urban institute nonprofit social and economic policy research

Recent Changes in Colorado Welfare and Work, Child Care, and Child Welfare Systems

Read complete document: PDF

PrintPrint this page
Share on Facebook Share on Twitter Share on LinkedIn Share on Digg Share on Reddit
| Email this pageE-mail
Document date: October 01, 2001
Released online: October 01, 2001

Assessing the New Federalism State Update No. 9

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

About the Series

This state update is a product of Assessing the New Federalism, a multiyear project to monitor and assess the devolution of social programs from the federal to the state and local levels. Alan Weil is the project director. The project analyzes changes in income support, social services, and health programs. In collaboration with Child Trends, the project studies child and family well-being.

In 1996 and 1997, the Urban Institute conducted case studies in 13 states that provided a baseline for understanding changes emerging from welfare reform. This set of state updates describes changes occurring between 1996-97 and 1999-2000 based on a second set of case studies completed in 1999 and 2000. Programs covered include income support through the Temporary Assistance for Needy Families program, employment and training supports for low-income welfare and non-welfare families, child care, and child welfare. It also looks at interactions among these programs.


In the years before welfare reform, Colorado's Aid to Families With Dependent Children (AFDC) and Job Opportunities and Basic Skills (JOBS) programs focused on income support and education and training as means to achieve the long-term self-sufficiency of AFDC recipients. Accordingly, recipients were encouraged to participate in education and training programs with the idea that such activities would eventually lead to high-wage jobs. This philosophy prevailed in the years leading up to welfare reform, even as the federal government granted states more freedom through waivers to experiment with innovative welfare policies. Indeed, while other states used federal waivers to make their welfare programs more work-focused, Colorado's welfare demonstration project, the Colorado Personal Responsibility and Employment Program, continued to emphasize skill development through education and training.

In addition to the emphasis on skill development, Colorado's pre-reform welfare system was also characterized by strong supportive services. In 1995, for example, Colorado's spending per poor family on services such as employment and training, child care, and child support were significantly higher than the national average. Colorado's pre-welfare reform system was also characterized by the strong county role in the administration of social services. County agencies had primary responsibility for the delivery of income support, child welfare, and child care programs.

Colorado's welfare reform legislation, Colorado Works, significantly departs from the state's traditional emphasis on education and training while further embracing a decentralized administrative structure. Colorado's strong economy, low unemployment rate, and prevailing belief that welfare recipients should be required to participate in work activities in order to receive assistance helped to shift the focus of Colorado's welfare system from an education and training model to a "work-first" model. The two primary goals of Colorado Works are to develop strategies and policies to ensure that participants are in work activities as soon as possible, and to provide counties with increased responsibility for the design and administration of the Colorado Works program.

In June 1997, shortly before Colorado Works took effect, researchers from the Urban Institute and Child Trends visited Colorado to document and analyze how the state was preparing to implement the program, and more broadly, to describe the services in place to assist low-income families. Two years later, researchers returned to Colorado to examine changes to the state's social service system. This brief presents findings from the more recent visits to Colorado, and serves to update the 1998 report, Income Support and Social Services for Low-Income People in Colorado.

This report begins with a short profile of Colorado's demographic, economic, and political conditions and then moves into a discussion of different aspects of the state's social safety net. The next three sections offer more detail on specific programs and services. The first section describes Colorado's Temporary Assistance for Needy Families (TANF) program, Colorado Works, including the state's work-related component for TANF recipients. The second section covers the state's system for providing child care for both Colorado Works recipients and other low-income families. The third section describes the child welfare system, with particular attention paid to the interaction between child welfare and welfare reform. Each section discusses important policies affecting the pro-gram and the clients that the program serves, the program's administrative structure and general service delivery, and information on recent changes. The brief concludes with a summary of noteworthy trends across the different social policy areas.

The information presented in this report comes from in-person interviews and focus groups with relevant front-line program staff, as well as focus groups of welfare recipients and other low-income families. In addition, researchers conducted interviews with various state-level officials responsible for child care and child welfare to obtain a statewide overview of the system and to learn about new policy directions in these areas. Most interviews were conducted with program staff in Denver, the state's largest urban area. Thus, most of the information in this update paints a picture of the social service system in Denver, and does not necessarily reflect how social services are administered in other counties throughout the state. However, telephone interviews were carried out with child welfare administrators in 12 additional counties to get a sense of how the collaboration between child welfare and TANF was working at the county level; of how child welfare caseloads were affected by welfare reform; and of how financing had changed at the county level.1 In addition, the primary data collected from the interviews and focus groups is supplemented with reports produced by other research organizations and by Colorado state agencies.

To gather information for this update, researchers returned to Colorado three times during late 1999 and early 2000: September 1999 (child welfare interviews); November 1999 (child care interviews and focus groups); and June 2000 (Colorado Works and work-force development interviews). Telephone interviews with county child welfare administrators took place from February through April of 2000.

Social and Political Context

Social and Economic Conditions

In 1999, Colorado's population reached over 4 million, growing throughout the 1990s at a rate almost two and a half times that of the nation as a whole. Colorado's teen birth rate and the percentage of births to unmarried women are both lower than those of the nation overall (table 1). The state's population includes a significant share of Hispanics, but relatively small proportions of other racial or ethnic minority groups. More than one in six Colorado residents are of Hispanic descent (14.9 percent), with most being either native-born or naturalized citizens. Non-Hispanic blacks, however, make up fewer than 1 in 20 residents (4.3 percent). In addition, most of Colorado's population lives in metropolitan areas, despite the fact that most counties in Colorado are predominately rural. In 1996, only 16 percent of Colorado residents lived in non-metropolitan areas compared to 20 percent nationally.

TABLE 1. Colorado State Characteristics, 1999
  Colorado United States
Population Characteristics
Population (1999) a (in thousands) 4,056 272,690
Percent under age 18 (1999) b 26.3% 25.7%
Percent Hispanic (1999) c 14.9% 11.5%
Percent Black (1999) d 4.3% 12.8%
Percent non-citizen immigrant (1998) e 5.2% 6.3%
Percent non-metropolitan (1996) f 16.0% 20.1%
Percent change in population (1990-1999) g 23.1% 9.6%
Percent births to unmarried women 15-44 (1998) h 25.6% 32.8%
Percent births to unmarried teens 15-19 (1997) i 8.6% 9.7%
Birth rates (births per 1,000) females age 15-44 (1998) h 15.0 14.6
Birth rates (births per 1,000) females age 15-19 (1998) h 48.7 51.1
State Economic Characteristics
Per capita income (1999) j $31,546 $28,542
Percent change per capita income (1995-1999) j 16.0% 10.8%
Unemployment rate (1999) k 2.9% 4.2%
Employment rate (1999) l 83.0% 81.5%
Percent jobs in manufacturing (1998) m 10.1% 14.8%
Percent jobs in service sector (1998) m 30.2% 29.9%
Percent jobs in public sector (1998) m 15.7% 15.8%
Family Profile
Percent children living in two-parent families (1999) n 66.1% 63.6%
Percent children living in one-parent families (1999) n 21.3% 24.8%
Percent children in poverty (1998) o* 12.3% 17.5%
Percent change in child poverty rate (1996-1998) o* -16.3% -15.0%
Percent adults in poverty (1998) o* 9.0% 11.2%
Percent change in adults in poverty (1996-1998 ) o* -11.8% -10.4%
Governor's affiliation (2000) p Republican  
Party composition of Senate (2000) q 18D-17R  
Party composition of House (2000) q 27D-38R  

* 1998 National adult, national child poverty estimates show statistically significant decreases from the 1996 estimates at the 0.10 confidence level, calculated by the Assessing the New Federalism project, The Urban Institute.
Table 1 notes.

Colorado enjoys a strong economy that has outperformed the national economy for the past several years. Colorado's unemployment rate is significantly lower than the nation as a whole (2.9 percent compared to 4.2 percent) and the state's per capita income has grown significantly faster than the national average since 1995. The average per capita income of Colorado residents in 1999 was about $3,000 higher than the per capita income nationally. In terms of employment sectors, Colorado has a lower than average percentage of manufacturing jobs but similar percentages of service and public sector jobs compared to the nation overall.

The strength of the economy in Colorado has not translated into significantly higher revenues for the state. Colorado's revenue is subject to limitations imposed by the Taxpayers Bill of Rights amendment, passed in 1992, which stipulates that government revenues may not increase more than the combined percentage of population growth and inflation. In 1998 and 1999, the government of Colorado returned $563 million and $679 million of surplus revenue, respectively, to Colorado taxpayers. During the 1999 and 2000 legislative sessions, the Colorado legislature passed a series of laws outlining "refund mechanisms" for the surplus; among these was a state Earned Income Tax Credit.

