Kentucky’s Per Capita Income: Catching Up to the Rest of the Country

Mark C. Berger

A goal among many leaders in Kentucky is to see the state’s per capita income equal or exceed the national per capita income average. Although Kentucky has narrowed the income gap recently, its per capita income still stands at only 81 percent of the national average. Matching this national level would require significant changes in Kentucky. The state would need large increases in the number of high school and college graduates in the state, and/or in the percentage of private sector employment per capita. Based on previous rates of increase, it will still be many years before Kentucky’s per capita income is equal to the national average.

Introduction

A frequently used indicator of a state’s economic health is per capita income. Historically, Kentucky’s per capita income has been below that of the U.S. average, although that gap has narrowed in recent years. In 1995, per capita income in the U.S. stood at $23,208 in 1995 while in Kentucky the level was $18,849.1 Many believe that an important goal for Kentucky is to narrow the gap between its income and that of the rest of the country. Kentucky Governor Paul Patton, in a recent speech to the Hopkinsville Chamber of Commerce, said that his goal was to see per capita income in Kentucky above the national average.2 Although this may be a lofty goal, there is cause for optimism given the recent history of income levels in Kentucky. Indeed, while per capita income in Kentucky stood at only 78.3 percent of the national average in 1985, by 1995 it had increased steadily to 81.2 percent of the national average.

In this article, I examine long-term trends in Kentucky’s per capita income relative to the national average. In the process, I address several questions: 1) Has the recent increase in Kentucky’s per capita income relative to the U.S. average been part of a long-term increase or has it been confined to more recent years? 2) Has Kentucky’s experience mirrored that of other states, or has it been unique? 3) What determines differences in per capita income at the state level? 4) Can these determinants explain why Kentucky’s per capita income is below the national average? 5) What can explain the increase in Kentucky’s per capita income relative to the national average in recent years? 6) How different would Kentucky have to be today to be at the national average of per capita income? 7) How long will it take for Kentucky to reach the national average per capita income?

Per Capita Income as a Measure of Well-Being or Standard of Living

Per capita income is often used by policymakers and the public as an overall index of well-being or standard of living in an economy. Thus, before proceeding with the analysis, it is important to examine what per capita income measures and to look at its strengths and weaknesses as an indicator of economic well-being.

Personal income data are collected by the U.S. Department of Commerce’s Bureau of Economic Analysis as part of the National Income and Product Accounts. These data comprise wage and salary disbursements, other labor income, proprietor’s income, rental income of persons, personal dividend income, personal interest income, and transfer payments to persons (e.g., Social Security, Aid to Families with Dependent Children, etc.). The majority of personal income comprises wage and salary disbursements, followed by transfer payments to persons and personal interest income. Table 1 shows the 1995 breakdown of personal income into its components for the U.S. and Kentucky.

TABLE 1
Personal Income and Its Components, U.S. and Kentucky, 1995

Kentucky1

U.S.2

Amount     Percent

Amount    Percent

Wage and salary disbursements

40,644,369

86%

3,423,330

85%

Other labor income

5,476,497

12  

423,799

11   

Farm proprietors’ income

5,282,519

11  

19,529

0   

Nonfarm proprietors’ income

623,446

1  

449,257

11   

Less: contributions for social insurance

-3,650,670

-8  

-294,013

-7   

Less: adjustment for residence

-250,831

-1  

-873

-0.02

Net earnings by place of residence

47,501,884

65  

4,021,029

66   

Dividends, interest, rent

10,879,281

15  

1,054,107

17   

Transfer payments

14,380,955

20  

1,022,841

17   

Total personal income

72,762,120

100  

6,097,977

100   

Population (000s)

3,860

262,755

Per capita income (dollars)

$18,849

  

$23,208

  

1 In thousands of dollars unless otherwise noted.

2 In millions of dollars unless otherwise noted.

Source: U.S. Department of Commerce, Bureau of Economic Analysis,

Thus, personal income is just the total amount of income earned or disbursed to individuals in the economy in one form or another in a given year. Individuals then use this personal income to purchase goods and services, pay taxes, or place in savings or investments. It is thus a broad-based measure of economic well-being for the economy. Per capita personal income is simply the total personal income divided by the total population, which gives a per person measure of the income earned or disbursed to individuals in the economy. As a result, per capita income adjusts for population differences over time or across states.

