Improving Earnings per Job:The New Economic Development Challenge in Kentucky

Paul Coomes

The Kentucky Economic Development Partnership was created by the state legislature in 1992 to oversee industry recruitment, incentive programs, and other efforts to improve the standard of living of Kentucky residents and workers. The national economy was red hot between 1992 and 2000, and our region performed well, particularly in manufacturing assembly and distribution operations. The state border cuts through many large economic markets, most notably those of Cincinnati, Louisville, Evansville- Henderson-Owensboro, Nashville, and Huntington-Ashland. Consistent with the larger bi-state economic areas, I identify nine regions in Kentucky and report on their recent relative economic performances. I find solid job, business, and population growth, especially in the Louisville, Northern Kentucky, and Lexington regions. However, jobs in all regions still pay on average less than in the United States, and the gap widened during the decade. The gap is most likely due to the low rate of college attainment of adults in Kentucky. Preliminary data from the 2000 Census suggest that the college attainment rate of Kentuckians improved during the nineties, but not as much as it improved nationally. I discuss some economic development policy issues to address the new challenges in the state. For example, I suggest that state government institute pay hurdles in its tax incentive programs, so that companies only qualify if they pay employees more than the current average pay in each industry and region.


The decade of the nineties was one of great economic progress around Kentucky. The national economy was the strongest in thirty years. The automobile manufacturing industry continued to favor our region as it moved southward along and around Interstates 65 and 75, and particularly in the center of the state. Kentucky state government invested billions of dollars for transportation and education infrastructure in relatively undeveloped areas. Major air hubs developed, for passengers in Northern Kentucky and for freight in Louisville. Federal welfare reform induced hundreds of thousands of previously dependent persons to join the labor force, and just at a time when the major urban areas were getting desperate for workers.

This strong economic tide has lifted all regional boats in Kentucky. All economic regions in the state have posted gains in businesses, jobs and payroll, personal wealth, per capita income, and local tax bases. Naturally not all regions grew at the same pace, in the same way, or for the same reasons. The Mountain region posted gains in manufacturing employment and earnings per job, but lost on net around one thousand residents since 1992. The Cumberland region, while gaining residents during the last decade, supports jobs that pay on average less than $20,000 per year. The Northern Kentucky, Louisville, and Lexington regions grew the most in terms of nearly all demographic and economic measures. These regions have benefited from the intersections of major interstate highways and the national movement of people and industry from the northeast to the southwest. The Bowling Green-Hopkinsville area posted solid economic and demographic and growth, as did the Paducah-Purchase area and the Owensboro-Henderson area. The Ashland area posted the slowest job growth, and lost one-fifth of its manufacturing jobs, but nonetheless managed to add two thousand residents and to post decent gains in personal wealth.

In this article, I summarize the findings of a study of the recent economic performance of regions in Kentucky, and highlight patterns that might lead to better economic development policies.1 In the study, we describe the economic regions of Kentucky, and analyze a large flow of public data on the performance of these regions since 1992 - the year the Kentucky Economic Development Partnership was created. We also compare the performance in each region with that of similar regions outside Kentucky. Because the state includes such a variety of economic assets and opportunities, it is important to benchmark the regions of the state not just against each other, but to places that have comparable potential. Finally, we examine some of the most critical causes of differential regional economic performance, including topography, demography, and human capital.

Major Findings

The state border cuts through many economic markets, most notably those of Cincinnati, Louisville, Evansville, and Nashville. Any state-level analysis and policy should begin with an understanding of these large bi-state economies. For this study, we found it useful to subdivide the Kentucky into nine economic regions. The regions roughly correspond to labor, housing, and retail markets, or at least Kentucky’s share of those markets. The economic activity inside each region is linked to the distinct topography, infrastructure, and human capital of the region.

Nine Economic Regions in Kentucky


nineregions_crop.gif (13236 bytes)

The three most urbanized regional economies - Louisville, Northern Kentucky, and Lexington - account for 55 percent of the state’s population, 60 percent of the jobs, and 65 percent of the payrolls. The three most urbanized regional economies also have the most educated citizens, as measured by the percentage of adults with a high school or a college degree. However, all Kentucky regions are below the national average in the educational attainment of its adults.