Finally, Colorado compares favorably when common indicators of family well-being are examined in light of national averages and trends. Since 1996, adult and child poverty rates have decreased more rapidly in Colorado than in the nation as a whole. In addition, a slightly larger percentage of children in Colorado live in two-parent families compared to children nationally, and the percentage of children in poverty in Colorado is more than 5 percentage points lower than the national average (12.3 percent compared to 17.5 percent).

Political Framework

Divided government has been the norm in Colorado for the past several years. Republicans controlled both houses of the Colorado General Assembly throughout the 1990s, while a Democrat, Roy Romer, served as governor from 1987 until January 1999. Term limits precluded Romer from running for governor again in 1998. He was succeeded by Bill Owens, the first Republican to be elected governor in Colorado since 1974. During the first two years of Owens' term, Republicans controlled both the House (40R-25D) and the Senate (20R-15D). However, Democrats took control of the Senate (18D-17R) in the 2000 election and gained two seats in the House (38R-27D).

Owens has made tax relief, education, and transportation major priorities of his legislative agenda. With large budget surpluses in Colorado, Owens believes the increased income gained through tax relief is the best means of improving the well-being of Colorado families. In line with this belief, several laws enacted during the 1999 and 2000 legislative sessions permanently reduced taxes by $550 million. In addition to tax relief, Governor Owens has worked to increase funding for public schools and to accelerate funding for critical transportation projects. Since Owens took office, he has signed into law a number of policies passed by the General Assembly directed toward improving the lives of children and families. These include bills to improve the quality and safety of child care,to expand funding for the Children's Basic Health Plan, and to expedite the adoption process.

Colorado's Social Safety Net

Colorado consistently ranks in the bottom half of the 13 Assessing the New Federalism focal states across a number of characteristics measuring the strength of the social safety net (table 2). Colorado's welfare benefit level has remained unchanged since 1989. At $357, Colorado's maximum benefit for a single mother of two with no income is significantly below the national median of $421, and ranks 9th among 13 focal states. Colorado also ranks toward the bottom among the focal states on the ratio of children receiving welfare to all poor children (8th) (a rough measure of welfare "coverage"), and on the percentage of children covered by health insurance (9th). Despite these poor rankings among the 13 focal states, Colorado's percentage of children without health insurance is lower than the national average (11.1 percent compared to 12.5 percent). In addition to the low benefit level and relatively small percentage of poor children covered by TANF, Colorado also has relatively low income thresholds for Medicaid/SCHIP and child care eligibility. With the maximum eligibility level at 185 percent of the federal poverty level (FPL) for both Medicaid/SCHIP and child care (equivalent to 57 percent of state median income for child care), Colorado ranks 12th and 7th, respectively, among the 13 study states on these factors.2

TABLE 2. The Safety Net in Colorado, in National Context
  Colorado United States
Welfare Benefits - Maximum Monthly Benefit
(Family of Three, No Income)
1996 (AFDC) a $356 Median: $415
1998 (TANF) a $356 Median: $421
2000 (TANF) a $357 Median: $421
Ratio of Children Receiving Welfare to All Poor Children
1996 (AFDC) b 44.4% 59.3%
1998 (TANF) b 32.7% 49.9%
Percent of All Children Without Health Insurance
1997 c 13.7% 12.2%
1999 c 11.1% 12.5%
Income Cutoff for Children's Eligibility for
Medicaid/State Children's Health Insurance
Program (Percent of Federal Poverty Level)
1996 d,e 94.4% 123.8%
1998 d,f 185.0% 178.4%
2000 d,g 185.0% 205.1%
Income Cutoff for Children's Eligibility for
Child Care Subsidy (Percent of State Median
Income/Federal Poverty Level)
1998 (January) h 58% / 185% 57% / 182%
1999 (June) h 57% / 185% 59% / 178%

Table 2 notes.

Caseload Dynamics

While welfare caseloads have been declining in Colorado since 1993, the rate of decline accelerated in the years after Colorado Works began. According to the interim report of a Colorado Works evaluation, caseloads declined by 22 percent in the last 18 months of the AFDC program (January 1996-June 1997), but declined another 39 percent in the 18 months after Colorado Works began (July 1997-December 1998).3 This decline has significantly outpaced the caseload decline nationally. For example, from January 1998 to June 1999, caseloads declined by 49 percent in Colorado compared to 33 percent nationally.4 Almost the entire decline in TANF caseloads in Colorado comes from one-parent TANF cases, effectively increasing the proportion of two-parent and child-only cases remaining on the rolls. In June 1999, child-only cases represented 36 percent of all TANF cases.5 Trends in the City and County of Denver, the focal county of this report, mirrored these state trends.

Food Stamp participation has also been on the decline in Colorado. Between 1996 and 1999, the number of households participating in the Food Stamp program fell by 25 per-cent. This decline is slightly smaller than the decline for households nationwide (27 percent).

Welfare and Work

When the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) replaced Aid to Families with Dependent Children with the Temporary Assistance for Needy Families program in 1996, the Colorado General Assembly enacted Colorado Works to serve as the state's TANF program. Colorado Works began in July 1997, replacing both Colorado's AFDC program and the Colorado Personal Responsibility and Employment Program, a five-county welfare demonstration project initiated in 1994 under a federal waiver. The waiver project, like the state's JOBS program, emphasized education and training as a means to increase employment and earnings. Colorado Works represented a major shift in programmatic emphasis—from education and training to a "work-first" model focused on moving recipients into available jobs as quickly as possible. While the waiver project ended before it could be evaluated, the design of the Colorado Works program was influenced by two lessons learned from the state's waiver experience: (1) welfare recipients need a support system of services in order to stay employed once they have jobs; and (2) moving families from welfare to work is expensive and requires a long-term commitment.

In addition to a new "work-first" emphasis, Colorado Works also made significant changes to the administrative relationship between the state and county governments. Before Colorado Works took effect, cash assistance in Colorado was state-supervised and county-administered, with counties having both fiscal and administrative responsibilities for the AFDC/JOBS program. Unlike many other states, Colorado used welfare reform as an opportunity to devolve authority for much of the design and implementation of TANF to the county level. With the exception of eligibility rules, benefit levels, and time limits, Colorado Works made counties responsible for most aspects of their TANF programs, including the design, implementation, and functioning of employment and training services, and the choice of whether to operate a diversion program.

Colorado Works Policy and Program Emphasis

Colorado's TANF program puts a strong emphasis on moving TANF participants from cash assistance to work as quickly as possible. In line with the federal mandate, the state legislation required that counties assess all applicants for employability, and that recipients enter into an Individual Responsibility Contract (IRC). This contract defines the terms and conditions of assistance and focuses on the client's employment goals and responsibilities. However, because Colorado gives much discretion to the counties, it is difficult to define an overall program emphasis for the state. In fact, counties have so much discretion in Colorado that the state's TANF program has been described as "not one statewide program but 63 county programs with common eligibility rules and benefit levels."6 Below we highlight the characteristics of the TANF program for the City and County of Denver.

Local Colorado Works Focus

Denver has sought to develop a system of support that can continue to work with clients long after they have left TANF. In addition, Denver has sought to use its TANF funds fully by providing a wide array of support services and making them available to individuals who do not qualify for cash assistance. While Denver has definitely moved from an education and training focus to "work first" under welfare reform, the Denver program is not as harsh as TANF programs in many other states. For example, Denver allows TANF recipients to participate in numerous work activities beyond the federally approved activities, and has implemented a graduated sanction policy. In addition, the county has created a Family Counseling Program to address the mental health, substance abuse, and domestic violence needs of TANF recipients. Attending Family Counseling Program activities counts toward an individual's TANF participation.

Diversion. Diversion is an integral component of Colorado Works. The state allows counties the option of adopting two diversion programs. Under the first program, state diversion, counties may offer cash or in-kind services to TANF-eligible applicants who require only short-term aid to meet immediate needs. Under this diversion program, applicants receive a one-time lump sum payment in lieu of ongoing cash assistance. Under the county diversion program, counties may adopt a separate "nonassistance" program for families do who not qualify for basic cash assistance under Colorado Works.

The City and County of Denver offers both state and county diversion. TANF applicants have the option to request assistance through the state diversion program. Payments are typically less than the equivalent of three months of cash assistance; and recipients may only receive state diversion assistance three times in a lifetime. As in other states, the financial diversion alternative is used only rarely—in the first five months of 2000, an average of 12 recipients a month opted for state diversion.

Denver's county diversion program is called Working Family Assistance (WFA). WFA can provide cash, child care, housing, transportation, and other supportive services, and is available to both TANF- and non-TANF-eligible families. To qualify for WFA, a family must have a child under 18 and an income of no more than 225 percent of the federal poverty level. Except in unusual circumstances, WFA is provided no more than twice a year (with a three-time lifetime limit) and is limited to a maximum of $1,000 per request. In response to low utilization rates, the Denver Department of Human Services (DHS) made a decision to market the program heavily and increase accessibility in early 2000. For example, the office provided extended hours with the primary activity focusing on taking WFA applications. As a result, approved WFA applications have steadily increased from 10 in January 2000 to 60 in May 2000.

