Statement of Robert W. Burchell, Ph.D.
Distinguished Professor, Principal Investigator
Center for Urban Policy Research
Rutgers University, New Brunswick, New Jersey

prepared for

Senate Committee on Environment and Public Works
Subcommittee on Transportation and Infrastructure
United States Senate
Tuesday, July 14, 1998
9:30 a.m.

SECTION I
PUBLIC WORKS PROGRAM PERFORMANCE EVALUATION

STUDY OVERVIEW

The purpose of the research described here was to evaluate all 205 Economic Development Administration (EDA) Public Works Program projects that received their last payment in FY 1990. This means that, as of that date, the projects were completed and structures associated with them either occupied or soon to be occupied. Thus, at the time of this research_six years later_these projects had been sufficiently established to make their evaluation possible.

Since 1965, EDA's mission has been to promote the long-term recovery of economically depressed areas by assisting local governments via public works project grants in generating and retaining jobs and in stimulating commercial and industrial growth.

STUDY PROCEDURES

The study was undertaken from November 1996 through March 1997 by research teams from five universities and a major professional organization. All principals of the research teams have extensive experience in both economic development and infrastructure studies. Each principal spent significant time in the field researching individual projects and talking to grantees. Each principal and affiliated staff participated in some aspect of research analysis and in writing the final report. All concur with the findings presented below.

The research team contacted by mail and telephone 205 grantees of public works projects. To help the grantees better understand the purpose and types of information necessary to undertake the evaluation, all grantees were invited to attend seminars conducted by the research team at 13 locations nationally. Sixty (60) project sites were visited to conduct in-depth discussions with grantees to learn more about their individual projects' impacts and to validate the information that they were in the process of providing.

The analysis uses performance measures developed by EDA specifically to evaluate public works projects. Performance measures relate primarily to numbers of various types of jobs created or retained and amounts of private- and public-sector funds leveraged.

PROJECT TYPE AND CONTEXT

From a universe of 205 EDA public works projects receiving a closeout payment in FY 1990, all 205 were successfully contacted.

The composition of the 203 completed \1\ public works projects is as follows:

\1\Two projects aborted and were not constructed because of local financial or market reasons.

DISTRIBUTION OF PROJECTS

Number__Percent
Buildings 27__13.3
Industrial Parks 59__29.1
Roads 17__8.4
Water/Sewer 87__42.8
Marine/Tourism 13__6.4
Total 203__100.0

In terms of the context of the above projects, EDA public works projects take place in locations where levels of unemployment and percents of the population below the poverty level are 40 percent higher than state and national averages. These are also locations where per capita income is typically 40 percent lower than averages at the state and national levels.

PROJECT COMPLETION

Of those public works projects contacted by the research team, 99 percent (203) were completed as planned. Ninety-one percent (185) of the projects were completed on time.

Fifty-two percent (105) were completed under budget.

PROJECT IMPACTS

Project-Related Direct Impacts

Ninety-six percent (195) of the public works projects produced permanent jobs six years after completion. Eighty-four percent (171) leveraged private-sector investment over the period. On average, each public works project produced 327 direct permanent jobs for every $1 million of EDA funding.

Based on average EDA funding of $660,557 per project, $3,058 in EDA funds was spent per job created or retained. Total cost (all sources of funding, including EDA) per job created or retained was $4,857.

Not including public projects, for every $1 million of EDA funding, $ 10.08 million was leveraged in private-sector investment.

For all projects, for every $1 million of EDA funding, another $ 1 million was leveraged in federal, state or local investment.

15.0 FTE (full-time-equivalent) construction jobs were created per $ 1 million of EDA funding, carrying out solely the grant-supported component of capital infrastructure.

Nonproject-Related Direct and Indirect Impacts

Nonproject-related direct or indirect jobs (those that occur because of the project or the project's jobs) were found to be present in 30 and 35 percent, respectively, of all public works projects.

Considering all projects' ability to generate nonproject-related direct or indirect effects, for every $1 million of EDA funding, an additional 50 jobs and $ 1.18 million in private-sector investment were generated in nonproject-related direct effects, and an additional 64 jobs and $126,180 were generated in indirect effects.

