Addendum

The pages following provide more detailed information on the following common performance measures:

1.   Federal housing assistance program performance
2.   Job Training and employment
3.   Wildland Fire Management
4.   Flood Mitigation
5.   Participation in Disaster Insurance
6.   Health
7.   Environment (Note addition to measures covered in guidance
      letter above).
      A.  Rural Water Project Performance
      B.  Non-Point Source Water Pollution Program Performance
      C.  Wetlands

The above information was provided to affected agencies under separate cover.

A Comparative Measure of Federal Housing Assistance Program Performance

Description of the Measure:

The relative performance of federal housing programs with similar functions and purposes can be assessed using a uniform cost-effectiveness measure. This comparative measure is based upon a discounted net present value of the subsidy costs required to serve an eligible family of a particular income level over an extended period of time. This measure can be adjusted for family size, household income, timing of occupancy, and other relevant factors. This analysis uses a 30-year life cycle to conform to standards in the housing industry.

This statistic of relative cost-effectiveness can be used to derive another important measure of program performance – the number of households served with any given level of budgetary resources. A more sophisticated version of the measure might account, on the cost side, for displacement of private investment and, on the benefit side, for the estimated value of the housing units and their locations for the beneficiaries.

Applicable Programs:

This measure can be applied to a range of housing programs at HUD (Housing Vouchers, Section 202 for the Elderly, Section 8 New Construction, Public Housing, HOPE VI), as well as housing programs in other agencies, such as Military Housing at DOD and Section 515 Rural Rental Housing Program at USDA.

Analytical Issues:

The Federal government has, over the years, used an evolving set of programs to address the affordable housing needs of lower-income households. Initially, subsidy was provided for direct production of low-income housing. Beginning in the 1960s, HUD used several combinations of subsidized or guaranteed financing, construction grants, and ongoing rent subsidies to support the development and operation of privately owned, assisted housing with "project-based" subsidies. Currently, five major Federal programs support production, including the Low-Income Housing Tax Credit, construction grants for the Elderly and Disabled, revitalization of severely distressed public housing under HOPE VI and local construction or rehabilitation using HUD's HOME and CDBG block grants. Together these programs provided $6.6 billion for subsidized production in FY 2001.

Beginning in the 1970s, a new form of tenant-based Section 8 rental assistance was developed that has evolved into the housing voucher program. We would expect a measure to help evaluate the relative cost-effectiveness of providing housing assistance through a tenant-based or a project-based approach. In doing so, the additional criteria should be considered:

Costs

Targeting of resources.

Minimizing substitution for private investment

Long-term federal liability or exposure

Benefits

Quality of unit and neighborhood

Promotion of self-sufficiency

Housing stability

An Example:

Attached is an example illustrating how this comparative measure has been used to compare a set of housing assistance program models. The assumptions are detailed in order to clarify how such an approach can account for a wide range of factors and treat each approach in a similar manner to facilitate comparison.

Description of Alternatives and Costing Assumptions

The following cost analysis compares four alternative ways to spend $5 billion to strengthen affordable housing efforts. Modeled on existing programs, the four alternatives are (a) Construction using Project-based Subsidies for Mortgage and Operating Costs, (b) Production Grants plus Operating Subsidies, (c) Housing Vouchers, and (d) Low-Income Housing Tax Credits plus Vouchers. In order to compare costs for different programs, all costs have been discounted to their present value in FY 2002. To provide a rough idea of how many households each alternative will reach, the cost per household is then divided into $5 billion.

Present Value of Federal Cost Per Unit of Various
Subsidized Housing Alternatives

Net Present Value

Households
Served with $5B

Construction using Project-based
Subsidies for Mortgage and Operating Cost
(Modeled on former Section 8 New Construction)
$123,409

40,515

Production Grants plus Operating Subsidies
(Based on Section 202 Housing for the Elderly)
$131,973

37,886

Housing Voucher Program
$86,827

57,585

Low-Income Housing Tax Credit Plus Vouchers
$97,895

51,075

General Assumptions for Cost Comparison

Net Present Value discount rate of 6.3%.