Eligibility. Eligibility determination is one of the few areas of Colorado's TANF policy where counties have no discretion. The state does not impose any pre-eligibility requirements for TANF applicants. Denver, however, opted to strongly urge applicants to complete work-related activities before TANF benefits are approved. Applicants in the City and County of Denver have 10 days in which they are asked to complete 22 hours of an activity. Activities may include such things as locating child care or addressing housing problems.

Work Activities. Colorado Works has a 24-month "work trigger." That is, adult Colorado Works recipients are required to engage in work-related activities within 24 cumulative months of receiving cash assistance. The state adopted federally approved work activities for the purpose of calculating county work participation rates, but allowed counties to include additional activities to satisfy the state's work activity time trigger.

Despite the seeming latitude of the 24-month work trigger, all new applicants in Denver are expected to engage immediately in work related activities. Denver's policy is to develop an IRC when clients first apply to TANF, listing the activities the client is required to complete and the timetable for completion. Activities begin immediately; and some must be completed before the client receives final approval for cash assistance. At the time of our visit, there were two types of work-related activities that counted toward the client's participation requirement—countable and allowable. Countable work activities were those that counted toward the state's Federal Participation Rate, including unsubsidized and subsidized employment, work experience, on-the-job training, job search and job readiness assistance, community service programs, vocational education training, job skills training, and other activities "designed to lead to self-sufficiency." TANF recipients were also able to participate in allowable work activities that were defined by the county and approved by the state. Allowable activities were considered "necessary to prepare TANF recipients for work" but did not count toward the county's federal participation rate. Allowable activities in Denver included orientation and assessment, mental health evaluation and treatment, substance abuse evaluation, domestic violence prevention therapy, parenting skills training, job search/job readiness activities beyond the six-week countable criteria, basic skills training, GED classes, and applying for needed services (e.g., SSI and housing).

Clients with children over six years of age must complete 134 hours of countable work activities per month to be included in the state participation rates. Those with children under the age of six years must complete 96 hours of countable activities per month. However, clients may meet Denver's work requirements with a mixture of countable and allowable work activities.

Exemptions. There is no state exemption from work activities based on the age of the youngest child. However, counties have the option to set a youngest child exemption. Counties may also opt to set additional exemptions from work activities for circumstances such as mental or physical disabilities, caring of a disabled child, or homelessness. In Denver, participants are exempt from work activities if they have a child under six weeks of age. Case managers may also exempt clients for other reasons, but in cases of homelessness or disability the IRC may be written to include steps the client is required to take to address the problem (e.g., apply for SSI).

Sanctions. Colorado has a three-tiered sanction policy for noncompliance with work activities, failure to cooperate with the Office of Child Support Enforcement, or failure to obtain necessary child immunizations. Sanctions become progressively steeper with repeated occurrences of noncompliance. For the first occurrence, 25 percent of the cash grant is cut until the case becomes compliant, for a minimum of one month but not more than three months. The second occurrence results in a 50 percent reduction in cash assistance for a minimum of one month or a maximum of three months. The third occurrence results in the cessation of all cash assistance for three to six months or in closing the case completely. Once a client reaches a third sanction, all subsequent sanctionable offenses are at the third sanction level.

Individual counties are allowed to develop their own sanction policies within state parameters. Under Denver's sanction policy, the first offense results in a 25 percent decrease in cash assistance for one month. If the case does not become compliant within one month, the second level of sanction—a 50 percent reduction in benefits—is imposed for one month. If the sanction is still not "cured" at the end of month two, a level three sanction—a 100 percent decrease in benefits—is imposed for a minimum of three months. The sanction stays in place until it is cured. The two- and five-year time clocks continue to tick throughout the first and second sanction period, but not the third.

A 1999 class action lawsuit alleging that the Colorado Department of Human Services (CDHS) and several counties, including Denver, had unlawfully reduced or stopped TANF cash assistance benefits without sufficient written notice has resulted in several changes regarding sanctions and other case closures. As a result of the settlement, all existing sanctions were erased. New presanction warning letters and detailed procedures were developed to ensure that all clients are well aware of their rights and the steps involved in sanction and case closure. The county's settlement led it to establish a Customer Service, Quality Assurance, and Quality Control Team. The team is charged with reviewing each TANF case to ensure that it was handled correctly according to the action taken—especially in case closures. The staff examine all cases that have a level three sanction imposed. They also examine cases that are closed as a result of the 24-month time trigger to make sure that clients were given multiple opportunities to participate in activities.

Organization of Welfare and Work Programs

State Level. Colorado Works is a state-supervised, county-administered program. At the state level, Colorado Works is overseen by the Office of Self-Sufficiency in the CDHS. CDHS has overall responsibility for all social service programs, including TANF, food stamps, child welfare, child care, child support enforcement, youth corrections, mental health services, vocational rehabilitation, alcohol and drug abuse, homeless programs, and adult and veterans services (table 3). A Welfare Reform Task Force was created in 1999 to examine the implementation of welfare reform in Colorado and develop recommendations regarding ongoing program implementation.

TABLE 3. Administration of Income Support and Social Services in Colorado
Federal or generic
program name
What the program is called in Colorado Which agency administers the program in Colorado
TANF Colorado Works Colorado Works Division, Office of Self-Sufficiency, Colorado Department of Human Services (CDHS)
Workforce Investment Act Workforce Investment Act Colorado Department of Labor and Employment (CDLE)
Food Stamps Food Stamps Food Assistance Programs Division, Office of Self-Sufficiency, CDHS
Child Care Development Block Grant Colorado Child Care Assistance Program (CCAP) Child Care Division, Office of Children, Youth, and Families, CDHS
Child Welfare Child Welfare Child Welfare Division, Office of Children, Youth, and Families, CDHS
Medicaid Medicaid Colorado Department of Health Care Policy and Financing

In 1994, the Colorado Workforce Consolidating Council (CWCC) was established to identify areas for reform to move toward the goal of consolidated employment and training services for all Coloradans. As with the TANF program, the state's intent is to facilitate responsive, locally-planned and -directed employment and training programs. In consultation with CWCC, the Colorado Department of Labor and Employment (CDLE) set policy, developed performance standards, and allocated funds for One-Stop Career Centers. While counties were heavily involved in providing job training services, their involvement increased through changes adopted in 1997, which included consolidating the Job Training Partnership Act (JTPA) and Job Service Programs with nine Workforce Development Regions.

At the state level, Colorado has made a number of changes in response to the Workforce Investment Act (WIA). In September of 1999, the governor issued an executive order creating the Colorado Workforce Development Council that replaced the CWCC. A majority of the voting membership of the Council consists of business representatives as is required by WIA. This contrasts with the CWCC, which had business representatives but was dominated by public sector appointees. The Workforce Development Council sets general policy and provides resources, while the counties have considerable flexibility in how they administer the programs within the guidelines established at the state level. The Office of Workforce Development was created within the governor's office to provide staff support for the Workforce Development Council and to advise the governor regarding matters related to workforce development. The CDLE, which served as the administrative entity under JTPA, continues to act in that function under WIA and has direct responsibility for overseeing federal funding.

State-Local Relationships and Local Administration of Colorado Works. Although CDHS monitors compliance with state and federal regulations, much of the authority over TANF policy, program design, and implementation resides with each of Colorado's 63 counties. TANF program elements that apply to all of these counties include cash grant levels; a five-year lifetime limit for cash assistance; a 24-month trigger for participation in a work activity; and basic eligibility criteria, including cooperation with child support enforcement and child immunizations.

When Colorado Works began, the state gave county commissioners the authority to set county-level TANF policy. Because the City and County of Denver does not have a board of county commissioners, the mayor created a Welfare Reform Board to serve in this capacity. Board members are mayoral appointees and include representatives from the mayor's office, Denver Department of Human Services, and community agencies. A Colorado Works recipient also serves on the Board. The Welfare Reform Board was responsible for overseeing the implementation of welfare reform for the City and County of Denver. The Board continues to set direction and policy, and makes recommendations regarding how TANF funds should be used. The city council and the mayor's office approve the final budget for Colorado Works in Denver.

In the City and County of Denver, the local human services agency is called the Denver Department of Human Services. DHS is a city/county agency accountable to the mayor. The Family Employment Resources Division of the Denver Department of Human Services is responsible for determining eligibility and providing benefits for TANF; Medicaid; child care; food stamps; and homeless, burial, and general assistance programs. DHS has a central office that administers these programs. Three satellite offices accept applications and provide ongoing case management and services, though clients are usually required to attend orientation at the central office.