Except in cases where the project was tax-exempt, public works projects increased the local tax base at a level of $10.13 million per $1 million of EDA funding.

PROJECT IMPACTS (GENERAL)

Public works projects' economic impacts generally increase with time. Jobs resulting six years after completion were, on average, twice the number witnessed at project completion.

EDA public-sector economic stimuli create private-sector jobs at high levels of success and low levels of cost.

CONCLUSIONS

Most of the public works projects achieved EDA's objective of providing communities with the necessary infrastructure to expand their economic base.

Jobs and private investment have occurred in many areas that would not have experienced these benefits without EDA assistance.

EDA offices as an instrument of government, and EDA field representatives who interact with grantees, are well-regarded by their constituencies.

SECTION II
DEFENSE ADJUSTMENT PROGRAM PERFORMANCE EVALUATION

STUDY OVERVIEW

The purpose of the research described here is to evaluate all (190) Economic Development Administration (EDA) Defense Adjustment Program grant projects approved during the period FY 1992 through FY 1995. The primary objective of this program and its projects is the restructuring of local economies to diversify away from dependence on former defense bases or defense contractors impacted by closure or cutback.

Direct appropriated funding to EDA for the Defense Adjustment Program began in FY 1994. From 1992 to 1994, EDA received transfers of funds for defense projects from the Department of Defense's Office of Economic Adjustment (OEA). The program, therefore, is relatively young, and as of 1997, the defense construction, capacity building (planning and technical assistance), and revolving loan fund (RLF) projects analyzed here were just taking hold. While their relative recency does not allow for an evaluation of these projects at full maturity, their accomplishments at this early phase can certainly be quantified.

As indicated above, the Defense Adjustment projects, even if completed, have had only a short time to mature. With time, the permanent jobs that they create will increase and the cost per job created will decrease. The present evaluation provides a snapshot view of the projects' effects during an early phase of their existence. A concurrent EDA study of the Public Works Program and other similar studies have shown that the effects (both direct and indirect) of these projects will increase substantially over time.

STUDY PROCEDURES

The study was undertaken from November 1996 through September 1997 by research teams from five universities and a major professional organization. All principals of the research teams have extensive experience in both economic development and infrastructure studies. Each principal spent significant time in the field researching individual projects and talking to grantees. Each principal and affiliated staff participated in some aspect of research analysis and in writing the final report. All concur with the findings presented below. The research team contacted by mail and telephone 190 grantees of defense adjustment projects. To help the grantees better understand the purpose and types of information necessary to undertake the evaluation, all grantees were invited to attend seminars conducted by the research team at 13 locations nationally. Forty-two project sites were visited to conduct in-depth discussions with grantees to learn more about their individual projects' impacts and to validate the information that they were providing.

The evaluation is undertaken using performance measures developed by EDA specifically to assess the productivity of defense adjustment projects. Performance measures for defense construction and revolving loan fund projects primarily involve numbers and types of jobs created or retained and amounts of private-sector funds leveraged. For capacity- building projects, the performance measure is a grantee self-rating of the quality and impact of the EDA capacity- building effort.

PROJECT TYPE AND CONTEXT

From a universe of 190 EDA defense adjustment projects that were approved from FY 1992 through FY 1995, all 190 were contacted.

The 187 grant-funded projects analyzed in this study\2\ include 162 single-element projects, twenty double-element projects, and five triple-element projects. These sum to 217 total project elements funded via the 187 EDA grants.

\2\ Three projects were never funded due to grantee financial problems (2) or cross purposes between the grantee and the EDA regional office (1).

[Table Omitted]

Since 1987, approximately 2.5 million defense-dependent jobs have been lost due to defense downsizing. EDA's Defense Adjustment Program is a direct response to base closures, base downsizing, and/or reduced defense contracting. Cutbacks are often sudden and severe for their host communities. In addition, projects are in locations where minority populations and percents of the population below the poverty level are 20 percent higher than state and national averages. These are also locations where per capita income is 25 percent lower than averages at state and national levels.

PROJECT COMPLETION

Of those 190 defense adjustment projects contacted by the research team, 98.5 percent (187) were initiated as planned.

Of those undertaken, about 97 and 98 percent of defense construction and capacity-building projects, respectively, moved to completion; 100 percent of the RLFs moved to completion.