Tenancy begins in FY 2005.

Construction of project-based programs begins in 2002 and extends through 2004.

Operating costs and rents increase at GDP Price Index from OMB Economic Assumptions (2 percent in most years).

Amortization terms: 30-year loan at an interest rate of 9.0%.

Households occupy 2 Bedroom units.

National median income for FY 2000 is $45,180 adjusted for a family that might occupy a two-bedroom unit.

Tenant contribution is assumed to be $200/month for all alternatives.

A. Construction using Project-based Subsidies for Mortgage and Operating Costs

Description:

Modeled on the former Section 8 New Construction program. Privately owned and
operated multi-family housing projects would be constructed with long-term rental
assistance and mortgage financing (either FHA or conventional). Tenants pay 30% of
their income for rent.

Net Present Value: $123,409
Households Served with $5 Billion: 40,515

Assumptions:

Total development costs estimated at $82,024 in 2002 dollars based on Cummings and Dispasquale's research on the LIHTC program. There are some concerns that these cost projections underestimate expenses. For example, in a number of projects, land costs are not included.

Total Development Costs amortized at 9% over 30 years.

Operating Expenses estimated at $3,288 in 2002 based on FHA, Office of Housing data. Similar estimate derived from Institute of Real Estate Management data.

Administration fee added 3% of gross rent.

No net cost assumed for FHA-guaranteed financing.

Tenant Contribution $200/month in 2002.

B. Production Grants plus Operating Subsidies
Description:

Based on the Section 202 Housing for the Elderly program. An upfront grant covers all construction costs. Projects do not carry any debt service. Ongoing operating costs are covered. Tenants pay 30% of their income for rent.

Net Present Value: $131,973
Households Served with $5 Billion: 37,886

Assumptions:

$79,417 production grant, expenses in equal amounts over three years.

$4,222 operating expenses in 2002 based on 202 Project estimates.

C. Housing Vouchers

Description:

Vouchers cover the difference between 30% of recipient's income and local Fair Market Rent. Vouchers provide access to the lower half of the private rental market (roughly the 40th percentile of market rents). Portable assistance enables tenants to move when their lease expires and retain housing benefits.

Net Present Value: $86,827
Households Served with $5 Billion: 57,585

Assumptions:

$5,950 per unit costs in 2002 dollars.

Administration Fee of 3% included.

Tenant Contribution $200/month in 2002.

National average FMR in 2000 approximately $625 for a two-bedroom unit.

D. Low-Income Housing Tax Credit Plus Vouchers

Description:

The LIHTC provides state-administered tax credits to private equity investors in affordable housing. A small ($800) one-time capital grant is used to encourage certain types of development, such as larger units in non-poverty neighborhoods. Vouchers allow "worst case needs" households to afford higher rent levels.

Net Present Value: $97,895
Households Served with $5 Billion: 51,075

Assumptions:

Equal distribution of credit over 10 years.

Tax credit $38,000 in 1996 dollars. (Cummings and Dipasquale)

LIHTC rent for a household at 60% of median income is $678 in 2000.

Administration fee added 3% of gross rent.

Job Training and Employment
Comparison Measure

Last year we reviewed nearly 50 federal job training and employment programs. That review showed that there was little consistency across programs in measuring outcomes. That review also highlighted that not all job training and employment programs provide the same services. Some programs focus on classroom or on-the-job training; others on employment and re-employment assistance; while still others include a combination of services. We also found that few serve the same target populations. Some programs focus on adults, some on youth and young adults, and others on special target populations like veterans, the disabled, or American Indians. Even within a particular group, certain programs may serve people with different economic needs. But what was clear was that programs did not always measure what is their primary goal—participants leaving their programs with a job. The purpose of this proposal is to develop a common performance measure that addresses the goal of getting a job for participants in all affected programs.