Local Administrative Model. When TANF first began, the Denver DHS adopted a new, integrated staffing model in which eligibility workers assumed both eligibility and assessment responsibilities and were reclassified as case managers. Staffing and allocation of responsibilities have been modified several times since the implementation of Colorado Works. At the time of our visit, Family Assessment Workers conducted intake and eligibility, completed preliminary assessments, and developed initial IRCs.7 Once cases were approved, clients were transferred to ongoing case managers who conducted additional testing and assessment as needed, modified the IRC when necessary, and made referrals to appropriate services. The Mayor's Office of Employment and Training (MOET) staff had primary responsibility for TANF employment services as described below.

The Denver DHS continues to explore ways to configure staff responsibilities and functions most effectively. For example, in an effort to improve identification of client barriers to employment among hard-to-serve clients, at the time of our visit, DHS was planning to bring senior social workers on board to conduct reassessments of ongoing TANF cases to address barriers to employment and other issues. The addition of these social workers arose out of concern that many of the Colorado Works recipients who are nearing their five-year time limit may not have received a thorough initial assessment that adequately identified barriers and offered appropriate services when they originally applied for TANF.

The Local Administration of Workforce Development Services. In the City and County of Denver, the Denver Workforce Development Board oversees the use of employment and training funds received through JTPA/WIA, Wagner-Peyser, and U.S. Department of Labor Welfare-to-Work grants. The Mayor's Office of Employment and Training is responsible for administering these programs. In response to the mayor's desire to have all city/county workforce development services under one agency, responsibility for the employment component of TANF was transferred from DHS to MOET in 1999.

The primary service delivery system for supporting employment and training activities is the Denver One-Stop Career System. This system began in 1997 using federal one-stop implementation funding. Eight One-Stop locations throughout the City and County of Denver provide services to both employers and job seekers (including job placement, training, and Adult Basic Education). Services available through the One-Stops and their network of community based service providers include: the Denver Youth Opportunity Grant, Economic Dislocation and Worker Adjustment Assistance, Targeted Assistance Grant (for refugees), Welfare Employment Contractors (both TANF and Department of Labor Welfare-to-Work programs), and the Youth Offender Demonstration Grant. The Job Service, Veterans' Employment Service, DHS, and the Community College of Denver are involved in various One-Stop operations.

At the time of our visit, MOET staff provided up-front workforce development services (e.g., orientation, job listings, initial screening for training eligibility, job search skills, job referral and placement, testing and assessment, and a resource room) while extended training programs were offered by a large network of community-based organizations under contract to MOET. Thirty-three contractors provided additional, more intensive employment and training services. Three of the One-Stops were primarily serving Colorado Works recipients, one focused on youth employment programs, and one was largely an Employment Services Job Service site. Few contractors are physically located at the One-Stops, although a "no wrong door" policy provides linkages to other partner offices and agencies through staff referrals.

Denver's workforce development system has experienced considerable upheaval and change, both structurally and programmatically. This is due in large part to (1) the consolidation of administrative responsibility for the majority of employment and training pro-grams under MOET, and (2) the transition from JTPA to WIA. For example, the implementation of WIA has meant a shift in program and service priorities for MOET and its contractors, from the JTPA emphasis on training to the work-first model endorsed by WIA.

To better fulfill its new role and meet the overarching goals of a streamlined, nonduplicative employment and training system, MOET was considering a significant reorganization at the time of our site visit in June 2000. One option under consideration was to move MOET away from an organizational structure based on funding streams, toward a more horizontal and integrated organization that blends program staff and services based on the WIA-mandated concepts of core services (e.g., eligibility determination, outreach, assessment, job search and placement, and labor market information) and intensive services (e.g., specialized assessment, counseling, case management, and training). In addition, MOET planned to further integrate WIA activities into the TANF offices and expand services to the working poor and other non-TANF populations in all One-Stops.

Colorado Works and Workforce Development—Service Delivery and Linkages in Denver

MOET and DHS have a long history of partnering and coordination on behalf of welfare recipients. MOET had served some welfare recipients through contracts funded under the JOBS program. In addition, when AFDC ended, the proportion of JTPA recipients who were receiving cash assistance was reported to be around 60 percent. When responsibility for TANF employment services initially moved from DHS to MOET, MOET assumed only a management role for DHS employment staff. Eventually, DHS employment staff became MOET employees; and space located within the DHS central office was reorganized and expanded to become a One-Stop Career Center.

Employment and supportive services for TANF clients are coordinated through the IRCs developed between clients and DHS caseworkers. The first IRC is developed during an initial intake meeting with the DHS Family Assessment Worker, and contains all the tasks the client must accomplish to be eligible for and to remain on cash assistance. The IRC is revised as needed throughout a client's time on assistance. Following eligibility determination, cases are transferred to a DHS case manager who is responsible for assigning the client to a work activity as well as ongoing client eligibility and other facets of case management. A Family Counseling Program within DHS provides mental health, substance abuse, and domestic violence counseling services to TANF recipients.

The TANF employment and training services provided by MOET staff, many of whom are simply former DHS staff, are essentially the same as those provided before the administrative change. There are three primary employment programs designed to help TANF recipients move to work. Many non-job-ready TANF clients are referred to Group Services, a two-week job readiness program designed to assist clients in developing independent thinking skills, self-esteem, problem solving, and other work-oriented skills. Job-ready clients are often referred to industry-specific Worker Development activities. Additional services, including job leads, labor market analysis, a resource center, job club, job development, and GED classes, are available to the general public.

As the lead administrative agency for all workforce development activities in the City and County of Denver, MOET contracts with an overlapping, but not necessarily identical, set of service providers to provide JTPA/WIA and TANF services. A single nonprofit organization may offer contracted services under several different funding streams and pro-grams. At the time of our visit, MOET staff and 22 contractors were providing services to TANF recipients. In addition, MOET in-house staff administer Wagner-Peyser-funded Employment Services such as labor market information exchange and unemployment insurance.

As with TANF Worker Development activities, Denver's Welfare-to-Work program, funded through the Department of Labor, is also industry-focused; the five targeted industries are manufacturing, construction, hospitality, retail, and financial and office services. The program has a strong postemployment focus, and is considered to be an integral component of employment-related services provided to TANF recipients. In fact, Welfare-to-Work eligibility determination is always completed at the time of TANF application. This has let Denver avoid problems experienced elsewhere with identifying individuals eligible for Welfare-to-Work. However, as with Welfare-to-Work programs across the country, there are still far fewer people using the program than originally anticipated. Recently the Welfare-to-Work program has put an increasing focus on trying to recruit noncustodial parents, but Denver has struggled to locate individuals willing to participate.

At the One-Stop career centers, MOET directly provides what are defined as core services under WIA. Previously, under JTPA, contractor staff provided services that they deemed appropriate for individual customers. However, at the time of our visit, training vouchers were being provided to individuals under WIA. This is a significant shift from the contract-for-services nature of JTPA. Individuals may receive training vouchers worth up to $3,000, though the limit can be increased in exceptional circumstances. The state is responsible for judging whether providers can be on the approved list for Individual Training Accounts (ITAs), and is still in the process of setting standards.

Program Innovation and Challenges

Colorado Works has presented both significant implementation challenges and the opportunity for program innovations. At the local level in Denver, many of the implementation challenges during the first few years were associated with organizational and staffing changes, including expanding the role of eligibility workers, moving the TANF work component to MOET, and integrating additional services and supports into the new "work-first" program model. Denver's workforce development system has also undergone significant change in terms of organization and policy direction.

One of the reported benefits of shifting responsibility for TANF employment services from DHS to MOET is that it creates a seamless, citywide workforce development service delivery system that can be accessed by all residents. However, this shift has also created a new set of coordination challenges, including a need to define clearly the division of responsibilities between DHS TANF workers and MOET staff. For example, MOET and DHS staff reported having had difficulties coordinating modifications to clients' IRCs. DHS case managers sometimes bypass the on-site job readiness components (i.e., Group Services and Worker Development activities) and refer clients directly to contractors, even though MOET staff has primary responsibility for TANF employment and training services. Yet MOET staff indicated that these up-front job readiness activities are critical to client success in later employment and training activities. In addition, new contracting arrangements needed to be put in place for vendors providing TANF employment and training services.

Denver is continuing to adjust its TANF cash assistance program to the changes resulting from welfare reform by using TANF funds to create a broader set of services for TANF clients and other low-income families. The City and County of Denver has sought to fully utilize its available TANF funds to provide services to eligible clients. As a result of the decline in TANF caseloads and limited use of services, it became apparent that Denver would have a significant TANF surplus going into the 2000-2001 fiscal year. At the time of our visit, the Welfare Reform Board had just recommended allocating approximately $3 million of the estimated $7.6 million TANF surplus for a variety of activities in the Denver area. This was the second time in a year the board acted to allocate surplus funds. Recommendations for funding included housing and transportation services, child care resources and referral, MOET summer youth programs, dental services, and skills enhancement for working families. The largest single allocation was $5.9 million to provide housing assistance to TANF clients. These funds were needed to address the problem of soaring housing costs resulting from the strong economy and Denver's growing population.

