Chapter 5.6: Examples of a Regional Approach
The cost/benefit model examines the quantifiable costs and benefits of expanding transit opportunities to low-income job seekers. The cost of various assistance programs, and the success of Wisconsin's W-2 program in moving household into self-sufficiency, is directly linked to success in low-income households and assistance program participants finding or improving their jobs. If wages are depressed by lack of access to reasonable wage jobs, the costs to the State rise. The model, therefore, measured the directions and magnitude of changes in the State's balance sheet given changes in transit investment.
The model was designed to estimate the approximate changes in costs to the state, and to W-2 participants, as transportation options change. The model uses a host of causal relationships to estimate change given a set of changes from an approximate base case. The model thus demonstrates change in costs given changes in independent variables, rather than predicting actual costs.
Qualitative considerations and assumptions must be considered as well. For instance, the relationship between the number of job placements dependent on transit could be greatly affected by either W-2 participants' willingness to bear costs associated with increased travel times, or the willingness of employers to subsidize employees' transportation costs. The quantitative model thus cannot be used without applying the entire research model or some reasonable proxy.
State and Participant Tests
The BRW/Biko Associates quantitative model in Wisconsin estimated the fiscal effects of expanding or extending transit services in any Wisconsin urban area from two perspectives: Magnitude of savings available to the state, and effects on participants. Each urban area will have different inputs related to expected wage levels and wage differentials, numbers of cases in each job category, health care costs, child care costs, and transportation costs. The model's assumptions can be changed for each urban area in which the model is applied. The Participant and State test results are based on the causal and interactive relationships of a number of independent variables, such as assistance program benefit schedules and income tax rates, that effect state and participant costs.
Some welfare program parameters, such as baseline caseload estimates and job placement assumptions, change for each Wisconsin county. Local economic conditions also change, such as prevailing wage rates, availability of employer health benefits, and levels of transit dependence. As these variables change, the model assumptions can similarly be changed.
The Milwaukee study examined how total costs to the state changed, as the state was able to place higher numbers of W-2 participants in unsubsidized jobs. The costs examined included the following:
- Costs of subsidized jobs.
- Cost of health care.
- Cost of program administration.
- Childcare costs.
- Cost of income taxes (Earned Income Tax Credit, EITC).
What is the Baseline
The Milwaukee study was conducted before W-2 was actually implemented. The baseline caseload and cost figures, and some program parameters, used in the cost/benefit model were based on estimates produced by several agencies that had studied the planned implementation of W-2. The Wisconsin Legislative Fiscal Bureau (LFB) had estimated the costs of the program to the State based on the Workforce Development (DWD) planning assumptions. As these assumptions did not consider the spatial mismatch between jobs and W2 participants, the BRW/Biko model assumed that improved access was required in order to meet the DWD and LFB projections. The baseline was, therefore, the forecast costs assuming projections were achieved. The model measured the change in costs to the State from lowering the number of successful placements in unsubsidized jobs.
Modeling the Relationships
Using a variety of scenario analyses from the Participant perspective we can ascertain a range of savings associated with each of the following scenarios:
- Moving from a Transitional Placement to an unsubsidized job.
- Moving from a Community Service Job to an unsubsidized job.
- Moving from a Trial Job to an unsubsidized job.
- Health Care savings, varying the wage of the new job and the availability of employer health care benefits.
- Child Care savings, varying the wage of the new job.
- Stipend or job subsidy savings.
- Food Stamp program savings, varying the wage of the new job.
- Net tax effect, varying the wage of the new job.
Based on the transit deficiencies identified earlier in the report, the study assumed that half of the placements in unsubsidized jobs were dependent on transit that did not exist or was extraordinarily inconvenient or expensive (thus likely to result in loss of job). The study also assumes that without transit, these unsubsidized placements are likely to instead result in Community Service Job (CSJ) placements (50%), Trial Job (TJ) placements (20%), or in the participant dropping from the program (30%). Low-income workers are often better off taking a part-time or seasonal job than a CSJ; the combination of the tight labor market and the new federal minimum wage ensures that CSJs are generally not attractive to workers who can find part-time work. The job gap studies show that a fairly large number of part-time and seasonal positions are available in the metropolitan area. As a consequence, if full-time unsubsidized jobs are not available, the worker may choose to drop out of the program and work part-time or seasonal jobs. These workers may, however, still participate in the Childcare and Health Care components of W2.