While we recognize that some agencies may not have data now, over the summer we will develop a framework and time line for getting the data for performance on the primary outcome measure of getting a job where those data are deficient or non-existent. Among other things, this will involve ensuring we are measuring the same thing (e.g., "placement" may be defined differently among programs), and we have common definitions for these measures.

Agencies proposed to be involved in this common performance measure initiative include the Departments of Labor (including its Workforce Investment Act programs for youth, adults, and dislocated workers), Education (including such programs as Vocational Rehabilitation, Vocational and Adult Education), Housing and Urban Development (Youthbuild), Veterans Affairs (Vocational Rehabilitation and Employment), and Interior (Bureau of Indian Affairs programs). We are proposing four possible measures for program outcomes, but are open to other possibilities. Three are outcome measures and one is an efficiency measure.

Measure 1—Attainment of a job

To determine how effective programs are in meeting the outcome goal of job training programs—placement in a job. Agencies would report their methodology for measuring job attainment and the data collected by them or by their grantees to determine what data exists, how and when job placement is measured, and what proportion of participants actually exit the program into a job.

Measure 2—Attainment of a certificate or degree by program participants

To determine how effective programs are in meeting intermediate goals that can lead to better jobs and long-term earnings. Even though the primary outcome goal of job training and employment programs is a job, a significant intermediate outcome measure can be whether a program increases participants' skills needed to get and retain a job. We suggest attaining a degree or certificate as a possible common measure since this often is an intermediate step to another training or employment program before gaining a job and, as such, is a reasonable indicator for eventual success in the job market. As above, agencies would report on data now collected and available and describe the basis for that data.

Measure 3—Earnings gains

To determine whether programs have an effect on participants' earnings compared to their earnings prior to program enrollment. While job attainment has many benefits, having a job that doesn't pay more than the participant was earning before program enrollment undermines the programs' long-term outcome goal of improving employment and earnings. Agencies would report on surveys or on other program data for this outcome.

Measure 4—Total program cost per placement in a job

This efficiency measure would aggregate total annual program costs (something else that will need to be standardized) divided by the number of placements in a job or in postsecondary education.

Wildland Fire Management
Proposed Comparative Performance Measures

In response to the historic 2000 fire season, the Departments of the Interior and Agriculture both received large increases in funding for hazardous fuels reduction activities. In fact, fuels reduction funding for USDA increased by 193% to $205 million while Interior funding increased 298% to $195 million. These increases are premised on the common sense notion that around homes and communities in the wildland-urban interface (WUI), it is better to prevent fires than to try and fight them after they start. However, little data exists on how best to make these programs effective at reducing these risks.

To improve performance, OMB proposes that the agencies develop common measures to track and compare agency progress in reducing community fire risks. Further, we reiterate the importance of developing measures that focus on the efficiency of delivering a particular level of service (i.e., unit cost). Thus, the following measures also include a comparative measure (with an implied efficiency objective) relating to fire preparedness and suppression.

Measure 1: Percent of adjacent at-risk communities removed from the at-risk list as a result of agency actions or assistance.

This measure would provide a direct assessment of the impact of federal fuels treatment programs on reducing risks to communities and would allow for a comparison of DOI and USDA efforts. Other assistance programs could also be included, allowing for cross-comparison of the effectiveness of various types of assistance.

In order for this measure to be useful, a number of questions must be answered:

Should the size of a community be a control factor or should communities be grouped by size?

What standards should be used to determine whether a community can be removed from the at-risk list? The extent to which a buffer is established around the community? Whether or not adequate defensible space is created to allow firefighter access in an emergency?

What agency actions or assistance should be included? Fuels treatments only? Or other community actions resulting from federal assistance programs like FIREWISE?