Cash incentives are provided to TANF recipients in order to encourage their participation in countable work activities. Clients can receive anywhere from $200 to $400 extra if they complete their work requirement, effectively increasing the TANF grant for these clients. There was some disagreement among Denver Welfare Reform Board members regarding the use of cash incentives for TANF clients and working poor households. Some policymakers in Denver thought these resources would be better spent on supportive services and training; so the issue of cash incentives is likely to be revisited.

Denver also developed a number of other programs in the years since its transition to TANF. In 1998, DHS issued a Request for Proposals to identify and fund support services for TANF clients, including mental health, substance abuse, and domestic violence services. The Family Counseling Program grew out of this effort. The Family Counseling Program now contracts with 13 community-based organizations, five of which have staff on-site at DHS, to provide support services using TANF funds. TANF clients in need of such additional support may fulfill their work participation requirement by engaging in Family Counseling Program services. The program's services are also available to Working Family Assistance recipients.

In state fiscal year 1999, DHS allocated TANF funding for a kinship care program designed to address a growing child-only caseload. The initiative provides support for children receiving TANF cash assistance and living with nonparent relatives who have "care and control" of the dependent children. Additional support payments, as well as supportive services such as a clothing allowance, respite care, and counseling, are available to these families.

For parents in prison, "Children Doing Time" was designed as a one-year pilot program. This program will provide parenting classes, counseling, assessment for family reunification, and other activities for children in TANF child-only cases who are in the care of relatives and whose mothers are in the correctional system. The Denver Women's Correctional Facility agreed to support and participate in this effort. This project is targeted to begin in the fall of 2001.

DHS was also seeking to increase the use of TANF services by putting an increased focus on job retention and working poor families. Several policies had been modified to support families who are working, but not necessarily earning more than the poverty threshold. DHS renamed its county diversion policy "Working Family Assistance" to demonstrate that these services are meant not only as an alternative to regular cash assistance for TANF-eligible families, but they can also help the many working families who require short-term financial and other assistance. A Work Incentive Policy for former TANF recipients was being expanded to increase the value of cash incentives. These incentives provide payments of $500 to TANF recipients who have obtained work and have become ineligible for cash assistance payments; these payments are also provided for certain job retention milestones (i.e., $500 at six months and $1,000 after one year).

As these examples illustrate, the City and County of Denver is continuing to develop a mixture of specialized services that responds to the diversity and intensity of needs of different populations. Services are designed not just for mandatory work recipients, but also to meet the needs of working families and work-exempt populations. Administrators recognize an ongoing need to improve client participation, staffing, and policies to continue to meet the needs of changing caseloads.

At the time of our visit, MOET was still making changes as part of the WIA implementation process. MOET administrators were using WIA implementation as a mechanism to examine structural reorganization throughout their systems to create a single integrated employment and training system. In addition, the myriad community-based providers were having to adapt to new contracting procedures as well as funding and reporting procedures. MOET was experimenting with the design of its One-Stop Centers seeking to maximize the use of core services and to encourage interested clients to use intensive services.

Child Care

As welfare programs have shifted dramatically toward requiring recipients to work or engage in activities leading to work, child care is now a cornerstone of state efforts to support these activities. People leaving TANF because they have found employment, often referred to as transitional (for the period of transition off of welfare), also often need child care to make their transition a success. Though PRWORA eliminated the requirement that states provide child care assistance to these families—by eliminating any entitlement to child care for them—most states continue to give these families a high priority for child care subsidies. This study examined the ways in which TANF and post-TANF families gain access to child care subsidies. We studied nonwelfare working families as well, since they also need child care but often cannot afford it, and many of the states in this study find themselves in the situation of having to make choices between providing subsidies to TANF clients or to nonwelfare working families.

In Colorado, child care subsidies are available to all TANF recipients who are engaged in an approved work activity. In addition, the state also provides transitional child care subsidies to recipients who leave TANF because they have found employment. This assistance is available until the family's income exceeds the eligibility limit.

When PRWORA passed in 1996, then-Governor Romer thought of child care as the foundation upon which to implement welfare reform in Colorado. Thus, welfare reform provided the impetus for major changes to Colorado's child care subsidy policy. For example, while policies relating to child care were defined by the state before Colorado Works, that legislation amended state policy to allow counties to define their own child care income eligibility levels (within state parameters), and to allow flexibility in the amount that counties pay providers. As a result, income eligibility cutoffs for child care assistance and provider payment levels vary significantly from county to county in the state. Below we discuss these and other child care policies, and how they are administered in the state.

Child Care Eligibility and Assistance

Colorado state law asserts that counties must provide child care assistance to families earning less than 130 percent of the FPL level, but gives them the option of providing subsidies to families with incomes as high as 185 percent of FPL. Therefore, eligibility cutoffs in the state for a family of three can vary anywhere from a monthly income of $1,504 (130 percent of the FPL) to $2,139 (185 percent of the FPL).8 Currently, no county in Colorado sets its eligibility level as low as 130 percent of FPL—eligibility levels range from 144 percent of FPL (Custer County) to 185 percent of FPL. Forty-three of the 63 counties in Colorado set their eligibility cutoff at or near 185 percent of FPL.9

Administrative Structure and Funding

Child care assistance is provided through a state-guided, county-administered subsidy program. At the state level, since 1989 the Division of Child Care within the Colorado Department of Human Services has been the lead agency responsible for providing child care assistance to welfare and other low-income families. In addition to managing the child care assistance program, the Division of Child Care has been in charge of licensing and monitoring child care facilities, and has planned and implemented public policy on issues related to child care. County departments of social services administer the child care assistance program and play an important role in child care policymaking, setting eligibility levels and reimbursement rates. The increased county flexibility has resulted in cross-county variation on both of these dimensions of child care policy. A major exception to county flexibility is the parental fee, which continues to be defined at the state level and is therefore uniform throughout the state.

Funding for child care has increased in Colorado since welfare reform. In state FY1999, Colorado spent approximately $70 million on child care subsidies—more than double the $32 million spent in state FY1996. The Colorado Child Care Assistance Program (CCAP) was funded at $62.4 million in state FY1999, with an additional $10.5 million being transferred from TANF to cover cost overruns in certain counties. While federal law allows states to transfer as much as 30 percent of their TANF block grant to pay for child care, this transfer represented only 7 percent of Colorado's TANF grant.10

Child Care Fees and Reimbursement Rates

In Colorado, all employed families must pay a parental fee beginning with their first full month of employment. Families on welfare who are not employed but are in an education or training program or work activity are exempted from the parental fees. This is also true for teenaged parents attending high school, who draw from the low-income component of the child care assistance program.

The parental fee is based on family income in relation to the federal poverty level, family size, and the number of children in care. Depending on a family's income, the fee can represent anywhere from 6 to 11 percent of its monthly income. In times of hardship, however, counties may reduce parental fees to as little as $5 per month for a period of three months.

Many counties in Colorado have recently changed their reimbursement rates to reflect the results of a new market rate survey conducted in 1999. While a state agency—the Division of Child Care—conducts the market rate survey and publishes rates that reflect the 75th percentile of the market rate, county departments of social services have the flexibility to set their own rates.11 Therefore, unlike other states, reimbursement rates vary from county to county in Colorado. These rates tend to be higher in the Denver metro area and resort counties compared to the rest of the state.

Every county in Colorado uses a three-tiered reimbursement structure based on the type of care used. Licensed child care centers almost always receive the highest reimbursement rates, followed by licensed family homes and license-exempt providers. Rates also vary for full-time care (defined as five hours or more per day) and part-time care, and in almost all counties are higher for infants and toddlers than for children age two and older. In addition, Colorado allows a number of "alternative rates" for different types of specialized care including "mildly ill" care and extended hours care.

In most counties, including the City and County of Denver, providers are paid retrospectively and receive the payment directly from the county.12 The state relies on a voucher system that pays the provider on a monthly basis two to three weeks after a child care attendance and billing form is submitted. In most counties, this payment is made by direct deposit into the account of the provider. Parents, however, must pay their fee directly to the provider in all Colorado counties.

Other Forms of Child Care Assistance

The Child Care Assistance Program is only one of a number of programs supporting the early care and education of Colorado's children. The Colorado Preschool Program supplies school districts with the funds to provide half-day services to four- and five-year-old children who have risk factors that relate to child development. School districts in Colorado are also mandated to serve three- and four-year-old children with special needs. Finally, the Head Start program provides funds to local public or nonprofit agencies designated as Head Start grantees to provide early childhood education and other services to three- and four-year-olds.