CONTEXT OF PROJECTS

AT TIME OF APPLICATION (Medians)

(187 Initiated Projects)

Median__ Ratio to State__ Ratio to Nation
Unemployment Rate (%) 7.0__ 0.98__ 1.02
Per Capita Income ($) 13,034__0.72__0.73
Below Poverty Level (%) 15.5__15.5__1.18
Minority (%) 27.3__1.21__1.39

Of those undertaken and completed, 80 and 81 percent of the defense construction and RLFs, respectively, were completed on time. About 56% of the capacity-building projects were completed on time.

Of those undertaken and completed, about 90 percent of defense construction projects came in at or under budget; the figures for capacity building and RLFs are 97% and 100% respectively\3\.

\3\ RLFs, by their nature, cannot come in over budget. They lend what they have.

PROJECT IMPACTS\4\

\4\ As projects age and mature, project accomplishments will likely increase over time.

Project-Related Direct Impacts: Defense Construction On average, completed defense construction projects (49) have produced 30,870 permanent jobs to date, or 124 jobs per $ 1 million of EDA funding. These jobs were produced at an EDA cost of $8,052 per job and a total cost (all sources of funding) of $12,045 per job.

Defense construction projects produced 18.0 FTE\5\ construction jobs per $1 million of EDA funding.

\5\ Full-time-equivalent.

Completed defense construction projects (43)\6\ leveraged $722 million in private sector investment, or $2.2 million per $1 million of EDA funding.

\6\Forty-three of forty-nine defense construction projects have private-sector investment. Six projects are public sector and have no private-sector investment.

DEFENSE CONSTRUCTION AND CAPACITY-BUILDING PROJECTS:

PERMANENT JOBS: (Medians)

(49 Completed Defense Construction and 31 Completed TA* Capacity-Building Projects)

Defense Construction__Capacity Building)
Jobs Per $1M EDA 124__63
EDA Cost Per Job $8,052__$13,633
Construction/Professional Jobs 18.0 FTE__13.7 FTE
Private-Sector Investment Per $1M of EDA Funding $2.2 M__N/A

*Technical Assistance

Capacity Building

Capacity-building projects, by their definition and design, are not intended to create jobs directly, but to increase the planning, organizational, and technical skills needed for local economic development. Nevertheless, some jobs result as an indirect byproduct of those project goals. Completed capacity-building (technical assistance) projects (31) have produced 63 permanent jobs per $1 million of EDA funding at an EDA cost of $13,633 per job and a total cost of $19,393 per job\7\.

\7\Direct job creation is an incidental benefit of capacity-building projects, which generally support subsequent projects having direct job creation.

Permanent jobs coming from capacity-building technical assistance projects reflect developments such as stalled businesses being matched with new markets, workers being more employable due to training, and businesses generating more money because they have been made more efficient.

Completed capacity-building (technical assistance) projects have produced 13.7 FTE professional consultant jobs for every $1 million of EDA funding.

Completed capacity-building projects have, in addition, produced adjustment strategies, heightened community involvement and planning, created workable implementation strategies, and undertaken market/feasibility studies. EDA capacity-building efforts have been rated by grantees as seen in the following table:

GRANTEE RATING OF CAPACITY -- BUILDING PROJECTS (Means)

(70 Completed Capacity-Building Projects)*

(Scale of 1-10; 10 = best)

Quality of Adjustment Strategy 8.2
Extent of Community/Business/Government Participation 8.5
Consistency of Implementation Efforts and the Adjustment Strategy 7.8
Quality of Technical Assistance Effort 8.8
Impact of Technical Assistance Effort 8.9
Quality of Feasibility/Market Study 9.1
Impact of Feasibility/Market Study 8.7
*These include all types of capacity-building projects, not just technical assistance.

Grantee Observations:

Across the board, grantees report that the products they are delivering with EDA oversight are both well done and have a significant impact.

Capacity building empowers local areas to respond in a proactive and forward-moving way to the adverse impacts on their economies.

Grantees further report the following:

Capacity-building projects are responsible for significant networking among various forms and levels of economic development agencies. This enables greater use and leveraging of public and nonprofit funds.

Capacity-building projects comprise technology transfer efforts wherein sophisticated methods of enhanced productivity are used to measure business adjustment to new technology.