The chief limitation of personal income as a measure of well-being is that it does not measure activities or things that people value that are not traded in the marketplace. For example, environmental quality or other amenities are not reflected in personal income, nor is the value of leisure time or the value of services provided inside the household. Nevertheless, personal income covers a broad base of economic measures better than any other indicator. For instance, another indicator such as the unemployment rate only gives the percentage of persons without work, not the well-being of those with work. Similarly, the employment rate tells the percentage of persons that are working but not the earnings of those workers. On the other hand, average wages would provide the earnings of workers but not the income non-workers have at their disposal. Consequently, personal income is the best measure of economic well-being that is readily available.

Per Capita Income in Kentucky Relative to the U.S.

Figure 1 shows the ratio of per capita income in Kentucky to the U.S. average from 1929 to 1995, the entire time period for which per capita income data are available from the National Income and Product Accounts. Two series are shown in Figure 1: the first spans the period from 1929–94, and the second shows the new series recently published by the Bureau of Economic Analysis that covers the period from 1969–95 but is not comparable to the earlier series.3

berger97fig1.gif (7464 bytes)

Figure 1 tells an interesting story. Per capita income in Kentucky relative to the U.S. average rose steadily until about 1979 or 1980, exhibiting the long-run convergence familiar to regional and growth economists. For instance, Barro and Sala-i-Martin argue that marginal returns to capital may be higher in states with low income levels, and thus growth may be higher, promoting convergence.4 Convergence may also occur if there is mobility of businesses and workers across states. Businesses will tend to migrate where land and labor costs are lower, expanding economic activity and raising per capita income. In contrast, workers will tend to migrate where wages are higher, increasing the supply of workers in certain areas and exerting downward pressure on income. The net effect of such mobility would be an equalizing of incomes across states and higher rates of growth in per capita income observed in low income states.5

In the long run, with such mobility of businesses and workers, incomes would be completely equalized across states except for differences reflecting location-specific factors. Blomquist, Berger, and Hoehn examine such differences due to location-specific amenities such as climate, air and water quality, and other natural conditions.6 For example, if people find Kentucky to be a pleasant place to live because of its climate or natural features such as rivers or mountains, then per capita incomes may remain below the national average; in other words, Kentucky residents are willing to accept a lower income to live in a desirable location. Per capita incomes in undesirable locations would lie above the national average to compensate individuals for living in unpleasant conditions. Nonetheless, excepting location-specific amenities, both growth theories and regional models of economic behavior predict an eventual convergence of per capita income for Kentucky and the U.S.

Contrary to the long-run pattern of convergence, however, Kentucky’s relative per capita income fell rather sharply in the early and middle 1980s. This fact suggests that the recession and economic restructuring of that period affected income in Kentucky more than in the rest of the country.7 Since about 1985, though, Kentucky’s per capita income has been rising relative to the national average, so that the state’s relative income now stands approximately at its 1979–80 level. Viewed in this light, the recent increase in Kentucky’s income has represented a catching up to a level relative to the national average that had been reached previously.

What will the future hold and how quickly can we expect Kentucky’s per capita income to converge to the national average? We can get some clues about the process of convergence by looking at the experiences of other states. I turn to this analysis in the next section.

Kentucky’s Experience Compared to Other States

Has this convergence to the national per capita income average been unique to Kentucky, or has it occurred in other states? Table 2 shows that convergence has been proceeding on a nationwide basis regardless if considering the entire period of available data (1929–94) or the last 10 years. This table shows the average change in the ratio of state to U.S. per capita income, both for those states that began each time period above the national average and those that began below the national average. As would be expected from convergence, the average change for those states above the average is negative and positive for those below the average. States like Kentucky that are below the national average are catching up over time and those above the national average are falling toward it. Figure 2 focuses on the experience of Kentucky and surrounding states over the last 10 years. It shows that the pattern of convergence to the national average has also occurred in states neighboring Kentucky.

berger97fig2.gif (7476 bytes)


TABLE 2
Convergence of States’ Per Capita Income to
U.S. Average, 1929–94 and 1985–95

 

Time Period

Number
of States

Average Change in Relative Income

States above U.S. average, 1929

1929–94

14

-0.1780  

States below U.S. average, 1929

1929–94

34

0.1825

States above U.S. average, 1985

1985–95

17

0.0004

States below U.S. average, 1985

1985–95

33

0.0092

Source: U.S. Department of Commerce, Bureau of

Economic Analysis, unpublished data.