Program Innovations and Challenges

One major child care challenge facing Colorado policymakers is a low subsidy utilization rate among families who have left TANF. According to a survey of respondents who left Colorado Works in the first three months of its implementation, only 14 percent of these early welfare leavers reported receiving a government or employee sponsored child care subsidy.13

The ability of counties to set their own eligibility levels and reimbursement rates has had a number of specific implications for families receiving child care assistance. Because counties have different eligibility and reimbursement policies, respondents noted that parents who move from one county to another can become confused when faced with a new set of rules and procedures. In some cases, county eligibility policies may vary so much that the move might change a family's eligibility status.

The supply of child care providers is also a serious problem in Colorado. Respondents in Denver described the situation as a "provider crisis" caused by the state's strong economy. Colorado's economic expansion has helped more families to find work, increasing the demand for child care. At the same time, the economy has lured potential child care providers away from child care to more lucrative professions. As a result, respondents noted that child care centers have high turnover rates, and suggested that some centers are employing poorly trained staff.

To deal with some of the state's child care issues, Colorado has developed a Consolidated Child Care Pilot Program, which designated 12 communities to design "consolidated programs of comprehensive early childhood care and education services." The goal of the program is to allow communities to address their critical full-day, full-year child care needs in innovative ways not allowed by state law. Under the program, pilot communities may request waivers of state laws, rules, or regulations to consolidate funding sources into a seamless child care system and to ensure collaboration among stakeholders.

Child Welfare

Child welfare agencies seek to protect children from abuse and neglect. They may intervene in families in which such behavior is suspected; offer services to families or require that families complete service programs; and request that law enforcement or the courts remove children from their home and place them in state-supervised care if the children face imminent or ongoing risk of abuse or neglect in the home. Nationally, many policy-makers, researchers, and advocates have expressed concern that families who did not fare well under the new welfare requirements might be referred to child welfare agencies for child abuse or neglect. Thus far, however, welfare reform does not appear to have had a significant impact on child welfare caseloads in Colorado. However, welfare reform has significantly affected child welfare financing, and enhanced collaboration between TANF and child welfare staff in some areas.

In Colorado, child welfare services are county-administered and state-supervised by the Division of Child Welfare Services (DCW) within the Colorado Department of Human Services (DHS). Services are delivered through and administered by local departments of human services. This means that the state provides guidance and oversight, but counties have considerable decisionmaking authority over how to design and run programs to best meet local needs. Before the Child Welfare Settlement Agreement (CWSA)14 in 1994, the counties had greater flexibility in the types of services they provided. However, the CWSA mandated more statewide standards in the services that each county was to make avail-able. Counties still have the flexibility to provide some services directly or through a contract with other private or public providers.

Child Welfare Caseloads

Despite widespread concern in the state, thus far child welfare caseloads have not increased following welfare reform. In 1998, Colorado investigated allegations of abuse and neglect involving 39,141 children. In 1998, of the children investigated, 18 percent were found to be victims of maltreatment. This represented a decrease from 22 percent in 1996, and was lower than the national median of 30 percent. Colorado's victimization rate, 6.7 cases of abuse per 1,000 children, is lower than the national median of 11.5.

In 1998, 53,345 children were in foster care—a 1 percent increase since 1996. However, some counties in the state reported a decrease in the number of children in foster care while the number of referrals has remained relatively flat. State administrators attributed the continued overall increase in out-of-home placements to an increase in shelter placements and possibly to an increase in the number of caseworkers in the field. County administrators attributed the continued increase to greater dysfunction within families, more parental substance abuse, and greater numbers of severely emotionally disturbed children who cannot remain in the home. Counties reporting a decrease in the number of children in out-of-home placements attributed the decrease to an expanded ability to provide wraparound and prevention services, thanks to more money stemming from the CWSA. The number of children adopted through subsidized adoptions has significantly increased, from approximately 2,300 in 1995 to over 4,400 in 1999 due to changes in the types of post-adoption services offered, including the extension of Medicaid.

According to some county administrators, welfare reform has affected welfare families in ways that are not reflected in the caseload numbers. Some administrators stated that working clients' self-esteem has increased. Others mentioned that the increase in the coordination between child welfare and TANF staff has benefited clients; the two divisions are no longer requiring clients to meet competing demands.


The 1996 federal welfare reform legislation altered federal funding streams that many states have used to pay for child welfare services. The former AFDC Emergency Assistance program was eliminated, with the program's funds rolled into the TANF block grant. In addition, federal funds for the Social Services Block Grant were cut by 15 percent, and eligibility for Supplemental Security Income was defined more narrowly.15

In state fiscal year 1998, Colorado's total spending (federal, state, and local funds) on child welfare services was $242,245,638, a 27 percent increase from state FY1996.16 While federal funds decreased 16 percent during that time, the increase in funding was due to a 100 percent increase in state funds from state FY1996.17 State funding has been increasing since the CWSA in 1994.

According to state administrators, before welfare reform, Colorado was drawing down about $70 million in Emergency Assistance funds for family preservation services. However, the state has enhanced efforts to maximize Title IV-E funds as it has lost the ability to pull down Emergency Assistance.18 In 1998, federal funds represented 32 percent ($78,301,438) of the funding, while state and local funds represented 54 and 13 percent, respectively.

The CWSA significantly changed service delivery and resulted in increased staff resources to the child welfare system. Statewide, 300 to 500 new workers were added over the three years immediately following the CWSA. In 1998, a change in the structure of the state's personnel system resulted in child welfare workers becoming county employees. Child welfare workers were formerly state merit system employees, with the number of full-time employee (FTE) positions and costs associated with staff salaries being handled at the state level. With this change, many counties found themselves competing for staff due to the salary differences that now exist between neighboring counties. Staff are also taking early retirement. Both the competition and the retirements leave a gap in the number of front-line staff available, almost negating the influx of workers that occurred immediately following the CWSA.

Collaboration between TANF and Child Welfare Agencies

Many families receiving services from child welfare agencies also receive TANF cash assistance. These dual-system families may face competing demands. They must meet the new requirements imposed on TANF recipients in order to receive assistance, while at the same time they must meet case plan goals developed by child welfare agencies in order to keep their children or have their children returned to them. Despite the overlap in populations, historically there has been little formal collaboration between child welfare and welfare agencies.

On the state level, DCW initiated collaboration with the Colorado Works Division a year after welfare reform began. DCW originally believed that all TANF administrative and policymaking activities would be entirely the responsibility of the counties. Thus, the state DCW initially maintained somewhat of a "hands-off" policy. However, after realizing that TANF funds could be used for child welfare purposes, DCW was less hesitant to collaborate with the Colorado Works Division. Both divisions created liaison positions to promote collaboration and provide technical assistance to county child welfare and TANF offices. The liaisons travel as a team to the counties. However, no current research is being conducted at the state level regarding the impact of welfare reform on the child welfare system.

At the time of our visit, DCW was working on draft legislation to combine the Family Services Plan (FSP; the child welfare services plan) with TANF's Individual Responsibility Contract for dual system cases. In these cases, the FSP would be used in lieu of the IRC. Under new requirements, certain activities undertaken for the FSP are considered work-related; but DCW wants legislation to mandate that the FSP will override the IRC. The state child welfare agency is pushing for joint staffing on all dual-system cases.

Although the state of Colorado accepted the family violence option in its TANF plan, it gives no exemptions from work requirements or time limits for recipients involved with child welfare or experiencing domestic violence. However, they can be granted good cause for not participating in a typical work requirement by having counseling defined as the planned work activity. This option was also passed on to the counties.

Since child welfare in Colorado is county-administered, the extent and formality of collaboration between child welfare and TANF agencies varies across counties. Some of the counties we surveyed (El Paso, Pueblo, and Weld Counties) reported collaboration as formal as meshing the child welfare services plan and the TANF plan to limited collaboration in the form of information sharing around the cash grant when a child is removed from the home. A majority of counties reported that child welfare staff have received training on the changes brought about by welfare reform, and that staff have the ability to retrieve information on a client's welfare status directly through a data information system or through some other formal protocol. Denver administrators reported plans for collaboration with the local TANF agency to begin January 2000: TANF funds have been allocated for child welfare to establish two units of six to seven workers each to provide early intervention services and intensive reunification services.

TANF has also created the incentive for collaboration around specific types of cases such as kinship care. In Denver, additional cash assistance is now available to relative providers (child-only cases) through TANF as described earlier. Relative providers caring for children with no child welfare involvement are eligible for supportive services to assist them in preventing the placement of these children in foster care. This help comes in addition to incentive payments (additional cash assistance). Relative providers receiving child-only payments for children involved with child welfare are only eligible for the additional cash assistance. Case management and supportive services for these caregivers are provided by the child welfare agency.

