a benchmark that rates that are established at least at the 75^th percentile "would be regarded as providing equal access." The Preamble also suggests that rate surveys should show variations in market prices by age and also among categories of care, and any appreciable sub-state variations that may exist. States may then consider these factors when designing rate structures.

It should be kept in mind that states don't set rates, they set rate ceilings. In other words, states pay what the provider charges so long as it doesn't exceed the subsidy rate ceiling. A general principle expressed in the Preamble to the CCDF Final Rule is that "Federal subsidy funds can not pay more for services than is charged to the general public for the same service."

Copayments are another element of the rate system about which states have flexibility in meeting the statutory requirement. Families must share in the cost of care, unless the Lead Agency waives the requirement for families at or below the poverty level and for children in protective services on a case-by-case basis. As with subsidy rates, states may set the copayment at any level as long as they demonstrate that established copayments are "affordable." The Final Rule suggests that affordable copayment scales require a low-income family to pay no more than 10 percent of its income for child care, regardless of the number of children in care. Such sliding fee scales must be based on family size and income and states are encouraged to structure them so that small increases in wages do not result in large increases in copayments.

The principal tool for states to learn more about the price of child care services in a market is the market rate survey. A market rate survey measures the prices charged by providers and paid by parents in a given child care market. Conducting a market rate survey can be a daunting task for a Lead Agency and needs to be done with special consideration for the complexities and variations within the child care system in a state. Methodological and administrative considerations in conducting surveys in the child care field and using the findings for establishing rate policies are discussed in this report.

The Relationship of Price and Cost. Market rate surveys can only capture data on the price of child care, not on the cost of child care in a given market. The price of child care is the fee per child paid by non-subsidized families for child care services or, when serving a family receiving the subsidy, the price of child care includes the subsidy payment plus the family's copayment. The cost of child care, on the other hand, is the actual per-child cost of the program for providing child care services and may be paid for by public or private grants, gifts or in-kind donations, as well as fees collected from consumers.

To the extent that cost exceeds price in a market over time-as may occur in low-income neighborhoods where consumers often lack the financial resources to afford high-quality child care, or for segments of the market such as care provided to special needs children or during non-traditional hours-market forces may compel providers to reduce charges and cut costs to avoid going out of business. Cost reductions may mean reducing staff or lowering wages and are likely to result in reduced quality. Whether a provider goes out of business or not, when cost exceeds price in a market over time, both the quality of care and the supply of available care are likely to decline. This report discusses a number of efforts to adjust rate ceilings and narrow any gap between the market cost and the price of care.

Defining the Market and the Sample Population. Conducting a market rate survey begins with identifying the market to be surveyed, specifically its geographic boundaries, the providers to be included (and perhaps, excluded), and the consumers to be included or excluded. It may reasonably be predicted that surveys will yield different results if the geographic area is a city within the state or a rural region of the state. Similarly, results will vary if the survey excludes segments of the provider population or is a census of all providers within a jurisdiction. This issue is further complicated by the wide variety of child care arrangements that often exist within a market, from in-home care provided by a relative to care provided on a large-scale by national for-profit firms.

Instrument Design and Survey Timing. Designing an effective survey instrument must be done with care. It is essential to employ sound research methods and to obtain high response rates from different categories and types of providers. How long should the survey be? When should the survey be conducted? These are just some of the questions those conducting a market rate survey must address.

Information to Supplement the Market Rate Survey. While a market rate survey is a fundamental instrument for assessing the price of child care services, it is not the only source of information available to State Child Care Administrators when setting rate policies. For example, the CCDF Final Rule suggests that current data from child care resource and referral agencies or licensing bureaus also may be consulted together with the rate survey.

Conducting the market rate survey is only the first stage in developing a market-based child care delivery system. Once data has been collected, a second stage begins, involving data analysis, establishing rate policies, and rule-making regarding the operation of the subsidy payment system-all factors that will have an important effect on how a subsidized child care system interacts with the wider child care market. This report discusses establishing subsidy rate ceilings taking into account the survey-determined market prices; the use of differential rates to ensure equal access to special needs and other categories of care and to improve quality; as well as policies affecting how rate policies are applied and, thus, transformed into a payment to a provider.

The remainder of this paper is organized in two parts, one dealing with the market rate survey itself and the second with the process of establishing subsidy rate policies based on information from the survey and related sources. Part I examines the subject of conducting a market rate survey in three chapters. Chapter 1 ("What Is a Market Rate Survey?") summarizes the background on market rate surveys and highlights requirements and guidance in the CCDF Final Rule. Chapter 2 answers the question, "What Are the Steps Involved in Planning a Market Rate Survey?" Specifically, it outlines the five steps involved in planning a survey, building upon an earlier paper on the subject (Stoney, 1994) as well as the presentations and discussions from a November 2000 meeting conducted by the Child Care Bureau on Rate Setting Policies: Ensuring Access and Improving Quality (Schock and Daugherty, 2001). Chapter 3 examines the question "What Are the Steps Involved in Conducting a Market Rate Survey?" In Part II, we focus upon how to use the findings of the market rate survey, together with other information, to formulate rate ceilings and related child care subsidy administration policies. Issues discussed include quality differentials to ensure access and improve quality, "special needs" rate ceilings, supplemental rate-setting methods, payment policies and procedures, and alternative strategies.

Part I-Market Rate Surveys

CHAPTER 1. What Is a Market Rate Survey?

market rate survey is a survey to determine the price of child care. Child care providers are surveyed within a specific geographic area that has been identified as the market area.

A distinctive feature of the child care market is the difference between the price of care and the cost of care. The price of child care is the rate charged to families who receive no assistance from public or private sources in paying for child care. When all families in a child care center or family child care home receive help paying for care, then the price of care includes the subsidized portion of the fee plus the portion paid by the family.

The cost of child care is the market cost when all factors that make up the cost of providing child care services (e.g., personnel, facilities, equipment, and supplies) are accounted for, including gifts, grants, and in-kind donations or discounts from public or private sources.

Many advocates and analysts also use the concept of full cost-when all market costs are accounted for, including the opportunity costs and foregone income of staff who tend to be paid less than workers in other occupations who have comparable education and qualifications. This paper does not use the full cost terminology since it is not generally considered in the conduct of a market rate survey or used by states in setting rate policies based upon survey findings.

For reasons that are discussed in this paper, in many cases the price of child care is substantially less than the cost of care. Since a market rate survey focuses on the price of care, it generally provides only part of the information necessary for a full understanding of pivotal factors that shape the child care market or that should be considered by the state in setting rate policies.

Market rate surveys became common as a result of child care subsidies introduced under the earlier 1988 welfare reform legislation (the Family Support Act), the Job Opportunities and Basic Skills (JOBS) program and the Transitional Child Care (TCC) program. A market driven approach to the purchase of publicly funded child care was central to these federal initiatives. This approach mandated that child care reimbursement policies established by states be determined on the basis of market rates. The market rate approach tended to improve child care reimbursement rates in most states because, for the first time, states were required to establish rate ceilings in relation to market practices (Stoney, 1994).

The legislative and regulatory framework for market rate surveys was changed radically by the passage of the most recent welfare reform legislation, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996. PRWORA required that four major federal child care subsidy funding streams be administered as a unified program, which has been named the Child Care and Development Fund (CCDF).