As Kentucky’s relative income has risen, has its per capita income ranking among the states changed? Figure 2 shows that there has been no change in rankings over the last 10 years among surrounding states. Table 3 shows the top 10 and bottom 10 states in per capita income rankings in 1985 and 1995, expressed in terms of income relative to the U.S. average. Table 3 shows that even though convergence to the national average has been occurring, the state rankings change slowly. Kentucky was ranked 44th in per capita income in 1985, and after 10 years of convergence, it had only moved up to 43rd by 1995.

TABLE 3
Top 10 and Bottom 10 States Ranked by Personal Per Capita Income (PCI)
Relative to U.S. Average, 1985 and 1995

1985

1995

State

Relative PCI

State

Relative PCI

1

Alaska

1.31

Connecticut

1.37

2

Connecticut

1.28

New Jersey

1.29

3

New Jersey

1.24

Massachusetts

1.21

4

Massachusetts

1.17

New York

1.19

5

New York

1.15

Maryland

1.13

6

California

1.15

Delaware

1.13

7

Maryland

1.14

New Hampshire

1.10

8

New Hampshire

1.10

Illinois

1.09

9

Delaware

1.07

Hawaii

1.06

10

Illinois

1.06

Nevada

1.05

  

41

South Dakota

0.80

Idaho

0.81

42

Montana

0.79

Kentucky

0.81

43

Idaho

0.79

North Dakota

0.80

44

Kentucky

0.78

Oklahoma

0.80

45

South Carolina

0.78

Montana

0.79

46

Alabama

0.77

Utah

0.79

47

Utah

0.76

New Mexico

0.78

48

Arkansas

0.76

Arkansas

0.78

49

West Virginia

0.74

West Virginia

0.76

50

Mississippi

0.66

Mississippi

0.72

Source: U.S. Department of Commerce, Bureau of Economic Analysis, unpublished data

What Determines a State’s Per Capita Income?

On the most basic level, factors that affect per capita income are those which raise or lower the amount of income a person receives in a state. One such set include factors which raise or lower the productivity of the labor force. Most obvious among these is the level of education. Workers in states with higher levels of education among their residents will earn more in the labor market and thus increase those states’ per capita income. Not only productivity, but employment of workers in general will be a very important factor affecting per capita income across states. States with a higher percentage of their population working will have more people earning wages and salaries and thus are likely to have a higher per capita income. In addition, whether the state is primarily urban or rural will have an impact on the model. Rural states will have a disproportionate number of individuals working in agriculture, where wages and incomes will tend to be lower. Thus, the very nature of the jobs in rural states will tend to hold down per capita incomes.

I have constructed an econometric model of per capita income that explains variation in income across states in 1995. After experimenting with several different combinations of variables which account for the factors discussed in the previous paragraph, I have specified five variables that do a good job in explaining differences in per capita income across states.8 Table 4 shows these variables and the results of the estimated econometric model. This table also shows the average values of the variables across all the states and the Kentucky values of the variables which will help explain why Kentucky’s income is below the national average.

TABLE 4
Econometric Estimates Explaining Per Capita
Income by State, 1995 a

  

Estimated

Kentucky

Average

Variable

effect b

value

of states

% of population over 25 & high school graduate

0.0096 *

31.7    

30.9    

% of population over 25 & college graduate

0.0208 *

13.6    

20.0    

Private sector employment per capita

0.7679 *

0.4361

0.486  

Public sector employment per capita

-0.4528

0.0832

0.0965

% of population living in rural areas

-0.0039 *

48.1    

31.1    

Intercept

9.0830 *

Log of per capita personal income

9.844  

10.00  

a  The dependent variable is the natural log of per capita personal income. Fifty-one observations (including the District of Columbia) were used in the analysis. The R2 for the estimated model is 0.7615.

b  A * denotes statistical significance at the 5 percent level in a two-tailed test.

Source: U.S. Department of Commerce, Bureau of EconomicAnalysis,  unpublished data.

From these econometric estimates, the following conclusions can be drawn about the determinants of per capita income across states: States with higher education levels, as measured by the percentages of the population over age 25 that are high school and college graduates, have higher per capita incomes. States with higher private sector employment per capita also have higher income per capita. Interestingly, states with higher government employment per capita, holding other variables constant, have lower per capita income. This finding suggests that improvements in per capita income are more likely to be obtained if job growth comes from the private rather than the public sector. Finally, as expected, states with higher rural populations have lower per capita incomes.

Why is Kentucky’s Per Capita Income Below the National Average?