Revolving Loan Funds

With regard to revolving loan funds (RLFs), 304 jobs have been created per $1 million of EDA funding for 16 completed projects (fully loaned); for

those projects in process (21), there are 247\8\ jobs created. EDA cost per job is $3,312 for completed RLF projects and $4,079 for projects that are in process.

\8\ In-process RLF projects can be analyzed in the same fashion as completed projects because they behave similarly from the time of their first loan onward.

Completed RLF projects have leveraged $115 million in private-sector investment, or $2.5 million per $1 million of EDA funding. In-process RLF projects have leveraged $42 million in private-sector investment, or $2.8 million per $1 million of EDA funding.

Other statistics for RLFs include combined default and write-off rates for completed projects of 13% and for RLF projects in process of 1.9%\9\. For both completed and in-process projects, jobs produced per business assisted are about 22 and 24, respectively. In 50% of the cases the RLF involves a business expansion (as opposed to start-up or retention), and in 67% of the cases it involves the funding of manufacturing firms (as opposed to commercial or service firms).

\9\ A 12-15 percent combined default and write-off rate is well within industry standards for this type of loan.

PROJECT IMPACTS (GENERAL)

Due to the recency of defense adjustment projects, their results are just beginning to become evident. Most will likely contribute significant additional employment growth in the long term.

Defense construction, as well as RLF projects, are nonetheless producing permanent jobs at relatively low costs; capacity-building technical assistance projects are producing smaller numbers of permanent jobs at somewhat higher costs. Capacity-building planning efforts and market/feasibility/reuse studies are perhaps more importantly laying the groundwork for both defense construction and RLF projects. Capacity-building projects could easily be given credit for jobs produced under these two other types of implementation activities.

REVOLVING LOAN FUND PROJECTS (Medians)

(16 Completed and 21 In-Process RLFs)

Completed__In Process
Jobs Per $1M EDA 304__247
EDA Cost Per Job $3,312__$4,079
Private-Sector Leverage Per $1M EDA $2.5 M__$2.8 M
Default/Write-off Rates 13%__1.9%
Jobs Created/Business 22__19

CONCLUSIONS

As reported by grantees, EDA defense adjustment projects are one of the few avenues of flexible assistance available to communities faced with base closures.

EDA funding is critical to most of these types of activities and is usually the primary source of initial funding.

The lower rate for in-process loans reflects almost no write-offs at this stage of the loan.

SECTION III

MULTIPLIER AND EMPLOYMENT-GENERATING EFFECTS OF PUBLIC WORKS PROJECTS

STUDY OVERVIEW

The research described here evaluated the job-producing results of public works investments. It employed nearly 200 input-output and regression analyses to document the effects of Economic Development Administration (EDA) public works projects on the employment growth of their host counties. Comparisons were also made to counties where EDA projects did not take place. Both the input-output and regression analyses sought similar answers: Did public works projects produce attributable permanent jobs in counties where these projects took place? Did the resulting jobs, in turn, produce other jobs? The two analyses are different in that the input-output analysis is constrained by current conditions, whereas the regression analysis allows the current structure of economic activities to change. The first presents a static view of job-creation impacts; the second, a more dynamic view. Both types of analyses are rigorous and standard econometric procedures for determining relationships between public investment and permanent job growth.

INPUT-OUTPUT ANALYSIS FINDINGS

This research examined the role of EDA-funded direct permanent employment and private- sector investment in producing total (direct, indirect, and induced) permanent employment and private-sector investment in counties throughout the United States. In other words, what are the direct employment and private-sector investment multiplier effects? The analysis was undertaken using the IMPLAN Model to generate indirect and induced effects from direct effects, the latter obtained from a national survey of public works grantees. Thus, the national survey generates direct permanent employment and private-sector investment; the IMPLAN Model generates indirect and induced permanent employment and private-sector investment. The sum of direct, indirect, and induced employment and private-sector investment yields total employment and private-sector investment.

Total employment and private-sector investment divided by direct employment and private-sector investment produce "multipliers" of the two direct effects.