The CCDF Final Rule bases the requirement that states conduct market rate surveys on the statutory mandate that states ensure that payment rates are sufficient to ensure equal access to child care services for eligible families that are comparable to families that do not receive subsidies. The state is required to demonstrate that it considered three factors in determining that its child care program provides equal access:

These factors must be spelled out in the state's CCDF plan, which states are required to submit every two years (the next biennial plan is due in July 2001 for the period October 1, 2001 to September 30, 2003).

The following points are emphasized in the discussion of market rates in the Preamble to the CCDF Final Rule.

CHAPTER 2. What Are the Steps Involved in Planning a Market Rate Survey?

his chapter outlines the steps involved in planning a market rate survey and is distilled and adapted from an earlier paper on market rates (Stoney, 1994) and from deliberations, presentations, handouts, and resource materials at the meeting conducted by the Child Care Bureau in November 2000 on Rate Setting Policies: Ensuring Access and Improving Quality (Schock and Daugherty, 2001).

There are five steps, with related sets of decisions, that lead up to a market rate survey:

Step 1. Conduct a preliminary analysis of the child care market(s) within the state.

The child care industry is very different from other types of businesses in the United States. All child care is local, and there is no statewide market. Even within a single community, the child care delivery system can be highly diverse and largely unorganized. It includes a grandmother, other relative, or neighbor who may care for one or a few children and who may or may not be licensed or regulated, and may or may not be receiving subsidies from the state. It includes child care centers who may serve a small number or several hundred children, as well as family child care providers who may serve only a few children. It also includes providers who may be exempt from licensing in some states, such as faith-based organizations, Head Start, prekindergarten programs, and before- and after-school (out-of-school-time) programs administered by the public schools or other organizations.

Many of the above categories of care and types of providers will be unknown to the state, not recorded in licensing databases, and not tracked in subsidy financial systems. Yet they are included in the overall child care market and their services to children and families influence the price of child care within the neighborhood or area market of which they are a part.

Some knowledge of the parameters of the child care delivery system is essential even to begin thinking about how to plan a market rate survey. So too are other fundamental practices among child care providers. How do they establish rates-by the hour, day, week, month, session, year? Do they offer part-time services, and how is part-time defined? Are there separate fee categories for infants, toddlers, preschoolers, and school-age children, and what are the age breaks for defining these groups? How is "special needs" defined in the state and what are common practices of providers in recruiting and serving children with "special needs" and in offsetting the additional costs associated with their care?

Preliminary information about these and other aspects of the child care market is needed to plan the survey. The best approach to obtaining this background data is through a two-part process, conducting a market assessment study, either prior to, or as a part of, the market rate survey. If done as a part of the market rate survey, it might be feasible to design an in-depth initial survey to obtain comprehensive information, and later annual or biennial surveys could be briefer, using a simpler format, to update the baseline data.

At a minimum, an initial market assessment study should include sufficient information: 1) to enable the state to identify provider categories (who should be surveyed?); 2) to be able to decide how to sample providers; 3) to determine how to handle the issue of providers who serve children with "special needs;" and 4) to determine appropriate political jurisdictions and other areas to be surveyed.

Any assessment study could be supplemented by background information provided by the child care community. Child care resource and referral agencies (CCR&Rs), child care professional associations representing various categories of providers, Head Start and other early education programs, and parent groups can be invaluable sources of insight on the workings of the child care market. Moreover, it is essential to enlist their involvement and support early on in the process of planning a market rate survey. CCR&Rs are sources of data about child care supply and demand (and sometimes the price of care) that will go beyond data the state would normally obtain.

The Preamble to the CCDF Final Rule advises that the state survey the full range of categories (e.g., center-based, group, family, and in-home care) and types of providers (for-profit and nonprofit providers, sectarian providers, and relative providers). According to the Preamble, states must also include in the survey infants, toddlers, preschoolers, and school-age children, as well as children with "special needs." This does not settle the question of who to survey, however, since the provider groupings are far from homogeneous. The pivotal question is: What types of providers participate in and shape specific markets?

At the heart of the matter is the issue: What is child care? Does child care include Head Start? Prekindergarten programs? School-age care? Summer camps? Different states have answered these questions in varying ways. In some states (e.g., Ohio and Oregon), Head Start is an integral part of an overall strategy for upgrading the quality and coverage of child care programs, with both Head Start and child care utilizing multiple funded sources and often operating in partnership. Also in certain states (e.g., Connecticut and Georgia), prekindergarten (PreK) programs operate in close partnership with child care-in Georgia the majority of the PreK four-year-olds are in programs operated by child care providers. School-age out-of-school-time programs tend to be funded through CCDF and the 21^st Century Community Learning Centers education sources, sometimes blending funds from both sources-bearing in mind that there are approximately 100 other funding sources for out-of-school-time programs (Halpern, Deich, and Cohen, 2000).

The current circumstances in each state, together with changing trends, must be taken into account in deciding whether and how to include child care providers that receive funding from Head Start, PreK, and school-age programs. If states want to support blended funding approaches and ensure that parents are able to purchase child care from providers that access multiple funding streams, it is essential that these providers be included in the market rate survey. Simply excluding Head Start or PreK programs from the survey because they typically may not charge parent fees or are funded with sources other than CCDF is short-sighted, particularly since the services provided by these programs is a factor in the supply-and-demand dynamics of the child care market. In fact, as blended funding becomes the norm, fewer and fewer programs can be described accurately as exclusively Head Start or prekindergarten programs. Rather, they are comprehensive child development and early education programs with a range of funding streams and program options. However, some states have experienced that these programs are reluctant to respond to being surveyed as child care programs. One approach when asking rate survey questions of programs where no parent fee is assessed is to ask such program respondents to calculate what their rate would be if they charged parents fees and to reiterate the purpose of including these programs in the survey.

If the state elects to survey full-time, center-based child care providers that are exempt from regulation but which do charge parent fees (such as those operated by faith-based organizations or those located on federal military bases or tribal property), caution should be used. Their costs, parent fees, and overall pricing structure may be quite different, since they are not required to conform to state regulations. States may wish to design the survey instrument so that results from particular types of programs, as well as other information, can be coded (e.g., having respondents use tick boxes to identify which provider category they exemplify) for further discrete analysis and treatment.

There are particular difficulties in surveying informal, unregulated home-based providers. The size of this segment of the child care market, home-based and legally exempt from regulation, is unknown, making the statistical exercise of obtaining a representative sample size a challenge. One approach is to use the market assessment study to obtain an estimate of this particular market segment. Once such an estimate is obtained, identifying the individual providers and locating their contact information for surveying purposes may require significant resources and effort. Some states elect not to survey this group of providers because of the substantial challenge of obtaining a representative sample and some observers have recommended that states not even attempt to survey legally exempt home-based providers (Stoney, 1994).

Alternative strategies to surveying, such as applying rate adjustment factors, should be considered in the case of these providers. Such strategies are discussed further in Part II of this paper.

A child care market rate survey measures the child care rates charged by the providers-and paid by the consumers-in a specific geographic area (Stoney, 1994). A key step in planning the market rate survey is defining the geographic areas that comprise the particular child care market from which price information is to be gathered.