The results of the econometric model can be used to explain why Kentucky’s per capita income level is below that of the average across all states. This is done by calculating the differences in the predicted per capita incomes arising from differences in education levels, employment per capita, and the percentage of population that is rural between Kentucky and the U.S. Figure 3 shows this calculation. We see that 57 percent of the difference between Kentucky’s predicted per capita income and the predicted average of the states’ per capita incomes is due to education differences — primarily Kentucky’s low percentage of college graduates among the population age 25 and over. That Kentucky is a much more rural state than average accounts for 29 percent of the difference, and the remaining 14 percent comes from the fact that Kentucky’s employment per capita is lower than the average of the rest of the states.

berger97fig3.gif (7000 bytes)

Thus, the lion’s share of the difference arises from the lower education levels in Kentucky compared to the average of other states. If education levels were higher, Kentucky’s per capita income would be closer to the national average. In fact, the model suggests that if Kentucky’s education levels were equal to the national average, 57 percent of the gap between Kentucky’s per capita income and the national average per capita income could be closed.

Kentucky’s Per Capita Income from 1985 to 1995

In considering why Kentucky’s per capita income has risen relative to the rest of the country from 1985 to 1995, we need to look for trends in Kentucky that are different from the rest of the country. Education levels have been improving over time both in Kentucky and in the rest of the country, so education cannot explain the rising per capita income in Kentucky. Similarly, there has been a small decline in the percentage of the population living in rural areas in both Kentucky and the rest of the country. That leaves employment/population changes.

While the recession of the early 1980s was particularly hard on Kentucky, the opposite was true for the recession of the early 1990s. Kentucky barely felt that recession, and since then, job growth has been stronger in Kentucky than in many other places. At the same time, population growth in Kentucky has not been as strong as in the rest of the country. These two factors combined imply that employment per capita has been rising faster in Kentucky than in the rest of the country. Figure 4 shows the changes in private employment per capita in Kentucky and for the U.S. From this figure it is apparent that private employment per capita has been increasing faster in Kentucky than in the rest of the country, and this difference may be partially responsible for the relative gain in Kentucky per capita income from 1985–95. This employment growth has in part contributed to the resumption in the convergence of Kentucky’s per capita income to the U.S. average so that it is now back to the level it was before the recession of the late 1970s and early 1980s.

berger97fig4.gif (9453 bytes)

Making Them Equal

We can use the results of our econometric model to construct scenarios under which Kentucky’s per capita income would be equal to the U.S. per capita income. We must ask how different Kentucky’s characteristics must be for the state’s per capita income to be equal or greater than the U.S. average. In Table 5, I consider three different scenarios that might accomplish this goal. The first scenario increases Kentucky’s education levels until the predicted per capita income from the model matches the national average. Under this scenario, Kentucky would have the same number of jobs, but its workers would be more educated and hence more productive, all of which would raise incomes. The second scenario increases private sector employment per capita, increasing the number of jobs while holding education levels constant. More jobs might exist because there are more employers in the state, or labor force participation rates, which are lower in Kentucky than in most other states, might rise. In the third scenario both education levels and private sector employment per capita are raised. All three scenarios hold constant the percentage of the population living in rural areas and the number of government jobs per capita.

TABLE 5
Changes in Kentucky’s Education Levels and Employment Required
for Per Capital Personal Income to be Equal to or Greater than U.S. Average, 1995

Characteristic

Scenario 1

Scenario 2

Scenario 3

% of population 25 and over & high school graduate

20%

10%

% of population 25 and over & college graduate

50%

25%

Private sector employment per capita

60%

30%

Source: Calculated from results shown in Table 4.

Scenario 1 means Kentucky would have a 50 percent higher percentage of the population age 25 and over with a bachelor’s degree or higher and a 20 percent higher percentage of high school graduates. Kentucky would then lie almost exactly at the average of the other states for the percentage of college graduates (20.4 percent vs. 20.3 percent) and well above the average of the other states for the percentage of the population that are high school graduates that did not attend college (38.0 percent vs. 30.9 percent). In fact, such a 20 percent increase in the percentage of the population that are high school graduates only would place Kentucky ahead of all other states, including Pennsylvania, where 38.7 percent of the population age 25 and over are high school graduates.

 

Scenario 2 would correspond to a 60 percent increase in the number of private sector jobs per capita. This would put Kentucky far above the average of the other states. In fact, only the District of Columbia would have a higher number of private sector jobs per capita and many of its jobs are held by commuters who do not live in the District.