Kentucky added on net almost 300,000 jobs over the 1992-98 period, a gain of 15 percent. The Northern Kentucky area was the fastest growing, with a growth rate of 24 percent. Payrolls of workers rose by $15 billion, or 34 percent, over the 1992-98 period. Nearly three quarters of the payroll growth was generated in the three most urbanized regions. Average annual earnings per job rose from $23,000 to $27,000 per job statewide. However, jobs in all regions still pay on average less than in the nation as a whole. Even in the Louisville region, with its concentration of high paying professional and manufacturing jobs, the pay is still $3,000 below the national average.

Given the wide diversity of topography, infrastructure and economic bases among the nine Kentucky regions, it is not very valid or interesting to benchmark the regions against each other. Thus, we used cluster analysis to discover other regional economies around the United States that were statistically similar to the nine Kentucky regions in 1992 - the year the Kentucky Economic Development Partnership was created. We chose two comparison regions outside Kentucky for each of the nine inside the state. For example, we found that the Lexington economy is statistically similar to that of Columbia, Missouri and Knoxville, Tennessee. The Paducah-Purchase economy is similar to that of Minot, North Dakota as well as the delta counties of northwestern Tennessee. We then compared the economic performance of the Kentucky regions to those ‘peers’. Generally, the Kentucky regions performed much like their peer regions during the 1992-98 period. In terms of growth in population, manufacturing and other jobs, payrolls, and per capita income, the peer regions tracked each other fairly consistently.

Summary Economic Statistics
Nine Kentucky Regions and Comparison Regions

Table 1 (for frames enabled browsers)       Table 1 (no frames)       Table 1 (PDF)

There were some interesting deviations, however. Perhaps the most positive statistical finding is the strong growth in manufacturing employment in the Mountain region of Kentucky, posting a growth rate of 45 percent against a loss of manufacturing jobs in neighboring Appalachian counties of West Virginia and Tennessee, and but an 8 percent growth in the Ozark region of Arkansas. Other Kentucky regions also outperformed their peers in manufacturing job growth, especially Lexington, Owensboro-Henderson, and Northern Kentucky. Overall, Kentucky added manufacturing jobs at twice the rate of the United States over this period.

Growth in Manufacturing Jobs, 1992-1998
Nine Kentucky Regions and Comparison Regions
coomes02fig2.gif (21568 bytes)

Probably the most negative finding from our tracking of peer regions is that Kentucky areas gained so little in terms of average earnings per job. Only the Northern Kentucky, Paducah-Purchase, and Lexington regional economies outperformed their peers in this key measure.


Economic Performance since 1992
Nine Kentucky Regions and Comparison Regions

Table 2 (for frames enabled browsers)       Table 2 (no frames)       Table 2 (PDF)

We have calculated earnings per job by dividing total annual labor and proprietors’ income by the average number of jobs in each region. I believe this is a better measure of regional economic development than per capita income. Per capita income, as a measure of economic development has two problems. First, the numerator includes income from government and other transfer payments, so that a region receiving major infusions of retirement, disability, and poverty aid would post high growth in per capita income whether the local economy was healthy or not. This is not just an academic point in Kentucky, where transfer payments account for 34 percent of personal income in the Mountain region, and 29 percent in the Cumberland region. Second, the denominator of per capita income includes children and retirees, i.e., the entire population, not just those in the labor market. Thus, ceteris paribus, a region with many large families would appear poorer on a per capita basis than one with small families.

The modest performance in terms of earnings appears to be directly related to the low educational attainment of adults in Kentucky. The accompanying chart shows clearly the positive relationship between education and earnings for the nine Kentucky regions and the eighteen peer regions around the US. Note that all Kentucky regions fall below the US average in terms of both college attainment and earnings per job, but that the greater the education level the higher the earnings.

Education and Earnings per Job
Nine Kentucky Regions and Comparison Regions
coomes02fig3.gif (4175 bytes)

Factors other than formal education are no doubt at work Many have pointed to the low cost of living in Kentucky as a reason that wages and salaries are lower here2 . However, there is no reason to believe that the cost of living is lower in the Appalachian mountain area of Kentucky than the mountain regions of Tennessee or West Virginia. The cost of housing, utilities, groceries, clothing, and transportation do not vary much among the peer regions. Moreover, Nashville reports a lower cost of living index value than does Louisville or Lexington. If taxes on households were taken into account, Nashville would appear to be an even cheaper place to live. My impression is that, here in the middle of the US, the measured local cost of living increases with the degree of urbanization. Population density raises real estate and transportation costs (parking, pollution costs), while the price of most retail items varies little from city to city, or urban to rural.