Because child care rates commonly vary widely depending upon the community in which the providers are located, the determination of geographic areas can have a profound effect on market rate ceilings established by the state. States should carefully consider the factors that correlate with child care rates and select geographic areas for the survey that reflect critical differences. Whenever possible, the market assessment study or available market rate survey data should be used to determine the geographic areas, rather than using the state's established political jurisdictions, such as city or county boundaries.

Many states have established rate ceilings on the basis of political jurisdictions that have been used for the delivery of human services, such as counties or social service department service delivery areas. Child care subsidy administrators and providers tend to be comfortable with this decision because they are accustomed to thinking about service delivery within such a framework. The downside of ignoring these jurisdictional boundaries is lack of stability and consistency that can create political problems for administrators and frustration for providers who may experience rate decreases because they shift from one rate grouping to another. Tying rates to delivery areas also has advantages for authorizing workers, providers, and parents in terms of ease of understanding information.

Such political jurisdictions may not, however, accurately reflect variations in the price of child care in the market. To ensure that the rate ceilings derived from the market rate survey accurately reflect market practices, the geographic areas used for the survey should be based upon the actual price differences that prevail in communities, rather than preconceived assumptions about which areas belong together.

A two-part process is proposed. First, data from the market assessment study or initial market rate survey are used to define a set of parameters that correlate with rates, which can then be used to define geographic areas for the upcoming survey. Second, the geographic areas and political jurisdictions used to establish rate ceilings are not determined until after survey results have been analyzed. In Oregon, for example, market rate survey data are collected at the neighborhood level using zip code boundaries; then, during data analysis, the zip codes are combined to determine the jurisdictions used to set rate ceilings. This approach has the further advantage of collecting data in the smallest feasible geographic areas in order to assure optimal estimates of the price of child care in that locality.

Using survey data to define the market areas within which child care rate ceilings are established assumes that areas with similar rates should be clustered together, whether or not they are contiguous or coterminous with historically determined political jurisdictions. Factors that predict child care rates should be the basis for selecting geographic areas that reflect critical differences. Research has shown, for example, that areas with high median incomes or high housing costs tend to have higher child care prices. Similarly, areas with high unemployment tend to have lower child care prices.

While defining geographic areas in terms of similar child care prices may result in more accurate market rate survey data, this practice is likely to result in low market rate ceilings in low-income neighborhoods. Providers in such areas may find that the rate ceiling is too low to cover their costs and enable them to offer high quality care. In such circumstances, the state should consider establishing differential rates for providers who meet higher standards or offer additional services, as well as the possibility of combining lower rate and higher rate areas within counties or commuting areas, which can have the effect of moderating low rates in rural parts of a county, for example, without significantly affecting higher rate areas in the same county. A state also may use an alternative rate setting methodology as discussed in Part II.

States that use predetermined political jurisdictions to establish market rate ceilings may find that sample sizes may be too small to reliably calculate a rate ceiling. This issue, alternative strategies, and policy implications are also discussed further in Part II.

While there are advantages and disadvantages to using political jurisdictions as the basis for setting rate ceilings, states can design their survey instruments to collect data from more precise geographic areas, offering a more accurate reflection of the child care market, and then craft a rate structure using areas that may or may not align exactly with the geographic areas of the survey.

The CCDF Final Rule has three provisions that relate to the timing of the market rate survey: 1) the survey must be conducted in the context of reasonably current market conditions to ensure that the payment rates continue to provide equal access; 2) the survey must be conducted at least every two years; and 3) the state may elect to conduct a full market rate survey biennially, and also survey targeted sub-samples in specific areas as frequently as it deems necessary.

Establishing the timing of the market rate survey is the result of considering several tradeoffs. To ensure that rate data are current, the effective date of rate increases must be taken into account. If the survey were conducted in March, for example, and the state were to publish new market rate ceilings in December, to become effective in January, child care providers might have raised their rates in September. In such a scenario, the "new" rate ceilings would already be 3-4 months out of date in terms of common practices of providers. If the rate ceilings remain in effect until the next biennial survey, the gap between the rate ceilings established by the state and the price of child care charged in the market will continue to widen.

Conducting an annual market rate survey is one strategy to ensure that rate data are more up-to-date. Some states that have attempted this have found it to be an expensive proposition. An alternative strategy is to conduct targeted sub-samples in specific areas, utilizing decision rules to apply increases to all providers in the category.

Another option is to include questions about rate increases on the market rate survey. The New Hampshire 2000 survey asked providers when they planned to raise rates (the survey included four options: within 3 months, within 6 months, within 12 months, and no plans). Additionally, New Hampshire asked providers to report their rates on March 29, 2000 (the snapshot date selected for the survey) and on March 29, 1999. By comparing these two rates, the state could get a sense of how much rates changed in a one year timeframe.

Deciding when within the year to conduct the survey poses other tradeoffs. Child care and other early education programs tend to begin operations in the fall, generally in line with the school calendar. Fall enrollment tends to be peak enrollment. In theory, this would be an ideal time to conduct the market rate survey. However, child care providers tend to be busiest this time of year with program start-up, and it may be difficult to contact them in a telephone survey or to get their attention to respond to a mail survey. If the survey is delayed until late October or November, it may encounter holiday delays.

Conducting a market rate survey requires both an in-depth knowledge of the child care field and expertise in survey research methodology. This is a challenging and time-consuming task, even for individuals with the requisite knowledge and experience. The principal tasks include: 1) selecting a representative sample; 2) developing a survey instrument; 3) conducting provider training about the survey; 4) pilot testing the instrument; 5) implementing the survey; 6) analyzing the data; and 7) preparing a report of survey findings.

The state has three broad options in determining who is to conduct the survey. First, in-house staff can be assigned this responsibility. Second, the survey can be contracted out. Third, a two-stage process can be used-selecting a contractor who will be called upon to conduct the survey, and who will involve in-house staff in this process, with the aim of training the staff, and ultimately turning the process over to the state staff to conduct future market rate surveys.

It may appear at the outset that using in-house staff is cost-effective, either initially or after training by a contractor. This is not necessarily the case. State staff with expertise in both survey research and child care tend to be few in number. Additionally, state child care staff are busy people with workloads that have grown tremendously in recent years. Moreover, successful conduct of a market rate survey requires sustained attention over time which may be difficult for state staff that are simultaneously involved in other tasks. Nevertheless, a state may choose to develop in-house, dedicated expertise.

States that have elected to contract out have tended to turn to one of the following sources: university research groups, private consulting firms, or child care resource and referral agencies (CCR&Rs). Successful surveys have been accomplished by all three groups.

Using university research organizations offers states the benefit of strong technical expertise in survey methodology and statistical analysis. However, not all universities have staff who are knowledgeable about the child care market. Similarly, states are served best by private consulting firms that can offer both technical expertise in surveying and statistics as well as understanding of the dynamics of child care supply and demand. For example, the California Child Care Resource and Referral Network contracted with a survey research firm to do data collection for the California market rate survey. A challenge facing CCR&Rs, which are locally based, is collecting survey data from all parts of the state in a consistent manner, using identical questions, asked by interviewers (in the case of telephone interviewers) who are completely neutral. These issues can be resolved by states in negotiating with the CCR&Rs about the methodology to be used in conducting the survey and in ensuring that the staff involved in data collection, data entry, data analysis, and reporting are carefully trained and supervised.