Measure 2: Completion of Fire Management Plans (FMPs). This would include working with States on coordinated fire management plans covering federal, state and local lands. The Forest Service previously announced it would update its fire management plans by 2003, and then altered the date to 2004. DOI has a similar target. This timeline could be accelerated to achieve an earlier date in 2003. The agencies have already established a Fire Management Planning Workgroup to ensure a seamless, consistent, and coordinated approach to fire management planning efforts.
Measure 2a: Percent of total acres burned that is managed with wildland fire use. This is a byproduct of completed fire management plans.
Ensuring FMPs are completed implies that there will be more opportunities for fires to be allowed to burn within prescribed parameters. Managing wildland fires for beneficial uses is usually much less expensive than suppressing them, particularly over the long-term and in remote areas. (This type of management requires that the agency or agencies have a carefully crafted fire management plan in place.) Establishing this measure could encourage the two agencies to improve development of fire management plans.
Measure 3: Average cost per fire and cost per acre burned for large fires, small fires, WUI fires, non-WUI fires, management units with current FMPs, and units without current FMPs.
Agencies would also be required to prepare explanations on procedures used for any fires 50% above baseline. Agencies would also establish a routine audits process to assess actual costs and reasons for cost increases (checklist of actions and behaviors) whenever annual spending exceeds thresholds.
The goal is to improve agency evaluation of the reasonableness of suppression costs. The agencies need to establish procedures, whether in requiring the use of cost-benefit analysis tools for field personnel in ordering equipment, like tankers or helicopters, or establishing new national guidelines on equipment usage. There are no incentives presently to hold down costs in the field. In fact, the reverse is true. Moreover, there are concerns that the agencies have used the fire program to fund broader needs (indirect expenses, equipment purchases, construction needs, etc.).

Effectiveness of Flood Mitigation Programs
Multiple Federal disaster programs attempt to reduce, or "mitigate," property damage caused by flooding. The Office of Management and Budget has identified four programs that share this goal but that rely on differing strategies.

FEMA Hazard Mitigation Program (HMGP),

FEMA National Flood Insurance Program (NFIP),

U.S. Army Corp of Engineers (Civil Works) Flood Mitigation program, and

USDA Natural Resources Conservation Service (NRCS) Small Watershed Program.

To evaluate the relative effectiveness of these programs, ideally agencies should develop common performance measures that compare net societal benefits against federal costs of a program. Since the full range of societal benefits are very difficult to put in monetary terms, OMB recommends that a common performance measure be developed to estimate net economic benefits (e.g., flood damages avoided), per Federal dollar spent. Note that there is a similar comparison of benefit/cost measures in the Army Corps of Engineers chapter of the FY 2003 Budget (page 295).

Benefits from Each Dollar Invested in Selected
Federal Flood Damage Reduction Programs

Corps of Engineers

Federal Emergency Management Agency's 404 Flood Risk Mitigation (HMGP)

Department of Agriculture's Small Watershed Program (NRCS)

$1.90

$2.30

$1.40

Because the four programs are designed and operate differently, the challenge for agencies is to measure comparable benefits and costs. Attached is a basic analysis of agencies' current benefit cost methodologies and guidance on developing a benefit-cost measure for this exercise. OMB staff will be available to work with agencies to help develop a common approach for evaluating program effectiveness.

Discussion of Proposed Benefit Cost Calculations

The agencies should develop a measure of the ratio of net non-Federal economic benefits to Federal budget costs. Mathematically, this means:

(Total project benefits—non-Federal costs)
Federal budgetary costs

For FEMA's Hazard Mitigation Grants Program (HMGP), the NRCS Small Watershed Protection program and the Corps of Engineer's Flood Mitigation program the primary benefits are avoided damage to property due to mitigation measures. The estimate should be based on probabilistic estimates of flood events. The non-Federal costs would include the non-Federal cost of the project (e.g., cost share contributions). The net of the project benefits and non-Federal costs would be divided by the Federal share of the project costs. The cost of conducting assessment and feasibility studies as well as operation and maintenance cost should be included in the project costs, since it is necessary for completing a project.

For FEMA's National Flood Insurance Program (NFIP), the primary benefits are also avoided damage to property due to mitigation measures. Again, this estimate should based on probabilistic estimates of flood events. FEMA currently calculates the value of avoided property damage resulting from compliance with flood plain management requirements. However, it is not clear what method is used to calculate benefits. The NFIP should base its benefits calculations on the HMGP methodology, which uses time value of money and probabilistic analyses.