Other Changes Affecting Child Welfare

In addition to changes prompted by welfare reform, other policy changes have impacted Colorado's child welfare system in the past few years. The governor created the Task Force on the Welfare of Children in 1999 after several high-profile child deaths in 1998. A report with recommendations was presented to the governor in January 2000. In addition, the death of several children at the hands of their foster parents led to legislative support for protecting parental rights. The impact of the Task Force's report is not yet known, although changes in service delivery and enhanced monitoring of foster care agencies are anticipated.

Along with the mandates of the CWSA, child welfare caseworkers must also be in compliance with the mandates of the 1997 Federal Adoption & Safe Families Act (ASFA). ASFA imposes strict time limits on achieving permanency for children in foster care. The documentation required by both directives has led to less time spent with clients and more time spent in the office preparing and completing mandated reports and checklists, according to caseworkers, supervisors, and administrators.

In the past few years, Colorado's mental health services system has moved to a managed care approach. According to caseworkers in Denver, mental health services are now more difficult to access, and there are limitations on what services are provided and how long they can be used. There was concern that with the permanency time frames mandated by ASFA, the limit on mental health services may negatively affect some parents' ability to reach the goals established in the service plan.

In 1997, managed care pilots in child welfare began in three counties, and Colorado received another IV-E waiver in the fall of 1999 to extend these pilots to an additional three counties. The pilots vary in the population they have chosen to serve, ranging from child-only kinship care cases to adolescent/delinquent cases. According to state administrators, cost savings from the initial three pilots have varied. In addition, outcomes from the initial pilots are not yet known.


Welfare reform affected both the program emphasis and administration of Colorado's welfare system. For example, Colorado Works is much more work-focused than Colorado's former AFDC program or its welfare demonstration project. From an administrative point of view, welfare reform prompted the state to devolve TANF policymaking authority to the counties. With the exception of benefit levels, eligibility criteria, and time limits, Colorado Works has given the counties in the state wide discretion in how they design and administer their TANF programs. Colorado Works gave similar authority to counties for child care policymaking. As a result, counties also have broad discretion in setting eligibility levels for child care subsidies and reimbursement rates for child care providers.

A notable trend in Colorado has been the way in which counties have used surplus TANF funds resulting from the dramatic decline in TANF caseloads in the state. As a result of the Colorado Works legislation, counties are able to use TANF funding in creative ways to fund a range of services for the state's low-income population. For example, the City and County of Denver is using TANF dollars to fund the county's diversion program—Working Family Assistance—which can help low-income non-TANF families in need of short-term financial assistance by providing housing, transportation, and other services. Several counties are also using TANF funds to provide cash incentive payments to former welfare recipients who meet specific job retention milestones.

Colorado is one of only two states in this study (the other being Texas) that allow counties to determine their own eligibility levels and reimbursement rates for child care. While this discretion is welcomed by the counties, respondents noted that parents who move from one county to another can become confused when faced with a new set of rules and procedures. In some cases, county eligibility policies may vary so much that a move might change a family's eligibility status. In terms of child care funding, the state has more than doubled its investment in child care since State FY1996, supplementing increases from the child care funding stream with transfers from the TANF block grant. As a result, few counties place families on waiting lists for child care assistance.

Finally, unlike TANF and child care policy, county control of child welfare has been subject to greater state oversight as a result of the CWSA. The CWSA limits county discretion by creating statewide standards and earmarking state money for certain "core" child welfare services such as family preservation. While county discretion has been limited by the CWSA, welfare reform has significantly enhanced the collaboration between TANF and child welfare staff. Because TANF funds can be used for child welfare purposes, the Division of Child Welfare and the Colorado Works Division of the Colorado Department of Human Services have created liaison positions in an effort to foster collaboration between the two divisions.

Despite the concern that welfare reform might increase referrals to child welfare agencies, referrals have remained relatively flat since 1996, and child welfare caseloads in Colorado have not increased significantly. In fact, among the cases that are investigated, the percentage of children found to be maltreated has declined. State funding for child welfare has increased 100 percent since 1996, with counties using this money to provide wraparound and prevention services to families. Counties experiencing a decrease in out-of-home placements have attributed the decline to these services.


1. Administrators from Adams, Archuleta, Chaffee, Cheyenne, Dolores, Douglas, El Paso, Jefferson, Otero, Pueblo, Sedgwich, and Weld counties were interviewed by telephone.

2. Eligibility for child care in Colorado is determined based on a family's income in relation to the federal poverty level. We have also converted the eligibility level to a percentage of the state median income (SMI) to control for differences in the relative wealth of a state when comparing Colorado to other states. It is important to note that counties in Colorado have broad discretion in setting eligibility levels for child care. The 57 percent of SMI/185 percent of FPL eligibility thresholds noted here represent the maximum threshold at which a county can set its eligibility cutoff for child care. A number of counties in Colorado set their eligibility threshold lower.

3. London, Rebecca A., Vincent M. Valvano, Jennifer Kuiper, Kendra Lodewick. 2000. Evaluation of the Colorado Works Program Interim Report on Caseload Characteristics, Program Eligibility, and County Policies. Oakland, Calif.: Berkeley Planning Associates.

4. Caseload data were obtained from the U.S. Department of Human Services, Administration of Children and Families, http://www.acf.dhhs.gov/news/tables.htm.

5. London, Rebecca A., Vincent M. Valvano, Jennifer Kuiper, Kendra Lodewick. 2000. Evaluation of the Colorado Works Program Interim Report on Caseload Characteristics, Program Eligibility, and County Policies. Oakland, Calif.: Berkeley Planning Associates.

6. London, Rebecca A., Vincent M. Valvano, Andrea Cann, Will Fischer, Todd Glasser, Jennifer Kuiper, Kendra Lodewick, Kristin Porter, Lee Williams, Jean Demmier, and Judy Krysik. 2000. Evaluation of the Colorado Works Program: First Annual Report.Oakland, Calif.: Berkeley Planning Associates (page XIV).

7. Effective December 1, 2000, the Family Employment Resources (FER) Division has updated this intake process. The new integrated staffing model consists of an FER Counseling team that performs a thorough and holistic psychosocial and financial assessment with applicants.

8. Dollar amounts are based on the 1999 federal poverty level—the level currently being used in the calculation of eligibility in Colorado.

9. Twenty-two counties set their eligibility level at 182 percent of FPL while 21 counties set it at 185 percent of FPL. See http://www.cdhs.state.co.us/childcare/eligibil.htm.

10. London, Rebecca A., Vincent M. Valvano, Andrea Cann, Will Fischer, Todd Glasser, Jennifer Kuiper, Kendra Lodewick, Kristin Porter, Lee Williams, Jean Demmier, and Judy Krysik. 2000. Evaluation of the Colorado Works Program: First Annual Report. Oakland, Calif.: Berkeley Planning Associates.

11. The Division of Child Care actually publishes rates for three different child care markets: urban, rural, and resort. The market rate survey is conducted to determine the rates charged by child care providers in the community. The maximum rate a state will pay is set at the 75th percentile of the market rate, which means that it will be sufficient to pay the rates charged by three-quarters (75 percent) of the providers in the community. States will pay the amount the provider charges to private paying parents or the maximum rate, whichever is less.

12. A few counties are experimenting with prospective payments.

13. London, Rebecca A., Vincent M. Valvano, Andrea Cann, Will Fischer, Todd Glasser, Jennifer Kuiper, Kendra Lodewick, Kristin Porter, Lee Williams, Jean Demmier, and Judy Krysik. 2000. Evaluation of the Colorado Works Program: First Annual Report.Oakland, CA: Berkeley Planning Associates (page 110).

14. The CWSA was a result of negotiations between DHS and the Colorado Lawyers Committee on behalf of children in foster care being served by the system. It was a means of making needed changes in the state's child welfare system.

15. For additional information about these changes see Geen, Rob, Shelley Boots, and Karen Tumlin. 1999. The Cost of Protecting Vulnerable Children, Washington, D.C.: The Urban Institute. Assessing the New Federalism Occasional Paper No 20.

16. Based on data provided by the state in response to the Urban Institute 1999 Child Welfare Survey. Percent change is not adjusted for inflation.

17. Colorado's use of TANF funds for former EA-funded services has declined by 100 percent (or over $13 million) from State FY1996, according to the Urban Institute survey.