At a November 2000 Child Care Bureau issues forum on conducting market rate surveys, states were cautioned about relying on rate data regularly collected by CCR&Rs as a substitute for market rate survey data for several reasons (Schock and Daugherty, 2001). First, the data that must be collected for a child care market rate survey may be somewhat different from the rate information included in a CCR&R database. Providers may not provide CCR&R agencies with full rate information because they are marketing their services to prospective parents and therefore want to appear as affordable as possible. Businesses of all types typically advertise their cheapest rates. You don't find out, for example, that the bargain airfare you read about in the paper is only for travel on Mondays during the winter months until you call to make a reservation. Similarly, child care programs may attempt to make their prices look lower by the way they report information to the CCR&R. Or they may decide not to include any rate data and instead simply provide the CCR&R with their hours, ages of children served, etc., and request that parents call directly for rate information.

CHAPTER 3. What Are the Steps Involved in Conducting a Market Rate Survey?

Building upon the five steps in planning a market rate survey, there are six additional steps in conducting the survey:

These steps, and related decisions, are discussed further in the remainder of this chapter.

Formulating the questions to include in the market rate survey requires a combination of expert knowledge of child care and of survey research techniques. The questions, when posed appropriately to providers (see Step 7 on survey methodology), must be capable of generating reliable information that will enable decision-makers to establish rate ceilings and related child care policies.

At the outset, criteria should be established for which questions to ask in the survey. Suggested criteria are:

An in-depth discussion of market rate survey questions and related issues is contained in the proceedings report of the meeting sponsored by the Child Care Bureau in November 2000 on Rate Setting Policies: Ensuring Access and Improving Quality (Schock and Daugherty, 2001). The proceedings report also includes examples of actual surveys used by states.

As discussed under Step 2, to the extent feasible, the state should attempt to incorporate questions that include all categories and types of providers in the market rate survey. If this is determined to be infeasible, supplementary cost studies, alternative rate setting methodologies, or other innovative means must be used to address the issue.

The age groupings used by a child care provider are typically driven by the state's licensing requirements. State regulations might require, for instance, a staff-child ratio of one to four children 18 months or younger; a one to five ratio for children 19 to 36 months old; a one to seven ratio for three-year-olds; and so forth. When these ratios are translated into staff costs, the implications are clear-the younger the child, the higher the cost of providing care.

But the rates established by providers are affected by market practices as well as by licensing requirements. Despite the higher costs associated with serving infants and toddlers, some providers choose not to charge significantly higher fees for these children. The reasons for this decision vary. In some cases, the provider may decide that parents cannot afford to pay a higher fee for infants, or they may find that their competitors do not charge higher fees for infants, and they cannot keep their spaces full if they charge higher fees. In other cases, the provider may decide that it is simply easier to average the cost of serving all age groups and charge one fee for all children.

Understanding the child care market is crucial in selecting the most appropriate age breakouts to use in setting rate ceilings. CCR&R agencies and provider associations can be an excellent resource in this regard. States can also use the survey to inform this decision. The survey instrument might, for instance, ask the question do you serve infants? If the answer to the question is yes, the provider is then directed to provide information on the ages of children included in the definition of infants and the rates charged for children of this age. Data analysis may reveal that a preponderance of providers define infants as children 18 months of age or younger. This definition can then be used in establishing rate ceilings and conducting future rate surveys.

Another key variable that must be determined is the unit of service by which the rate will be measured. The term unit of service refers to the unit of measurement that is used to establish the price of care. Child care providers often use a number of different units to establish child care rates. Some charge by the hour, others by the day, the week, or the month. To be credible to providers, a market-driven system should collect data and base reimbursement rate decisions on the units of service they use. From the state's perspective, however, it can be overly complex, costly, and difficult to administer a system of rate ceilings based upon a wide range of payment units. A manageable number of units of service should be selected, whenever possible using multiple units of service.

When a limited number of units of service must be selected, formulas must be established to convert rates into appropriate units for providers who do not charge fees according to the units of service requested on the survey. Such formulas are typically referred to as conversion formulas (Stoney, 1994). The way the state defines units of service in the survey and any conversion formulas that are adopted will have a great impact ultimately on providers' budget processes.

The cost of providing care on a part-time basis is greater per hour than the cost of providing full-time care. In either case, providers must meet mandated staff-child ratios, health and safety standards, and overhead costs throughout the day, regardless of the hours a particular child is in attendance. With full-time care, economies of scale work to reduce per child costs; with part-time care, average costs are inflated.

In the attempt to balance costs and income and achieve long-term economic viability, providers must strive for full enrollment. Fee and enrollment policies are likely to encourage the use of full-time care. Parents, on the other hand, often have a need for part-time care. It may be infeasible for two part-time children to share a full-time slot. Providers tend not to allow parents to purchase care on an hourly basis.

For these and other reasons, provider practices regarding the hours programs operate and policies regarding part-time services vary widely. The specific number of hours that make up full- or part-time care can be difficult to define. Economic factors as well as market practices must be taken into account in shaping questions about part-time care. States should request that providers report full- and part-time rates separately in all provider categories.

Additional considerations arise in the case of child care partnerships with Head Start or prekindergarten programs. The child may be in care full-time throughout the day as a result of the combined partnership services. However, the state may wish to collect information through the market rate survey to inform the decision whether to reimburse the child care subsidy on a full-day or part-day basis.

In light of the above circumstances, it may be necessary to consider supplemental cost studies and alternative methodologies in establishing rate ceilings for part-time care.

Licensed capacity is determined by the state for each provider in the light of staff-child ratios, facilities, and other criteria set forth in licensing regulations. Enrollment is the number of children currently served by the provider.

Licensed capacity is measured in full-time equivalent children, rather than the number of children served. Capacity is generally greater than actual enrollment at any point in time. Providers have the option to expand if demand grows for the child care they provide. Child care centers hire staff and operate on the basis of expected enrollment and mandated staff-child ratios.

Enrollment is a more accurate measure of child care slots available to parents in the community than licensed capacity. For example, capacity data will not provide an estimate of the number of part-time child care spaces available.

If the market rate survey is to give an accurate picture of the providers' costs, care must be taken to obtain sufficient information to adjust for the impact of sliding fee scales and scholarship programs. Many providers attempt to ease the burden for families of the high cost of child care by establishing scholarship programs or sliding fee scales, sometimes using subsidies from groups such as the United Way and foundations.

Questions should elicit the actual rates charged to families in specific circumstances (e.g., taking into account family income, number of children), not "average" or "most frequently charged" rates. By asking for average rates, the state may inadvertently be penalizing providers who attempt to assist low- and moderate-income families to access child care.

The survey should ask about any special fees that may be charged by providers for services such as registration or enrollment, supplies, activities, transportation, and meals. This information can be helpful in interpreting the rates providers report charging. If special fees are charged separately, the reporting rates are understated in terms of the actual cost of providing care and do not represent the true price of child care. Moreover, if large numbers of providers levy special one-time fees, this is a barrier for low-income families obtaining care.