For NFIP, the non-Federal cost is the cost of complying with the NFIP's flood plain management requirements. Property owners must elevate structures in the flood plain to the base flood elevation and make other structural modifications, and local governments must expend resources to ensure compliance. If it has not already, FEMA should conduct a standard regulatory analysis of its flood plain management requirements. This should provide information on costs identified above.

Unlike the other three programs, NFIP is fully funded by policyholders and has no appropriated expenditures. FEMA should work to find a meaningful measure of Federal costs, considering perhaps appropriated funds for its flood mapping program which underlies its flood plain management activities or the implied annual Federal subsidy to the NFIP.

For all four programs, agencies will need to develop a common metric for measuring budgetary spending. Some options include the average cost per project, mean cost per project, or average annual program expenditures. Agencies should also agree on common time frame for referencing past spending.

Benefits and costs not covered by this analysis

Benefits and costs that cannot be easily put into monetary terms, such as environmental costs and benefits are important aspects of these programs. However, we currently do not have a good method of displaying these types of effects in numerical terms that would be consistent across different types of projects or programs. These benefits and costs are not included in this measure. However, to the extent that non-monetary benefits can be measured or described, OMB would like for those measures and descriptions to accompany information on the economic performance measure.

Projects and programs may also have affects on property values and/or future levels of development. Typically, these are not measured by the programs. For example, a flood prone area that is protected by a project may lead to greater confidence that flooding is unlikely in that area. This may result in induced development or increased property values for existing development. A program that takes property out of the market to reserve it as floodplain, may also affect the value of that property and the adjacent properties. These changes are not currently estimated by agencies in project development for several reasons. First, changes in property values and levels would be difficult to predict. Second, if decisions about whether to proceed with a project to protect a flood plain were made based on projected development rather than on existing development, the Federal government would find itself subsidizing the elimination of natural flood areas that serve important environmental as well as flood mitigation functions. These types of changes are also not included in the proposed performance measures. However, if there are data available in this area, OMB would like to be provided with summary information.

Participation in Disaster Insurance Programs

In response to market failures, the Federal government offers two forms of disaster insurance—FEMA flood insurance and USDA crop insurance. Yet despite the real risks of floods and other natural disasters, both agencies face difficulties in maximizing the number of units covered by their programs. FEMA does not regularly measure participation rates, though past studies report a wide variance in participation. USDA's reported participation rate averages approximately 75 percent. That is, 75 percent of all crops eligible for crop insurance are actually covered. Efforts to increase the participation rate by greatly decreasing the cost of such insurance have not been successful. Despite increasing the federal subsidy for such insurance by 96 percent over the last 5 years, participation only increased by 16 percent. Instead, the increased subsidy has resulted in participating producers shifting to higher levels of coverage, which is an additional goal of the program. However, increased participation in these programs is critical since insurance diminishes economic and personal suffering of families and offsets the cost of direct Federal disaster assistance. Importantly, program enrollment encourages the individual policyholders to undertake risk-reducing measures rather than relying solely on the Federal government's intervention.

To improve performance, OMB recommends that the Crop Insurance Program and National Flood Insurance Program develop a common performance measure for participation. Further, agencies should identify and share "best practices" that contribute to improving participation. As part of this process, OMB expects that agencies will consider factors such as statutory purchase requirements, the role of private insurance partners, the effectiveness of advertising, and subsidization and pricing of policies. (Note that program participation must not be achieved by compromising actuarial soundness.) OMB staff will be available to work with agencies on this effort.

FY 2004 Health Comparison Measures

Measure 1—DOD, VA, and HHS' Indian Health Service and Community Health Centers

To quantify the resources expended on direct and Federally-funded health care programs and better understand the cost differences, we propose the following measure.

Cost Measure: Per capita expenditures on direct and Federally-funded health care services.