Brown, Dane, and Wood Counties
The above method was consequently applied in three other counties in Wisconsin; Brown County, the location of Green Bay; Dane County, the location of Madison; and Wood County, the location of two small cities (Wisconsin Rapids and Marshfield). These analyses differed from the Milwaukee study in several ways:
- The population being examined was expanded from just W2 participants to include participants in the Food Stamp program. Quantitative modeling also estimated the savings from reducing the number of people dependent on Medicaid.
- Far fewer data were available on the transit dependence and employment barriers of the target population. In Milwaukee, an existing body of social science research provided a foundation for the quantitative analysis. In Wood, Brown, and Dane counties the study relied on more generalized statewide data and older Census data.
- For the working poor population (Food Stamp recipients in the workforce) the crucial independent variable was an increase in wages rather than a movement from unemployment (or subsidized job) to full employment.
The analysis models the potential savings from improving this populations' (W2 and Food Stamp recipients) access to jobs. The potential savings come from reductions in program benefits as income increases, and the tax effects of increased wages, as most of these households are eligible for the Earned Income Tax Credit (EITC). Most of the households in this population are already earning wages. These households were characterized as the "working poor," although the study made no attempt to quantify the effect of transit investment on low-income households that did not participate in either W2 or the Food Stamp program.
The Delaware Valley Regional Planning Commission (DVRPC), the MPO for the Philadelphia area, utilized GIS to do extensive map work as part of job access planning and identifying the unmet needs of the region. Information such as the addresses of the target population, locations of likely job opportunities, corporate parks, schools, existing public transportation services, childcare centers and the location of regional service centers was mapped as part of the analysis.
The region's existing network of rail, trolley and bus routes offers a high level of service to commuters traveling during peak hours. Services are designed to funnel large numbers of people to a select number of transit hubs located in downtown business districts of core cities, revealing that service does not meet the needs of the relatively low-density residential and commercial areas in suburban counties.
Also as part of the analysis, DVRPC completed a transit accessibility analysis of the region and developed separate GIS networks for the varying transit services in the area. The results were compared against 1990 demographic and employment data at the zonal level. (Population and jobs data at the Traffic Analysis Zone (TAZ) level is available only from the decennial Census.) Based on this information, the DVRPC calculated the accessibility of residents and jobs to existing public transit services at distances ranging from ¼ to 2 miles based the type of transit service: rail, bus or trolley service. As a result, the percentage of the population and jobs located within ¼ mile (¼ mile is considered acceptable walking distance by most commuters) was determined by mode of transit. Accessibility to transit services was also analyzed by time of day and day of the week. Because many jobs require travel outside peak hours, mid-day and evening service is critical for many new entrants to the workforce and was identified as lacking.
The identification of unmet needs in the Philadelphia region has been an on-going process using a variety of sources. For example, the Mayor of Philadelphia often receives phone calls from employers in need of employees and then informs the stakeholder group.
The DVRPC is also responsible for the evaluation of Job Access and Reverse Commute applications and the inclusion of new initiatives in their overall transportation plan. They play a primary role in coordinating the application process and fielding questions from potential applicants.
New Jersey planning efforts to identify the unmet needs of the target population occurred at the regional and county level. GIS was utilized to show the existing transit network, address mapping of WFNJ population, childcare centers and major employers. With help from the DVRPC, each county determined areas where fixed route service, shuttle service, vanpools, or assistance such as bus passes might be most beneficial to the target population. In each county, a lead contact and steering committee was formed to work in conjunction with a consultant provided by the State.
Each county prepared a transportation plan that includes: a transportation demand analysis, a transportation supply inventory, identification of gaps in service, and a plan to coordinate existing resources to fill the gaps identified. The plans helped to identify specific services that were incorporated into the FY 1999 Job Access and Reverse Commute proposals.