Carrying out a market rate survey that ensures the collection of valid and reliable information requires the involvement of experts in survey research methodology, including statisticians experienced in sampling. Parallel with those technical aspects, and of equal importance, is the necessity that the survey be conducted under the overall direction of decision-makers who will be responsible for setting rate policies and with the involvement of child care experts.

Mail surveys are cheaper; however, they tend to attain a significantly lower response rate. For states that elect to send a written survey instrument to respondents, extensive follow-up should be anticipated. Useful techniques include: 1) sending a self-addressed, stamped envelope with the survey; 2) sending reminder emails, post cards, and/or additional copies of the survey to non-respondents; 3) including the name and phone number of a person who can answer any questions; and 4) providing incentives for providers to respond quickly.

Whenever possible, market rate surveys should be conducted by telephone to obtain the highest feasible response rate. A high response rate is essential to avoid biasing the survey results for particular categories and types of providers. A skilled telephone interviewer can ensure that respondents understand the questions and clarify the intent of the question, as necessary. This demands that the telephone interviewer be neutral, unbiased, knowledgeable, and well trained.

Telephone interviewers must be sensitive to the fact that providers, particularly family child care providers, are responsible for the ongoing care of children, and may not always be able to take time to speak on the telephone. Interviews should be scheduled in advance, seeking times that are convenient for the provider, such as nap time or during the evening or weekend. A preliminary worksheet also helps to prepare providers for a telephone survey call.

Some states have chosen to balance the benefits of mail and telephone surveys by combining these techniques. Mail surveys are sent to center-based programs and telephone surveys are conducted of family child care providers. Another approach is to send out an initial mail survey to all providers, and to do the follow-up through telephone interviews of non-respondents.

Whatever the method of surveying, states have found that response rates can be improved by offering incentives for providers to respond in a timely fashion. In Vermont, for example, providers who responded promptly were entered into a drawing for prizes including $50 gift certificates for early childhood education materials and a spa weekend; the three CCR&Rs that generated the highest response rates each received a $500 reward.

The extent to which survey results accurately reflect the price of child care charged by providers in a particular market depends upon careful sampling. The sample selected must be representative of the targeted provider population in the area. If that is done, scientific statistical techniques can be used to analyze survey findings and make accurate generalizations about the providers included in the sample.

The number of providers selected for inclusion in the survey (i.e., sample size) is a critical factor. In general, a larger sample is better. In the interests of long-term cost containment, it is sometimes feasible to use a large sample in an initial survey and a smaller sample in subsequent surveys.

A competent statistician can advise on the size of sample that is needed to ensure that it is representative of the group of providers from whom data are being collected. Keep in mind that the ultimate goal is to obtain data about the price of child care that is representative of the child care market in general.

There are two critical aspects to survey sampling. First, the sample must be properly drawn and of sufficient size to be representative. This aspect focuses on the providers who are asked to respond to the survey. Second, providers who actually do respond must reflect the target respondents. If individuals do not participate in the survey, and in particular if the characteristics of those who fail to respond are different from respondents, the results of the survey can become biased. Severe bias can render survey findings worthless and, more seriously, can mislead policy makers and lead to rate ceilings based upon erroneous assumptions.

The careful selection of survey strategies, coupled with a lot of hard work, can optimize response rates. Target response rates of 80-85 percent are optimal (Stoney, 1994). California, for example, which surveys approximately 14,000 providers, manages to achieve a 71 percent response rate (Schock and Daugherty, 2001). The CCR&R, which has overall responsibility for the California market rate survey in collaboration with a market research firm, over-samples by 15 percent, calls back programs that don't respond, and makes appointments with providers to call them on the weekends or in the evenings, when necessary.

A critical question is how many providers in a geographic area need to respond to enable the state to have a reasonable degree of confidence that respondents constitute a child care market that would enable decision-makers to set rate ceilings. Statistical rules of thumb tend to suggest that 20 to 30 observations of a "cell" (in this case, a category of rates for each child age/unit of service/provider) are usually sufficient to approach a normal curve and enable the researcher to calculate a reliable mean. Below 20 observations, statisticians worry about skewed results.

Unfortunately, such rules of thumb are based on the assumption that the respondents are reasonably homogeneous in their critical characteristics. This assumption is frequently not true of child care providers where diversity, rather than homogeneity, tends to rule. Moreover, it also assumes that those responding to the survey represent a substantial proportion of the population of providers surveyed (a "high" response rate) and that those responding are representative of those initially surveyed.

To ensure accuracy, it is essential that all terms used in the survey be clearly understood by respondents.

It is essential that respondents not misinterpret questions. Terms and definitions may differ among different groups of providers (for example, "special needs" is a term that is likely to differ across and within states). Terms should be defined from the provider's perspective and should reflect common provider practices to the extent feasible.

Definitions should be made part of the survey questions, rather than being a separate section. For example: "The state's licensing regulations define infants as birth to 12 months of age. Do you provide care for children in this age range?"

For mailed surveys, survey format can be as important as the wording of the questions. Crowding a lot of information onto a page, or using small print, to make the survey appear less burdensome is usually a mistake. It is better to have a page layout that is easy to read and understand. Greater accuracy can be assured by eliciting provider input on survey layout and question phrasing, if not in the design stage then during survey piloting (see Step 9 on pretesting the survey instrument).

Telephone surveys should be in the form of a "script" that can be read quickly and easily. Telephone surveys should use "branching" techniques to make it easy for the interviewer to follow a related series of questions, depending upon the answers given by the provider. For example, if the provider does not serve infants, the interviewer then skips questions in that branch that relate to services to infants, and jumps to the next branch or series of questions focused on the age group the provider actually serves. Ideally, telephone interviews of child care providers should not take longer than 10 minutes to complete.

States with providers whose dominant language is other than English may need to translate the survey instrument into other languages. They will also need to utilize telephone interviewers fluent in the languages spoken by the providers.

Providers should be informed about the survey, including who will be conducting the survey, when providers will be contacted, how survey findings will be used, and what their participation contributes. In particular, providers who do not currently provide subsidized care will need to be informed about why and how the survey is designed to discover the price of child care in the total market.

Techniques the state can use to inform providers about the survey include sending out an information letter, announcements in the media, and seeking the help of CCR&Rs, provider associations, and other child care organizations to publicize the survey through their newsletters, meetings, emails, and other dissemination activities. Providers should be assured that any information they provide will be kept confidential.

Informational and outreach activities should give providers details about the conduct of the survey. For example: "If you are selected to participate in the child care market rate survey, you will receive a call from (insert name of individual or organization) during the period (insert range of dates). If you have any questions about the survey, please call (insert the name and number of contact person who can answer any questions)."

Training and group information sessions should be offered to providers to help them understand how to complete the survey accurately. Written training materials should be developed, including brief handouts to give to providers who attend the training and to send to those who may not be able to participate.

Training is particularly important for those categories of providers who may not have taken part in previous market rate surveys, such as Head Start and prekindergarten programs. For example, those programs may not charge fees, and they will need to be assisted in responding to a survey questionnaire that is tailored primarily to providers who charge fees. They may also be accustomed to a different way of defining "special needs" and other terms.