Methodology: Agencies calculate actual expenditures divided by the number of people served (each patient counted only once regardless of number of encounters). Health expenditure data would include all appropriated funds (including collections and reimbursements) for facilities, personnel, equipment, and other ancillary services. This would not include separate appropriations for research. In order to adjust for differences in patient health status and case mix, the Agencies will adjust the population numbers for age and sex differences.

Measure 2—DOD, VA, and HHS' Indian Health Service and Community Health Centers

To assist with evaluating the efficiency of these programs, we propose the following measure.

Efficiency Measure: Average number of patients seen per day per physician/nurse practitioner/physician assistant.

Methodology: Agencies calculate the total number of patients seen per year divided by the FTEE number of physicians/nurse practitioners/physician assistants seeing patients, divided by the number of days care was provided (i.e., work days). In order to adjust for differences in patient health status and case mix, the Agencies will adjust the population numbers for age and sex differences.

Measure 3—DOD and VA

To compare health outcomes, we propose the following measure.

Quality Measure: Total number of readmissions for all patients admitted for medical or surgical conditions, divided by the total number of inpatient admissions for those purposes.

Methodology: Agencies would calculate readmission rates for inpatient admissions by measuring the number of patients readmitted within a 72 hour, 15 day, and 30 day period after discharge, divided by the total number of patients admitted during the year. In order to adjust for differences in patient health status and case mix, the Agencies will adjust the population numbers for age and sex difference.

Environmental Common Measures

A. Comparative Measure of Rural Water Project Performance

Description of the Measure:

The federal government administers at least five different programs that serve the same dual objectives: to help communities located in rural areas of the country to improve the quantity and quality of their water. Because the programs may address different stages of water delivery, applying a suite of measures can help in evaluating program effectiveness.
The results could aid a review intended to improve the government's efficiency and
customer service in its efforts to address rural water issues.

We suggest a suite of three measures:

Number of water connections per million dollars;

Quantity of potable water delivered per million dollars;

Quantity of water treated to EPA drinking water quality standards per million dollars.

Although not every program would be able to use all of these measures, every rural water program should be able to use at least one, providing one comparison of program effectiveness. Different programs have different desired outcomes; for instance, some programs focus on developing water to spur economic growth, while others develop it for health reasons or to meet regulatory standards. Our proposed measures are outcome-oriented but would not handicap program comparability by being too specific.

Applicable Programs:

These measures can be applied to at least 5 programs that can be classified as addressing rural water issues in the Departments of the Interior, Agriculture, Housing and Urban Development, and Health and Human Services, and the Environmental Protection Agency. A specific list is attached. There may be other programs that we have not identified here which qualify as programs that address rural water needs, and these should be identified during the review.

Analytical Issues:

OMB has begun evaluations of rural water programs in the past, but they were not followed through to completion. The first step in evaluating these programs will be to conduct an inventory to identify all the programs addressing rural water needs. Second, programs need to be assessed to determine specific areas of redundancy. Despite serving the same goal, an uncoordinated proliferation of programs has developed, based on the different specific target groups. Some, such as the Bureau of Reclamation, construct large storage, conveyance, and treatment facilities, which Reclamation sometimes does on behalf of other agencies (for instance the Bureau of Indian Affairs). Others, such as USDA's Rural Utilities Service, focus on the needs of lower-income rural communities by providing loans and grants to finance water facilities and hookups. Still others, such as the Environmental Protection Agency's Drinking Water State Revolving Fund (which serves all areas, not just rural parts of the country) provides loans for drinking water infrastructure through a state revolving fund. While each individual program may meet its mission in serving its clients, these programs have developed in an uncoordinated fashion, resulting in a number of policy problems. Establishing a suite of common performance measures for these programs can help to evaluate their effectiveness, and will aid in making decisions about program priorities.

Specific Methodology:

In general, project cost calculations should be on a net present value basis.

Number of water connections per million dollars. Many rural water projects are designed to provide Municipal, Industrial, and Residential (MR&I) water; therefore we should measure the total extent of service provided to the community, not just number of families, businesses, or government entities. Counting the number of water connections should measure service to all of these.

