Stay Connected

Follow ERS on Twitter
Subscribe to RSS feeds
Subscribe to ERS e-Newsletters.aspx
Listen to ERS podcasts
Read ERS blogs at USDA

Impacts of Regional Approaches to Rural Development: Initial Evidence on the Delta Regional Authority

by John Pender and Richard Reeder

Economic Research Report No. (ERR-119) 73 pp, June 2011

What Is the Issue?

Regional development approaches are attracting increasing attention, particularly as vehicles for encouraging rural economic development. At the Federal level, the Denali Commission was authorized in 1998 to promote regional development in Alaska and the Delta Regional Authority (DRA) was authorized in 2000 to do the same in the Mississippi Delta region. Since then, four additional regional development commissions have been authorized, and startup funds have been appropriated for two of these. Despite increased emphasis on such regional approaches, evidence of their economic impacts is limited, especially for newer programs such as the DRA. In this study, we demonstrate an approach to investigating the initial economic outcomes of such programs, using publicly available data and the best available methods to examine the DRA as a test case. We focus on changes in personal income per capita (and its components), employment per capita, and population from 2002 to 2007.

What Were the Study Findings?

The DRA began funding projects in 252 economically distressed counties in the Mississippi Delta region in 2002. In its first 7 years, the DRA invested $75 million in basic public and transportation infrastructure, business development, job training, and employment-related education. Growth in annual personal income per capita averaged about $600 higher in DRA-recipient counties from 2002 to 2007 than in economically and demographically similar non-DRA counties in the Delta Region and in the Southeast. This greater growth represented an additional increase in per capita income in DRA-recipient counties (relative to similar non-DRA counties) of about 3 percentage points over this 5-year period. Comparison of trends in per capita income in the matched groups of counties showed that these trends were very similar from 1990 to 2002, but began to diverge after the DRA began operating in 2002.

The major sources of greater growth in personal income were greater growth in per capita net earnings and personal transfer payments, both of which were statistically significantly greater in DRA counties. (Transfer payments are payments from a Government to an individual, e.g., Medicare, Social Security, etc.) These impacts were greater in counties where DRA spending per capita was larger, with each $1 of additional DRA spending per capita associated with an additional $15 in growth of personal income per capita, including $8 in additional net earnings (primarily in the health care and social service sector) and $5 in additional transfer payments (mainly medical transfer payments).

The incremental impacts of DRA spending on personal income, earnings, and transfer payments suggest that the DRA may be leveraging additional public or private sources of funds. In particular, the DRA's health programs, including funding of medical facilities and its J-1 visa waiver program to attract foreign doctors, appear to be contributing to increased health sector earnings and medical transfer payments by increasing the availability of health care services.
The DRA's health awareness campaigns, such as those focusing on diabetes prevention and treatment, may also be increasing the demand for health services. Since other public funds, such as medical transfer payments, are apparently being leveraged by the DRA, the income increments associated with DRA spending cannot be seen solely as a return to DRA investments.

We did not find statistically significant differences in growth of employment per capita between DRA and similar non-DRA counties; this result may be due to the difficulty of measuring those impacts given the small size of the program and the relatively short timeframe considered. We found some evidence of slower population growth in the DRA counties, but this difference was found to be a continuation of differences in trends in population growth prior to initiation of the DRA.

(For a list of DRA counties, see:

How Was the Study Conducted?

We used a quasi-experimental matching approach to select non-DRA nonmetropolitan counties in the Delta region and elsewhere in the Southeastern United States that had similar economic and demographic characteristics to DRA recipient nonmetropolitan counties prior to implementation of the DRA and compared mean changes in outcomes between these groups of counties. We also used multivariate regression analysis on the matched groups of counties to identify the effects of the level of DRA program spending per capita on the outcomes. The findings are robust to alternative methods of selecting the comparison groups of counties; use of alternative sets of variables for matching; including or dropping groups of counties for which confounding factors were present (such as counties heavily affected by Hurricane Katrina or the presence of other development or health programs); and use of alternative starting and ending years in the comparisons.




Last updated: Sunday, May 27, 2012

For more information contact: John Pender and Richard Reeder