18. Title IV-E is a federal funding stream solely used for child welfare purposes.

Table 1 Notes

a. U.S. Bureau of the Census. "State Population Estimates: Annual Time Series, July 1, 1990, to July 1, 1999." http://www.census.gov/population/estimates/state/st-99-3.txt. [Date accessed: October 2000.]

b. U.S. Bureau of the Census. "State and County Quick Facts." http://quickfacts.census.gov/qfd/states/08000.html. [date accessed: October 2000.]

c. U.S. Bureau of the Census, Population Estimates Program, Population Division. "States Ranked by Hispanic Population, July 1, 1999." http://www.census.gov/population/estimates/state/rank/hisp.txt. [date accessed: October 2000.]

d. U.S. Bureau of the Census, Population Estimates Program, Population Division. "States Ranked by Black Population, July 1, 1999." http://www.census.gov/population/estimates/state/rank/black.txt. [date accessed: October 2000]. National population percentages are as of July 1, 1999, from U.S. Census Bureau. "Resident Population Estimates of the United States by Sex, Race, and Hispanic Origin." http://www.census.gov/population/estimates/nation/intfile3-1.txt. [date accessed: October 2000].

e. Urban Institute estimates based on a three-year average of data in the 1997, 1998, and 1999 March Supplements to the Current Population Survey. A Joint Project Between the Bureau of Labor Statistics and the Bureau of the Census.

f. U.S. Bureau of the Census. 1999. Statistical Abstract of the United States: 1999(119th edition). Table 42. Washington, D.C.

g. U.S. Bureau of the Census. Population Estimates Program, Population Division. "Percent Change in Population for U.S. States: 1990 to 1999." http://www.census.gov/population/www/estimates/stmap03.html. [date accessed: February 2001].

h. U.S. National Center for Health Statistics. March 28, 2000. National Vital Statistics Report. Vol. 48, No. 3. Tables B, 10, and 19.

i. Urban Institute estimates based on two sources: (1) U.S. National Center for Health Statistics. April 29, 1999. National Vital Statistics Report. Vol. 47 No. 18. Table 10. (showing total number of births in each state and the U.S., 1997) (2) Child Trends. 1999. CTS. Facts At A Glance. Table 1. (showing number of births to mothers age 15-19 and percentage of teen births to unmarried mothers) Washington, D.C.: Child Trends, December.

j. U.S. Department of Commerce, Bureau of Economic Analysis, News Release September 12, 2000: 1999 State Per Capita Personal Income Revised. Table 1: Per Capita Personal Income, by State and Region, 1995-1999 [Dollars]. http://www.bea.doc.gov/bea/newsrel/spi0900.htm. [accessed: February 2001]. Note: Percent change calculated in inflation adjusted dollars by the Urban Institute based on the CPI-U as published in the Economic Report of the President. 2000. Table B-60. Washington, D.C.: U.S. Government Printing Office, February.

k. U.S. Department of Labor, Bureau of Labor Statistics. 2000. "State and Regional Unemployment, 1999 Averages." Released February 2000. http://stats.bls.gov/newsrels.htm. [date accessed: February 2001].

l. Zedlewski, Sheila. 2000. "Family Economic Well-Being." In Snapshots of America's Families II: A View of the Nation and 13 States from the National Survey of America's Families. Washington, D.C.: The Urban Institute.

m. U.S. Department of Labor, Bureau of Labor Statistics. 1999. Current Population Survey. Data extracted February 10.

n. Vandivere, Sharon, Kristin Anderson Moore, and Martha Zaslow. 2000. "Children's Family Environment." In Snapshots of America's Families II: A View of the Nation and 13 States from the National Survey of America's Families. Washington, D.C.: The Urban Institute.

o. Zedlewski, Sheila. 2000. "Family Economic Well-Being." In Snapshots of America's Families II: A View of the Nation and 13 States from the National Survey of America's Families. Washington, D.C.: The Urban Institute. Percent change in poverty rates (1996-1998) calculated by the Labor and Social Policy Center, The Urban Institute.

p. The National Governors' Association. 2000. "The Governors, Political Affiliations, and Terms of Office 2000." http://www.nga.org/Governor/GovMasterList.htm. [date accessed: February 2001].

q. National Conference of State Legislatures. 1999. "Directory of State Legislatures, Online. Partisan Composition of State Legislatures House and Senate." D indicates Democrat; R indicates Republican; I indicates Independent; other indicates Other; and V indicates Vacant. http://www.ncsl.org. [date accessed February 2001].

Table 2 Notes

a. U.S. Congress, House Committee on Ways and Means. 2000. 2000 Green Book. Table 7-7: Maximum Combined AFDC/TANF Benefits for a Family of Three (Parent with Two Children), July 1994-January 2000. Washington, D.C.: U.S. Government Printing Office. January.

b. Two Urban Institute calculations: (1) Average monthly number of AFDC 1996 recipient children divided by number of children in poverty 1996: U.S. Department of Health and Human Services. Administration for Children and Families. 1997. Characteristics and Financial Circumstances of AFDC Recipients FY 1996. Table 18. "Percent Distribution of AFDC Recipient Children by Age (October 1995-September 1996)"; (2) Average Monthly Number of TANF 1998 Recipient Children divided by Number of Children in Poverty 1998: U.S. Department of Health and Human Services. Administration for Children and Families. 1999. Second Annual Report to Congress August 1999. Table 9:22. "Percent Distribution of TANF Recipient Children by Age Group (October 1997-September 1998)." The numbers of children in poverty in 1996 and 1998 are Urban Institute calculations from the National Survey of America's Families II.

c. Kenney, Genevieve, Lisa Dubay, and Jennifer Haley. 2000. "Health Insurance, Access, and Health Status of Children: Findings from the National Survey of American Families." Snapshots of America's Families II: A View of the Nation and 13 States from the National Survey of America's Families. Washington, D.C.: The Urban Institute.

d. Based on three sources: (1) Urban Institute's TRIM [Transfer Income Model]; (2) Vernon K. Smith. 1999. "Enrollment Increases in State CHIP Programs: December 1998 to June 1999": Health Management Associates, July; (3) Urban Institute analysis of 1999 SCHIP [State Children's Health Insurance Program] Annual Report. Rules for 1996 and 1998 are policies in place the majority of the year. Rules for 2000 represent plans approved as of January 1, 2000.

e. In 1996, the thresholds represent the state Medicaid thresholds for poverty-related eligibility or AFDC-related eligibility. Higher thresholds for separate state-financed programs (such as in New York) are not representedhere.

f. In 1998, some states' thresholds represent Medicaid eligibility, and others are either Medicaid expansions or stand-alone programs enacted under the SCHIP legislation.

g. In 2000, all states covered at least some children through SCHIP; certain groups in some states are eligible only through Medicaid.

h. Data for 1998 are from January 1, 1998, compared with 1997 federal poverty level (FPL). Gina Adams, Karen Schulman, and Nancy Ebb. 1998. Locked Doors: States Struggling to Meet the Child Care Needs of Low-Income Working Families. Table A1. Washington, D.C.: Children's Defense Fund. March. Data for 1999 are from June 1999, com-pared with 1999 FPL. The 1999 data are calculated using state agency information and information from Helen Blank and Nicole Oxedine Poersch. 1999. State Developments in Child Care and Early Education. Table 1. Washington, D.C.: Children's Defense Fund. The Urban Institute calculated 1998 and 1999 U.S. median income eligibility level as a percentage of state median income (SMI) and FPL.

About the Authors

Jeffrey Capizzano is a research associate in the Urban Institute's Population Studies Center. His research efforts focus on welfare and child care policy. He has published a number of papers on the child care arrangements of preschool and school-age children and is currently working on an investigation of child care policy in 13 states.

Robin Koralek is a research associate in the Urban Institute's Labor and Social Policy Center. Her work has focused on a wide range of social welfare issues, including income support, employment and training, and services integration.

Christopher Botsko is a senior policy associate at Health Systems Research. His most recent research focuses on welfare reform, food stamps, the State Children's Health Insurance Program (SCHIP), and model programs for integrating mental and physical health care.

Roseana Bess is a research associate in the Urban Institute's Population Studies Center focusing on child welfare related issues. She was part of a team of researchers examining the potential impact of welfare reform on child welfare systems in the Assessing the New Federalism case studies.

The project has received funding from The Annie E. Casey Foundation, the W.K. Kellogg Foundation, The Robert Wood Johnson Foundation, The Henry J. Kaiser Family Foundation, The Ford Foundation, The David and Lucile Packard Foundation, The John D. and Catherine T. MacArthur Foundation, the Charles Stewart Mott Foundation, The McKnight Foundation, The Commonwealth Fund, the Stuart Foundation, the Weingart Foundation, The Fund for New Jersey, The Lynde and Harry Bradley Foundation, the Joyce Foundation, and The Rockefeller Foundation.

This state update was prepared for the Assessing the New Federalism project. The views expressed are those of the authors and do not necessarily reflect those of the Urban Institute, its board, its sponsors, or other authors in the series.

The authors would like to thank all of the government and program officials whose willingness to share their time and expertise have contributed to this state update.

Topics/Tags: | Children and Youth | Employment | Governing | Poverty, Assets and Safety Net

Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact publicaffairs@urban.org.

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

Email this Page