Quantity of potable water delivered per million dollars, both gross and net, to account for programs that have a cost-share component, or are reimbursable.

Quantity of water treated to acceptable standards per million dollars; similar to the water delivery measure, this should track both gross and net costs.

Department Agency Program

Agriculture Rural Utilities Service Water and Wastewater Loans and Grants

Housing and Urban Development Community Planning and Development Community Development Block Grants: State Program

Interior Bureau of Reclamation Water and Related Resources account, on a project-by-project basis.

Environmental Protection Agency Drinking Water State Revolving Fund

Health and Human Services Indian Health Service Sanitation Facilities Program

B. A Comparative Measure of Non-point Source Water Pollution Program Performance

Description of the Measure:

The Federal Government administers several programs aimed at reducing non-point source pollution. Since the Clean Water Act became law in 1972, water quality has dramatically improved in many areas. However, these gains came largely from reductions in the discharges of point sources, leaving non-point sources as the largest remaining water quality problem. Without a significant reduction in non-point source pollution, it is unlikely that water quality will further improve.

A comparison of the effectiveness and efficiency of Federal non-point source programs could be helpful in targeting limited resources. The ideal situation might be to compare programs that address different impacts to a river or waterway, such that you can compare the relative impact of a program to reduce sediment loading with one that is designed to reduce nutrient loading.

We suggest the following measures:

Tons of sediments reduced per million dollars.

Pounds of nutrients reduced per million dollars.

Estimates of the percentage impact of agency actions on exceedences and contaminant occurrences.

Applicable Programs:

These measures can be applied to the Section 319 Non-point Source Management Program (EPA), the Environmental Quality Incentives Program (USDA), and the Conservation Reserve Program's continuous signup (USDA).

Analytical Issues:

EPA's 319 Non-point Source Management Program supports a wide variety of activities including technical assistance, financial assistance, education, training, technology transfer, demonstration projects, and monitoring to assess the success of specific non-point source projects. USDA's Environmental Quality Incentives Program (EQIP) provides educational, financial, and technical assistance to farmers and ranchers to implement conservation practices on land currently in production. Eligible water quality practices include animal waste nutrient and pesticide management, as well as livestock water development projects. The continuous signup option within USDA's Conservation Reserve Program (CRP) provides farmers with financial and technical assistance to enroll acres to prevent water resources and install conservation practices such as riparian buffers, grass waterways, and filter strips.

Work is needed to identify the components of each program and determine whether only a subset of each program should be compared. Data must be collected on easements and acquisition costs, and other various practices. Estimates of sediment and nutrient reductions must be ascertained from models if real monitoring data are not available. A NOAA study of the economic costs and benefits of methods to reduce nutrient loads in the Gulf of Mexico relied on modeling data from USDA's Economic Research Service (U.S. Mathematical Programming Model) as there was limited real data available.

A comparable baseline is needed across programs. Program data sources must be identified and reviewed. A literature search of academic and government studies should be conducted to determine if any previous work in this area would be helpful. Consideration should be given to whether other Federal or nonfederal non-point source programs should be included in the comparison.

Specific Methodology:

Below is a methodology to compare programs such that you can compare the relative impact of a program to reduce sediment loading with one that is designed to reduce nutrient loading.

Show nutrient levels. Identify loads by land use (i.e., farmlands, forests, and urban/suburban). Show total phosphorus in streams by tons. Also show total phosphorus in streams by gradient.

Show sediment levels. Identify loads by land use. Show total sediment in streams by tons and group streams by sediment load. For example, X percent of streams have loads of Y tons. Appropriate outcome measures for sediment loading needs to be further developed.

Data sources: U.S. Geological Survey National Water Quality Assessment Program for inland streams and rivers and EPA's Environmental Monitoring and Assessment Program for coastal areas.

By program, by year, and by land use estimate the tons of sediments reduced per million dollars; estimate the pounds of nutrients reduced per million dollars.