Scenario 3 corresponds to increases in education levels and private sector employment per capita that are half the sizes of those in Scenarios 1 and 2. Such a combination of characteristics would give Kentucky a percentage of high school graduates similar to Nebraska and Vermont, a percentage of college graduates the same as Wisconsin and Idaho, and a private sector employment per capita similar to Nevada and Colorado. In general, the scenarios show that, to have a per capita income level equal to the national average at present, Kentucky would need a far different economy and a much more educated workforce.

How Long Will it Take?

Following the scenarios presented above, Kentucky would require a long time to catch up to the average U.S. per capita income. It might take a generation to raise education levels as much as needed, and, if education levels were rising at the same rate in the rest of the country as well, per capita income in Kentucky would not rise at all relative to the national average. On the other hand, the process of regional convergence, where capital and labor flow to areas with the highest return, should naturally raise per capita income in Kentucky relative to the rest of the country, as it has done in the past.

How soon should we reasonably expect this convergence? Looking at the long-term trends in Kentucky’s per capita income relative to the U.S. average, we can see that it took over 30 years to increase Kentucky’s relative per capita income from approximately 60 percent to 80 percent of the national average. To obtain more precise estimates of the rate of convergence, I have estimated regression models of Kentucky’s relative per capita income over various time periods and reported the results in Table 6. As can be seen the estimates range from a predicted increase of 0.0045 per year (0.45 percent) over the entire 1929–94 time period of the old series to 0.0060 (0.60 percent) per year estimated from 1929–79. These estimates can be used to predict how long it will take Kentucky to move from its current level of 81.2 percent of U.S. per capita income to 100 percent of the U.S. level. Using the highest estimated rate of convergence (0.60 percent), Kentucky will catch up to the national average in 31 years and will reach 90 percent of the national average in 15 years.

TABLE 6
Estimated Rates of Convergence and Number of Years until Kentucky
Per Capita Income Equals U.S. Average Per Capita Income

Time period of estimation

Estimated annual convergence rate

Number of years until equality reached

Number of years until 90% of U.S. average reached

1929–94

0.45%

42

20

1929–79

0.60%

31

15

1986–95

0.51%

37

17

Source: Calculated using U.S. Department of Commerce, Bureau of Economic Analysis
unpublished data.

Using any of the three estimates, it is clear that the convergence of Kentucky’s per capita income to the national average is a long-run process and difficult to accomplish overnight. Even if Kentucky were to increase the highest estimated long-run rate of convergence by 50 percent, it would still take 21 years for the state to reach the national average level of per capita income.

Conclusion

Will Kentucky in fact reach this national average? Probably, given the progression toward convergence that has been and is still occurring in the U.S. Of course, if Kentucky is a desirable place to live and work, it may never completely reach the national average because residents will accept lower incomes to live here. Based on past trends of convergence, it will take many years for Kentucky’s per capita income to reach the national average. The process could be accelerated, but it would be difficult. It would require that education levels or jobs grow faster than the national average, which may be difficult for Kentucky to sustain.

Footnotes

  1. U.S. Department of Commerce, Bureau of Economic Analysis, “Total Personal Income and Earnings by Industry (SA05) 1969-1995,” unpublished data files, September 1996.

  2. Governor Paul Patton’s speech to the Hopkinsville Chamber of Commerce, July 23, 1996. Source: Office of the Governor Press Office, September 3, 1996.

  3. U.S. Department of Commerce, Bureau of Economic Analysis, “Total and Per Capita Personal Income by State and Region,” Survey of Current Business 76 (May 1996): 94-101.

  4. Robert J. Barro and Xavier Sala-i-Martin, “Convergence,” Journal of Political Economy 100 (April 1992): 223-251.

  5. These and other reasons for convergence are discussed in Edward Nissan and George Carter, “Income Inequality Across Regions Over Time,” Growth and Change 24 (Summer 1993): 303-320, and Rajiv Mallick, “Convergence of State Per Capita Incomes: An Examination of Its Sources,” Growth and Change 24 (Summer 1993): 321-340.

  6. Glenn C. Blomquist, Mark C. Berger, and John P. Hoehn, “New Estimates of Quality of Life in Urban Areas,” American Economic Review 78 (March 1988): 89-107.

  7. James S. Fackler, “Economic Overview: National and State Economic Activity,” Kentucky Annual Economic Report, University of Kentucky Center for Business and Economic Research (1995): 43-48.

  8. I also tried to add other variables to the model, but with the five variables already included, these variables were not statistically significant. Most notable among these were variables measuring the age distribution of the population, such as the percentage of the population under age 18 and the percentage over age 65.

University of Kentucky home pageCarol Martin Gatton College of Business and Economics home pageCenter for Business & Economic Research home page