Conducting a pilot test of the survey questionnaire is an essential step, no matter how much care went into preparing the instrument. Provider training may offer an opportunity to begin the pretest process. In addition, a formal pilot test should be conducted.

To the greatest extent possible, providers should be contacted during the pilot in the manner they are expected to be contacted during the actual survey. The pretest can confirm assumptions about how providers will interpret the questions and the clarity of the definitions of terms. The reactions of categories of providers should be analyzed separately. For example, what center directors may find clear and easy to answer may strike family child care providers as confusing.

Telephone interviews should be pilot tested over the phone. Feedback on mail questionnaires can be obtained through follow-up telephone calls or by conducting focus groups with respondents.

In addition to pretesting reactions of respondents, the pilot test should include interviewers and data entry staff. To the extent feasible, data analysis should be simulated. Sample tables that will appear in the final report should be created and simulated findings should be analyzed to determine if the anticipated answers to the survey questions will in fact yield the desired information for setting rate ceilings and formulating policies.

Based upon results of the provider training and the pilot test, the questionnaire should be finalized and the survey methodology should be determined. Any necessary revisions should be made in the survey instrument, the sampling plan, or other survey procedures. The survey should be implemented according to plan.

Progress in implementing the survey should be carefully monitored. If mid-course corrections are necessary, decision makers should be consulted. The response rate should be carefully assessed for each category of providers and aggressive follow-up actions should be undertaken if the response rate shows signs of falling below predetermined targets. Selected responses should be periodically reviewed for quality control. Selected telephone interviews could be monitored to assure that questions are asked in a standardized way without bias.

Survey results should be analyzed as soon as possible after data collection. It is desirable to begin preliminary analysis even while the survey is still being conducted. This will be helpful in monitoring response rates, checking sampling, and for other purposes.

A public report of survey findings should be prepared. This is an essential step in ensuring provider cooperation for the next market rate survey and in building public support for rate ceilings and policies that are decided based upon survey results. While the public report is in preparation, an in-house summary version of the findings may be made available to decision-makers. This will enable policy makers to forecast the policy options which they are likely to address.

Part II-Establishing Rate Policies

Conducting the market rate survey is the first stage in developing a market-based child care delivery system. Once the data have been collected, a second stage begins, involving data analysis, establishing rate policies, and rule-making regarding the operation of the subsidy payment system. All of these factors may have an important effect on how a subsidized child care system interfaces with the wider child care market. In addition, how rate systems are structured and implemented may afford Lead Agencies additional tools with which to build child care capacity and improve the quality of care.

Setting Subsidy Rates: The 75^th Percentile. At first blush, once the market has been defined, the survey questions crafted and the data compiled, the task of setting rate policies may appear simple. However, complex factors involving fiscal constraints, market segmentation, and the promotion of equal access and improved quality, among other issues, keep rate setting a challenging area of program administration.

Federal regulations no longer limit states to setting market rate ceilings at the 75^th percentile. The Preamble to the CCDF Final Rule suggests that setting rates at or above the 75^th percentile of a market rate survey can be "regarded as providing equal access." In practice, the percentile at which states establish market rate ceilings is often a function of the level of state and federal funds available and the size of the population the state seeks to serve. Most states, however, have established subsidy rates at or above the 75^th percentile of their most current market rate survey (Blank and Poersch, 2000).

Determining the 75^th percentile is relatively straight forward, once data has been compiled and rate categories have been determined as discussed earlier. The 75^th percentile rate is that which is equal to or greater than the rates of 75 percent of the market rate survey respondents within a rate category. At the 75^th percentile, the subsidy rate would financially permit a family to choose to enroll a child with three out of four providers in a market area. If the rate ceiling were established above the 75^th percentile, this would further improve an eligible family's access to quality care.

In practical terms, however, the issue can be more challenging. To begin with, subsidy rates pegged to the market are only as reflective of market conditions as the rate survey which informs them. In addition to survey research methods discussed in Part I, the length of the rate-setting process is important in this regard. At times, two to three years may pass between the date of a market rate survey and the implementation of a new schedule of rates. With a regulatory requirement to conduct market rate surveys at least once every two years, some states may wish to adjust rate survey data during off years to keep the subsidy rate schedule current with market conditions, lest changes in price levels undercut a commitment to broadening access and improving quality. New Jersey officials, for example, recently added a cost-of-living adjustment to boost the subsidy rate structure (Blank and Poersch, 2000).

A second challenge in defining the 75^th percentile is determining whether to base that market level on rates by provider or by slot. Arraying price data by provider yields one price value per category of care for each provider, regardless of the number of children for which that provider cares. Rates arrayed by slot provide as many price values as a provider has slots, and in effect weight the provider's rate by the provider's size in the market. Calculating the 75^th percentile of market rates by provider or by child care slot are competing methods that are likely to yield different results, each with a claim to reflect market conditions.

Adjusting Subsidy Rates: Differentials to Ensure Access and Improve Quality. Simply setting subsidy rates at the 75^th percentile or higher still may not establish a rate sufficient to ensure equal access for child care services for all segments of the market or in all markets. In such cases, states may wish to consider differential rates to expand supply and to improve quality.

As noted earlier, the price of child care as revealed by the market rate survey, does not always convey the actual cost of providing child care for certain segments of the child care market or for different levels of quality of care. The actual cost of care for providing services to a "special needs" child usually is greater than the cost of more typical child care services, but legal considerations prevent those added costs from appearing in market rate survey responses. Similarly, costs associated with providing additional services such as social service referrals, or meeting higher standards including higher staff qualifications or accreditation standards, also may not be reflected in market rate surveys. Some child care programs are nonprofit or mission-based, part of a larger social service enterprise, and also receive in-kind contributions, United Way or other community philanthropic support; their actual costs intentionally may have little bearing on the rates they charge because they operate outside a traditional business model. Often, situations like these are related to specific segments of the child care market: low-income neighborhoods, rural areas, non-traditional hours, and special needs. To ensure sufficient capacity and support high quality of care, states have implemented various rate adjustments.

Many states already recognize variations in child care market conditions at a sub-state level by establishing regional, county, or urban/rural rates (Child Care Bulletin, 1997). In addition, in certain locations there may be a dearth of providers or a lack of capacity for certain categories of care such as infant/toddler, school-age, and "special needs." In such cases, the short supply may reveal itself in low cell size in market rate survey results, which inhibit arriving at a statistically valid determination of market rates. Beyond the surveying challenges, this issue can be particularly vexing in low-income communities where most residents cannot afford to pay the actual cost of child care and, as a result, the market price of care in these areas is below actual cost. To address this concern, some states have set aside their market rate survey results, for example in certain low-income neighborhoods, and taken the state subsidy rate ceiling and applied it to each center on the basis of the percentage of children in the subsidy system that a particular provider serves. If a center serves a disproportionately high number of children who receive subsidized care, instead of paying the statewide or regional subsidy rate ceiling, some states have paid the rate the center charges to non-subsidized families (i.e., 100 percent of that center's price of care).

In order to better reflect actual cost of care and expand capacity in certain underserved areas or categories, states can implement rate differentials or bonuses. One alternative to establishing rates on the basis of market practices is to establish rates on the basis of cost. In a state that negotiates contracts with a limited number of child care providers to serve large numbers of low-income children, this task may be relatively straightforward: request program budgets and negotiate rates according to the cost of providing the service.