C. Program Comparisons on the Environment—Wetlands

Programs

The Army Corps of Engineers construction program, Mississippi River and Tributaries program, and regulatory program; the Louisiana Coastal Wetlands Planning, Protection, and Restoration program; the Natural Resources Conservation Service Wetlands Reserve program; the Fish and Wildlife Service Partners for Fish and Wildlife program, Coastal program, Coastal Wetlands program, North American Wetlands Conservation Act program, and National Wildlife Refuge System; National Oceanic and Atmospheric Administration programs whose principal purpose includes the protection or improvement of wetlands; and the programs of a representative group of non-Federal entities.

Metrics

Total wetlands acres established or re-established in FY 2001 as a result of agency action.

Total wetlands acres that will be lost (before mitigation) as a result of FY 2001 agency action.

Acres of wetlands (1) established, (2) re-established, (3) rehabilitated, (4) enhanced, and (5) protected/maintained in FY 2001 primarily as a result of agency action, per $1 million in total costs.

Additional acres of wetlands (1) established, (2) re-established, (3) rehabilitated, (4) enhanced, and (5) protected/maintained in FY 2001 primarily as a result of action by representative non-Federal entities, per $1 million in total costs.

The quality of the wetlands (1) established, (2) re-established, (3) rehabilitated, (4) enhanced, and (5) protected/maintained.

Methodology

In characterizing the wetlands action (i.e., established, re-established, rehabilitated, enhanced, or protected/maintained), use the definitions developed by the Committee for Defining and Tracking Wetlands Gains and Losses of the White House Wetlands Working Group (1999)(attachments). For metric #1, do not count the restoration of farmed wetlands.

Provide a copy of any ranking factors that the agency uses to select the sites to be acquired or improved.

For metrics #1, #2, and #3, include only wetlands actions completed in FY 2001. For metric #2, include the losses that will result from decisions reached in FY 2001 (e.g., a permit issuance, or a construction start of a project that will result in wetlands losses before mitigation).

For metrics #1, #2, #3, and #4, do not include acres of related non-wetlands habitat.

For metrics #3 and #4, estimate the average cost for each of the five categories of wetlands action. Present all costs using October 2001 price levels. Estimate the full cost (Federal and non-Federal) of the action, including administrative and supervisory costs; overhead; technical assistance; engineering and design. Display Federal and non-Federal costs separately. Show land acquisition costs and provide a separate annual cost estimate for any long-term operation, maintenance, or monitoring.

For metric #3, the Army Corps of Engineers regulatory program may limit its estimates of average costs to the permits that involve 5 acres or more of compensatory wetlands mitigation. It should develop its estimates through a statistically valid sample of wetlands banks operators or by reference to other reliable data. Include the applicant's cost to develop a mitigation plan and to collect supporting data, and all costs incurred by the Federal and State agencies that review the permit application.

For metric #3, in addition to estimating the average cost of all wetlands actions, each agency should submit a table with acreage and cost data on each action involving 15 acres of wetlands or more. Please identify the type of wetlands involved using the Cowardin, Carter, and La Roe classification system (http://www.fws.gov/stand/standards/cl_wetl.html), and indicate how the agency scored the action under its ranking system (if any). Also, please prepare a graph showing the distribution of the actions listed on the table by cost (in acres per $1 million) and a brief assessment of the justification for any of these actions that cost more than twice the agency average on a per acre basis.

For metrics #1, #2, and #3, the Army Corps of Engineers should combine its submission for the construction program and Mississippi River and Tributaries program, but separate the data on ecosystem restoration projects from the data on compensatory mitigation actions.

The Fish and Wildlife Service will be responsible for metric #4. For this metric, it should treat the value of any land rights donated to non-profit entities as a non-Federal cost, and should include the value of any tax credits associated with such donations as a Federal cost.

The Army Corps of Engineers will be responsible for providing data on the Louisiana Coastal Wetlands Planning, Protection, and Restoration program.

On metric #5, the agencies will work with the Office of Management and Budget to select two or more geographic or regional areas for a qualitative, comparative evaluation of each agency's efforts.