The process of identifying unmet needs in the Detroit area was largely based on GIS analysis. A number of organizations worked together to coordinate information resources to develop a family of transportation options. The organizations involved, particularly the transit providers, the metropolitan planning organization, and the state transportation and social service agencies, recognized that transit services must be market driven, flexible, and responsible to the changing needs of the target population.
The location of the target population, jobs, childcare facilities and existing transit services were analyzed to identify areas under-served by transit. MAC and the DDOT worked extensively with local chambers of commerce and community organizations to identify and address employer transportation needs. Additionally, SMART utilized GIS to provide information to stakeholders and employers on how transportation changes might impact jobs.
As in many areas, a formal process was not undertaken to identify transportation needs; a valuable tool proved to be the sharing of information and experiences among stakeholders.
San Luis Obispo County
The identification of unmet needs in San Luis Obispo County was based primarily on GIS spatial analysis. GIS assisted the stakeholder group with identifying areas under-served by existing transportation services and the locations of potential employers. The analysis showed that 70% or more of the target population lived ¼ mile or less from an existing busline in the county, and that childcare facilities were very accessible from buslines as well.
The information also showed that existing transit services stopped at the city limits and did not provide adequate service to the airport and surrounding area. Earlier data identified the airport as a potential employer, and the number of available jobs was found to be dramatically higher than originally anticipated.
Traffic volume and traffic movement information assisted in determining where commuter routes would be most beneficial. GIS mapping identified an area highly concentrated with target population residents and no existing transit services. Although job opportunities are lacking in this area, planning efforts continue to develop fixed route, deviated route or demand response services.
A formal evaluation process was not used to select transportation alternatives in San Luis Obispo County. The unmet needs of the area were fairly easy to identify. One evaluation method considered was to count the number of individuals potentially served by each transportation alternative and compares the relative costs associated with providing the service. Planning efforts to develop alternatives for areas with highly concentrated target population residents are still in progress. A more detailed evaluation process is expected to be used as these efforts continue. Stakeholders plan to evaluate each strategy based on the anticipated cost and the estimated recovery rate from fares for each potential service alternative.
Key lessons learned as part of San Luis Obispo County's planning efforts focused on the quality of the data used in the analysis. Accurate and complete data is absolutely necessary to enable good decision-making. Confidentiality issues were also a problem in obtaining childcare provider information. Finally, coordination and a positive approach to welfare reform as a whole is important among stakeholders.
Each of the seven counties in the Minneapolis-St. Paul metropolitan area was responsible for identifying unmet needs in their area. Anoka County focused on transporting individuals to and from one central location for mandatory training classes. The stakeholder group in Anoka County utilized GIS to determine those needs. They were able to estimate how many individuals lived on or near a busline, how many lived in a condensed area not far from a busline, and the number of individuals living in dispersed rural locations. It was not known however, how many individuals had access to a vehicle. The stakeholder group estimated the number of individuals in each area who might be transit dependent, and of those, how many would need childcare transportation.
The stakeholder group determined that bus passes, demand response service, and loans to purchase and repair cars would be the most beneficial alternatives. They based the allocation of funds on the anticipated use of each type of service. The demand-response transit provider developed a cost per passenger per day amount for both individual trips and trips including a childcare stop along the way to the training center. The human services department scheduled members of the target population for mandatory training based on the maximum amount of service the demand response service could provide.
Communication between transit providers, job counselors and financial workers proved to be the most effective means to troubleshoot flaws in existing services and determine changes necessary to accommodate the greatest number of target population members.
Lack of readily accessible transportation statewide, particularly in rural communities, is a problem in Kentucky. The Human Service Transportation Delivery (HSTD) determined that the transportation needs of the target population could be better served by enhancing the quality of transportation services through random drug and alcohol testing of drivers and improved vehicle maintenance standards. The improvements were designed to provide better access for medical care, social services and job training. In an effort to streamline costs, a capitalization rate structure was developed for each region in the state.