But in a certificate/voucher system open to all legal providers, cost-based rate setting is far more difficult. So, in some states, a percentage bonus is paid above the prevailing subsidy rate for providers who offer infant child care services, care during non-traditional hours, or who offer services in particular geographic regions that may be underserved.

Special Needs Child Care. Under the Americans with Disabilities Act, providers must make reasonable accommodations to serve children with disabilities, but the ADA prohibits providers from charging the children's parents for any added cost of those accommodations. Thus, the added discrete cost of caring for special needs children goes unrecorded in providers' responses to market rate surveys. As a result, the gap between the cost and the price of child care for children with disabilities may be heightened. With no adjustment in the subsidy rate, providers would have little incentive to provide subsidized care to children with special needs because the reimbursement rate ceiling for CCDF subsidies is pegged to the price of child care for typically developing children and falls below the actual costs of providing the level of care required by children with special needs.

States have developed several methods to determine special needs rates (Cohen, 1999). In some cases, states multiply their subsidy rate by an adjustment factor. California, for example, multiplies the prevailing subsidy rate by a factor of 1.5 to establish the maximum rate for special needs care. Other states, including Vermont, add a flat amount to their standard subsidy rate. In some states, the practice is to reimburse actual costs based on individual documentation. Rather than administer a distinct special needs rate, Maryland allows for a rate adjustment of up to 15 percent above the prevailing subsidy rate, but only for accommodations above and beyond the "reasonable accommodations" required under ADA. At least one state, Pennsylvania, reimburses for special needs child care services based on the child's developmental age rather than chronological age. States may enhance their ability to adjust rates for special needs child care by supporting interagency collaboration and leveraging Title IV-B and IV-E child welfare monies in addition to CCDF funds.

The CCDF Final Rule leaves to each state the definition of which children are to be considered those with "special needs."

Tiered Quality Rating Strategies. A growing number of states have experimented with differential reimbursement rates tied to specific quality levels. Variously called tiered reimbursement or rated licensing, these differential rate structures place a quality enhancement mechanism atop a market-based subsidy system. These tiered quality rating systems usually use compliance with state licensing as a baseline measure of quality. States pursuing this approach establish some formal system for rating enhancements in quality upward from that baseline, and link the various quality rating levels to differential reimbursement rates. Twenty-two states, including the District of Columbia, have implemented some form of tiered quality rating strategy; another 16 are in various stages of considering or developing a tiered rating system (Collins, et.al., 2001).

In a tiered reimbursement system, a state pays higher subsidy rates to providers who achieve one or more levels of quality beyond basic licensing requirements. The reimbursement differential may be a flat dollar amount or a percentage above the subsidy rate. The quality standards generally combine a variety of approaches in defining each reimbursement level, including accreditation by a national organization and training or staff qualification provisions. Tiered reimbursement systems vary greatly. For example, Ohio pays five percent above the state subsidy rate for nationally accredited care, while Oklahoma, in its Reaching for the Stars program, has a four-tiered, multi-dimensional system-including teacher compensation, director and teacher training and credentials, and learning environment-capped by national accreditation at the highest reimbursement level.

While similar to tiered reimbursement strategies, which tend to reflect funding standards, rated licensing schemes tend to be based on licensing regulations. North Carolina has developed a "five-star" rated licensing system that ties higher reimbursement rates to standards set forth in state licensing, using provider education, program standards, and compliance history-but not accreditation-as benchmarks for reimbursement and licensing levels. (See Azer, 1999, and Gormley and Lucas, 2000.)

As with other differentials, tiered quality strategies present states with the choice of using either a flat rate supplement to the subsidy rate or a percentage enhancement above the subsidy rate. A fundamental question for states considering such an approach is whether to develop their own quality standards or to adopt those of one or more national accrediting bodies. In-house standards provide a state with more control in defining quality child care, but also add an administrative and resource burden in developing and monitoring standards, as well as added liability for the standards. Using an existing accreditation body's criteria for at least some of the tiers reduces a state's control over quality standards, but is a less costly option that eases development, implementation and liability burdens.

Supplemental Rate-setting Methods. Because rate surveys can only measure the price, and not the cost, of child care, state administrators seeking to fully take into account market conditions when establishing rate policies may want to supplement market rate survey data with other information. For example, in designing rate adjustments, states may wish to draw upon cost studies where and when they exist. Although there are few studies of the actual costs of providing child care, this is an area of inquiry drawing increasing attention. In rewarding quality, states can estimate how much it will cost a provider to meet an additional requirement and increase the subsidy rate ceiling by that additional amount, which may be a simpler task than determining the total cost of care. Basing deviations from the subsidy rate on actual cost information is in keeping with a market-based approach and has been employed as a reimbursement method for special needs differential in several states. (See Groginsky, Robison, and Smith, 1999.)

Payment Policies and Procedures. Perhaps the most under-recognized aspect of the rate-setting issue is the development and implementation of payment policies in state administrative rules and procedures. Understandably, state administrators invest enormous energy and resources in establishing market rate ceilings and rate adjustments that ensure equal access and hopefully quality as well. Once this process, involving social science research methods as well as the realities of state budgeting, is completed and a rate ceiling is established, however, this ceiling must be applied in the administration of child care subsidies.

For providers, the price of child care is the starting point in the process of turning child care services into operating revenue. Revenue from non-subsidized child care services is determined by multiplying the price by the number of children in care and the number of units of child care services provided-the formula is revenue equals price times number of child-hours/days/weeks.

In the case of the child care subsidy payment system, that equation is more complicated. Revenue is determined not only by the subsidy reimbursement rate and the number of child-units of care provided, but is also a function of the payment system rules and their implementation. Relevant rules include those concerning authorization, eligibility, copayment, absence policies, hours of care, and other factors, such as whether the provider is permitted to charge parents special fees greater than the subsidy reimbursement rate.

Eligibility Determination. How efficient the subsidized segment of the child care market is in translating a rate into a payment can have a dramatic effect on the ability of providers to keep their doors open or on their willingness to participate in the subsidy program.

One area that merits attention is initial eligibility determination. For non-subsidized families, entry into the child care market means selecting a provider and paying a fee, including a deposit in some cases. Supply and preference issues aside, this can be done quickly, perhaps in a day.

In the child care subsidy system, frequently providers are confronted on Friday afternoon with an anxious parent who is still awaiting eligibility determination but who needs child care for a job that starts Monday morning. The provider is forced to choose either to refuse to serve the low-income parent until eligibility determination is finalized, or risk providing services that may go unpaid if the family ultimately is determined ineligible. Both options threaten the efficiency of the market and the ability of the child care subsidy to support workforce development goals.

Streamlining and speeding the eligibility verification process can help quickly connect low-income parents with providers who have available slots. States may wish to consider extending a period of "presumptive eligibility" to a family based on information provided, but not verified, at the time of application. During that period, perhaps up to one month, providers can be assured of reimbursement if they agree to provide child care services to a low-income family that has applied for the subsidy. The presumptive eligibility period permits verification to be completed without interfering with the market transaction of fee-for-service or the achievement of workforce development goals. In one study, providers estimated that only 4.5 percent of children served while the provider awaited initial notification of eligibility ultimately turned out not to be eligible (Karolak, 2000).