Geographic and demographic differences also may help explain the variation in earnings per job. High pay in coal mining keeps overall earnings per job in Kentucky’s Mountain region above that in the mineral-poor Ozarks region of Arkansas, even though adults in both regions have relatively little formal education. The Ozark region, on the other hand, is a mecca for retirees and thus has posted strong job growth in the lower-paying service industries. The Lexington region has benefited greatly by its location at the center of the US population (east of the Rockies), picking up almost one-half of the net manufacturing job growth for the entire state of Kentucky during the study period. These jobs on average pay well above those in other local industries, giving the Lexington region one of the strongest performances in terms of earnings per job and per capita income.

Policy implications

There is little state or local governments can do about geographic and demographic factors in the economic development competition among regions. However, they can affect other dimensions of the competition. I consider three here – education, taxes and business climate, and incentives.

Kentucky has embarked on an ambitious plan to rationalize its public higher education system. The state has declared two universities to be research universities, with special funding initiates and high expectations for scholarship, reputation, patents, royalties, and ultimately business formation. Given the premium on knowledge in the economic environment of the last twenty years, and the major investments made in the research universities in surrounding states, this is long overdue. Regarding the educational attainment of Kentucky residents, there is a limit to what state policy can accomplish. Inexpensive higher education is available in nearly every corner of the state. However, enrollments and the number of graduates has been relatively flat over the past decade, partly due to the paucity of residents in college-going age cohorts, and partly to the lure of work in a strong economy. Initial indications from the 2000 Census suggest that the rate of college attainment for Kentucky adults improved from the 1990 Census, but actually lost ground against the national average. A more fruitful approach to raising the human capital in Kentucky might be to attract more educated persons to move here.

A case can be made that Kentucky has a good business climate for mobile, capital-intensive firms needing modestly educated and inexpensive workers, but a bad business climate for mobile persons with high human capital, entrepreneurial instincts, and high wealth. State taxes on individual income and expenditures are among the highest in the US, and higher education is only now getting the focused attention it needs to make Kentucky more competitive in terms of innovation, business formation and wealth creation. Decades of aggressive redistribution of income around the state has led to a poor business climate for those that might generate wealth and an entitlement culture for many of those that receive the payments and services. Talented and ambitious people leave, exacerbating the problem. This is especially relevant in a state like Kentucky, where two-thirds of the population live, work, and shop along and within the border of one of the seven adjacent states. Indeed, the biggest urban economy in Kentuckiana is Cincinnati, and the fastest growing city in the region is Nashville. The internet and the interstate have made state boundaries relevant only to tax collectors and regulators. State tax and expenditure policy needs to reflect these geographic and market realities.

Finally, Kentucky state government should reevaluate its incentive programs. In response to the rustbelt shakeout, unemployment and out-migration of the early 1980s, state officials devised a set of industrial incentive programs that have largely worked. KREDA, KIDA, and KJDA effectively lower the state tax liability on mobile firms while retaining statutory tax rates for immobile firms and workers. These programs have been used to attract hundreds of manufacturing plants, distribution centers, and office operations. Today, with essentially full employment, a growing population, and lagging pay scales around the state, the economic development challenge is different and calls for new programs. To raise earnings per job, we need to attract more scientists, engineers, health-related researchers, and entrepreneurs. That is, rather than attracting jobs for our people, we need to attract and retain smart and industrious people to create jobs and wealth. Our tax structure and incentive programs should reflect this goal. One simple improvement would be to prorate tax incentives to companies, with incentives rising for companies that pay workers above the current average in each industry and region. State policy would then clearly support raising earnings per job, rather than just encouraging more jobs and investment.


  1. "The Recent Economic Performance of Regions in Kentucky", by Paul Coomes and Michael Price, for the Kentucky Economic Development Cabinet, University of Louisville, May 2001, 67 pages. Available in pdf format at

  2. Berger and Bloomquist estimate that Kentucky moves up four ranks, from 40th to 36th, among states in terms of per capita income when adjustments are made for differences in cost of living and quality of life. However, none of the states Kentucky passed in the ranking have economic regions that we found to be most comparable to those in Kentucky. See "Kentucky’s Per Capita Income: What Should the Goal Be?", by Mark Berger and Glenn Bloomquist, Kentucky Annual Report 2000, pages 1-7. Also, the study examined per capita income rankings, not earnings per job rankings, so are not directly applicable to this discussion.

This document is a of CBER, Center for Business and Economic
Research, located at the University of Kentucky, Gatton College of Business and Economics.