Two sets of multipliers are shown in Table 1. These relate to two forms of direct effects_project-related and nonproject-related. The set of lower multipliers expresses total permanent employment and private-sector investment as a function of both forms of direct permanent employment and private-sector investment. The set of higher multipliers expresses total permanent employment and private-sector investment as a function of the solely project-related form of direct permanent employment and private sector investment. The lower multiplier for permanent employment and private-sector investment is the multiplier effect of permanent employment and private-sector investment at the site that the EDA grant specified, as well as other direct employment that located nearby; the higher multiplier for permanent employment and private-sector investment is the multiplier effect of permanent employment and private-sector investment solely at the site that the EDA grant specified.

[Table Omitted]

The multipliers shown in Table 1 are medians for five categories of projects and a weighted median for all projects. The overall median ratio for total permanent employment to both forms of direct permanent employment (project and nonproject related) is 1.50; the equivalent median for total private-sector investment is 1.44. Thus, if an EDA public works project creates 200 direct permanent jobs and $6 million in direct private-sector investment, total permanent jobs (direct, indirect, and induced) amounts to 300 and total private-sector investment to $8.64 million. For employment, there is some minor variation by type of project: Industrial parks exhibit the highest multipliers; water/sewer projects exhibit the lowest. For private-sector investment, there is much more variation by type of project: Buildings have by far the highest multipliers; tourism/marine projects, the lowest multipliers.

The multipliers for total permanent employment and private-sector investment versus only project-related direct permanent employment and private-sector investment are 5 percent higher for employment and 10 percent higher for private-sector investment. This finding indicates the relatively small amount of nonproject-related direct permanent employment and investment compared to project-related direct permanent employment and investment identified in the grantee survey.

REGRESSION ANALYSIS FINDINGS

This research evaluated the role of EDA public works investments in the creation of permanent private-sector employment and in enhancing employee compensation in U.S. counties.

Current models of the effect of infrastructure investment on private-sector productivity have yet to establish a firm connection between the two. Studies using these models often fail to control for the potentially important effect of variations in factor prices, especially wages, in response to public investments. A comprehensive model of county employment effects is provided in this study as a basis from which to view impacts.

The analysis reported here was undertaken using information from the Public Works Program_Performance Evaluation to specify the level of EDA investment in a public works project in a county. The resulting jobs produced in a county reflect the numbers of jobs counted annually as reported by County Business Patterns. Additional regression variables, taken from both County Business Patterns and the U.S. Census of Population and Housing, were used to help identify the independent effect of EDA investment on county employment growth.

The analysis employed multiple regression as the primary econometric technique, with separate equations constructed for both employment and compensation. Variables are expressed in their logarithmic form, reducing the influence of extreme values and enabling a closer fit of the regression planes and higher R2s in both equations. Regressions explain between 80 and 85 percent of the variation in county employment and about 70 percent of the variation in compensation levels.

Empirical results include the following:

EDA investments have a statistically significant and positive effect on county total employment levels (see Table 2).

EDA investments have no statistically discernible effect on compensation per employee. Thus, the resulting EDA jobs are produced at the average wage of all jobs locally.

The elasticity of total employment with respect to EDA investment is estimated at approximately .0074: that is, a 10 percent increase in EDA investment in a typical county ($4,650) is estimated to be associated with an increase of 4.2 jobs. Thus, a $10,000 EDA investment produces approximately 9 permanent jobs.

The cost per job for the EDA program is estimated at just over $1,100 (in 1997 dollars), counting all permanent jobs generated by the facility or the increase in productivity that the facility offers (direct, indirect, induced, and intangible). This estimate is comparable to the findings in both the Public Works Program_Performance Evaluation and the input-output analysis, which found that the cost of a direct permanent job was about $3,000 and that the multiplier for total jobs was about 1.5. But the input-output analysis considered only jobs created by the EDA facility. There are also other jobs created by the new assets themselves, leading to changes in the structure of county economies. Thus, the overall jobs multiplier might be even higher, bringing the cost per job more in line with the regression analysis.

CONCLUSIONS

This study found that EDA's Public Works Program does indeed produce permanent private-sector employment at a relatively low cost. The estimates clearly suggest that the program is having its intended effect. EDA appears to have converted its resources into permanent jobs at prevailing wages in its target counties. These counties are better off than similar counties where this type of effort is not taking place.