Absence Policies. Other market practices related to revenue generation may need to be considered for states to craft a market-based approach to subsidized child care that provides low-income parents with access to a full range of providers. How closely state policies mirror market practices regarding absence days (caused by illness or other reasons), vacations, or school-breaks can result in a reduction in provider revenue-in effect a cut in the subsidy-relative to private-paying families.

Special Fees. Similarly, some providers charge special fees, in addition to their base rate, for services such as registration or enrollment, supplies, activities, transportation, and meals. If these fees are not included in the subsidy rate, they can pose serious barriers for low-income families and effectively limit their access to care offered by these providers. If a significant portion of providers in the child care market charges such fees, states may wish to consider establishing a policy that permits payment of special fees on behalf of the eligible family or possibly build this supplemental fee scheme into the subsidy rate structure.

Communication and Training. Whatever a state's policies and procedures, clear communication in the payment process and sufficient training of both providers and government staff are essential to the smooth functioning of the subsidized child care system. When government agencies review and adjust invoices before executing payment as a third party payer, they should clearly communicate to providers the reason for any adjustment in the payment. Miscommunication and misunderstanding will add to the administrative overhead-for providers as well as state and local administrators-of implementing the subsidy and may deter providers from accepting or continuing to accept children whose care is subsidized, thus threatening the ability of states to ensure equal access for low-income families to the full range of child care options.

Alternative Financing Strategies. This paper has focused on the steps involved in establishing a per child reimbursement rate for child care services. In most cases, these reimbursement rates will be administered as portable subsidies, that is a subsidy that is tied to a specific family or child and follows them to whatever program is selected. In child care, portable subsidies are typically referred to as child care certificates or vouchers. Portable subsidies are one way to help make child care affordable. But they are not the only way. Another approach is to provide financial assistance directly to early childhood programs. For example, Head Start and prekindergarten funds are typically administered as direct, institutional subsidies. Recently, increased attention has been directed to this subject at the state level (Stoney, 1998; Vast, 1998).

As states have become more sophisticated at conducting market rate surveys and establishing rate ceilings, they have begun to discover some difficulties with the market rate approach. Rates can only increase, for example, if the price of care in the market increases. Thus, providers that seek to offer high quality services, attract and retain qualified staff, and keep staff/child ratios high often find that they have no choice but to raise fees. Each time the provider raises fees, more and more private, fee-paying families are priced out of the market. Additionally, the link between the price of care and the public reimbursement rate can serve as an incentive for providers to segregate children and serve all subsidized families (if the state's reimbursement rate ceiling is high) or all private, fee-paying families (if the state's reimbursement rate is low). And basing rates on the price of care in a market also allows the norm to prevail. That is, if most programs are charging minimal fees and providing poor or mediocre services, the few programs that are seeking excellence will find it difficult-if not impossible-to survive financially if they serve a significant number of subsidized children (Schock and Daugherty, 2001).

One way to address these difficulties is to develop financing strategies that combine portable and direct subsidies. Direct, institutional subsidies can be made available to programs to ensure that they are able to provide high quality services at affordable fees. State initiatives that currently provide this type of direct subsidy include Wisconsin's Quality Improvement Grants program, Washington's Wage Ladder, California C.A.R.E.S. and North Carolina's W.A.G.E.S. Each of these initiatives provides financial support to programs (or, in the case of CARES and WAGES, directly to staff members) so that they can offer high quality care at affordable prices. But in each of these states, it is assumed that direct and portable subsidies will be combined. Early childhood programs derive revenue from a combination of direct institutional aid, portable child care certificates (for low-income families) and parent fees (for families who can afford to pay full price).

CONCLUSION

Establishing child care rate policies and subsidy administration are challenging areas of public policy. Conducting a market rate survey that accurately reflects market conditions and practices is not a simple exercise. States and localities undertaking a market rate survey must consider a number of issues related to conducting the survey and analyzing and using survey findings as a framework for child care policy.

The market rate survey is a valuable tool to achieve the ultimate program goal of ensuring that eligible parents have equal access to a full range of quality child care options. Rate-setting policies and payment system procedures can have an enormous impact on the quantity and quality of child care accessible in a market. States should establish subsidy rate ceilings that are high enough to permit parents this freedom of access in the marketplace. Depending upon the location and the type of care, this may move states to use cost studies or other information to develop differential rates or adjustments or to consider alternative financing strategies.

REFERENCES

Azer, Sheri L. Overview of States with Differential Reimbursement. Boston, MA: Wheelock College Center for Career Development in Early Care and Education, December 1999.

Blank, Helen and Poersch. Nicole Oxendine. State Developments in Child Care and Early Education, 1999. Washington, DC: Children's Defense Fund, February 2000. Available on the Web at http://cdfweb.vwh.net/childcare/99_state_dev.pdf.

"States Utilize Differentials for Reimbursement," Child Care Bulletin 18 (November/December 1997). Available on the Web at http://nccic.org/ccb/issue18.html#6.

Child Care and Development Fund: Report of State Plans for the period 10/01/99 to 9/30/01. Washington, DC: Child Care Bureau, Administration for Children and Families, Administration on Children, Youth, and Families, April 2001. Available on the Web at http://nccic.org/pubs/CCDFStat.pdf.

Cohen, Abby J. "Special Needs Rates: Supporting Inclusion of Children with Disabilities in Child Care Programs," unpublished research paper, January 1999.

Collins, Judy, Collins, Ray, and Dry, Tracy. "Tiered Quality Rating Strategies for Child Care," National Child Care Information Center, Vienna, VA (working draft, July 2001).

Gormley, Jr., William T. and Lucas, Jessica K. "Money, Accreditation, and Child Care Center Quality," The Foundation for Child Development Working Papers Series. New York, August 2000.

Groginsky, Scott, Robison, Susan, and Smith, Shelley. Making Child Care Better: State Initiatives. Washington, DC: National Conference of State Legislatures, October 1999. Available on the Web at http://www.ncsl.org/programs/cyf/mccexec.htm.

Schock, Lisa, and Daugherty, Jane. Rate Setting Policies: Ensuring Access and Improving Quality. Proceedings of the Child Care Bureau Issue Meeting. Child Care Bureau, Administration for Children and Families, Administration on Children, Youth, and Families, March 2001.

Stoney, Louise. Promoting Access to Quality Child Care: Critical Steps in Conducting Market Rate Surveys and Establishing Rate Policies. Washington, DC: Children's Defense Fund, August 1994.

Stoney, Louise. Looking into New Mirrors: Lessons for Early Childhood Finance and System Building. Horizons Initiative, October 1998. Available on the Web at http://www.nccic.org/pubs/mirrors/index.html.

Vast, Teresa. Learning Between Systems: Higher Education as a Model for Financing Early Care and Education. Minnesota Early Care and Education Financing Partnership, 1998. Available on the Web at http://nccic.org/pubs/lbs1998.pdf.

NOTE: There may be publications on this page that are available as PDF (portable document format) files. To be able to read these files, download the free Adobe Reader.

Visit our Comments Page and tell us what you think about our site!
Page Updated: March 26, 2007

Table of Contents

INTRODUCTION