IV. Impact on Economic Productivity
One approach to measuring higher education’s effects on the state’s economy is through the economic gains resulting from improved productivity. This relationship directly addresses the crucial economic question of what higher education does to expand the productive capacity of the economy—quite apart from the dollars injected into the system generating spending and re-spending through the multiplier effect.
This question may be the most crucial of all because much of the support higher education receives comes from the government. This means that there is no market mechanism to directly assess the value of this spending. The importance of this lack of a market allocation mechanism is best demonstrated by considering the dilemma of government sponsored ditch digging. Public money spent digging a ditch will most certainly ripple through the economy creating many jobs directly and indirectly supported by the work. The same can be said of public funds used to fill up the ditch when it is completed. But the telling question is how much the productive capacity of the overall economy increased after paying for this work?
Productivity and Economic Growth
According to traditional economic growth theory, growth occurs by employing more basic inputs—more labor, capital or land. This approach, however, presents at least some circular logic. If we employ more basic inputs, we produce more output. The more fundamental question is how to get the additional resources to fuel the increased production?
One key to this process is utilizing the existing resources more efficiently. For example, if instead of taking 200 hours to build one car, a change in the assembly procedures now requires only 100 hours, then the labor supply effectively has been doubled. Thus, streamlining productivity—getting more done each hour of work—-has the effect of increasing the labor supply just as genetically engineering plants to produce more cotton effectively increases the “supply” of land.
Labor can become more productive if:
- there is more physical capital employed per worker (physical capital includes equipment and structures as well as public infrastructure);
- the health or skills of human beings increases—this is known as human capital;
- or, the stock of accumulated abstract knowledge grows, thereby increasing knowledge capital.
The primary economic impact of higher education is through augmenting the knowledge and skills of the workforce, although institutions of higher education also serve as repositories and transmitters of knowledge capital. The chief distinction between human capital and knowledge capital is that human capital cannot be separated from the human who possesses it.
Universities, colleges and other higher education institutions are crucial in improving productivity since they produce two kinds of capital—human capital and knowledge capital—reflecting their dual roles as both educational and research institutions. In practice, it is extremely difficult to measure separately the impacts of these two roles. For example, is it the technological sophistication and research skills of the faculty at a highly-respected engineering school that draws industry to the area? Or is it the skills and abundance of students at the institution that the industry hopes to employ that is the drawing card? Undoubtedly, both matter. In most cases, however, the teaching and research functions of most universities are joint products and attempting to measure the effects of just one of these functions on the local economy likely will result in measuring some of both.
This analysis focuses on the role of Texas institutions of higher education in augmenting the amount of human capital in the economy. That role is a particularly crucial one for the state economy. If our institutions stopped educating students, the flow of human capital into the economy would diminish almost instantaneously, barring massive out-migration of Texas students to institutions in other states followed by reverse migration back into the state.
The instantaneous impact of this imaginary shut-down of higher education on the amount of human capital available to the Texas economy contrasts to the longer-term effects it would have on the stock of knowledge or research spin-offs. Clearly, the sum of human knowledge would not evaporate if Texas’ higher education institutions ceased operating. But over the long term, the Texas economy would surely suffer.
Labor Productivity Responses to Education
There have been a few studies in the United States attempting to measure the effects of human capital investments on productivity, and many stumble with subjective measures of output, making it difficult to generalize their results. Other efforts, such as a study of training investments by Ann Bartel, have been able to measure more objective productivity response gains from education and training. But a low, six percent survey response rate makes relying on these estimates very uncertain.
Recent investigations have employed a more substantial survey base, tying data from the National Center on the Educational Quality of the Workforce to output, sales and other firm-level economic data from the US Bureau of the Census’ Annual Survey of Manufactures. This broad database generated responses from 3,358 business establishments representing a 64 percent survey response rate. From this data, Sandra Black and Lisa Lynch of the National Bureau of Economic Research estimated that a 10 percent increase in the average educational level of workers resulted in a 4.9 percent to 8.5 percent increase in productivity in manufacturing and a 5.9 percent to 12.7 percent productivity improvement in non-manufacturing industries. In this sample, manufacturing workers averaged 12.5 years and non-manufacturing workers averaged 13 years of education.
Productivity-Response Function for Texas Higher Education
For the purposes of this analysis, it was assumed that the productivity-response function appropriate for higher education lies at the low end of Black and Lynch’s estimates. That is, a 10 percent increase in the average education of a worker would result in a 4.9 percent productivity gain in manufacturing or a 5.9 percent gain in non-manufacturing. The use of these low productivity-response relationships is the result of three considerations.
First, the average education of the workers in the sample employed by Black and Lynch is below that of the average person in Texas’ higher education system. While this in itself would not dictate lowering the productivity-response function, if Black and Lynch had estimated their models using a sample of just those with some college or more, it is likely that their measured productivity responses to increasing education would have been slightly lower because there is a declining effect on productivity of increased education. Accordingly, we expect a slightly lower effect than found by Black and Lynch because we are applying it to a more highly educated population.
Second, Black and Lynch were concerned with what the productivity effects of increased education were and not how these effects were generated. The distinction, reflected in the “alpha factor” discussed in Chapter III, concerns how the effects of additional education are influenced by the institution, as opposed to the natural abilities of the students. In the same way estimates of the wage gains attributable to the institution of higher education must be lowered to reflect the natural abilities of the students, so should the productivity-response function.
Finally, in order to produce the most reasonable estimates, it is necessary to err on the conservative side in producing estimates of the effects of higher education on productivity. The decision to choose methods that will not overstate results leads us to assume a productivity-response function that should be at the lower end of Black and Lynch’s estimates.
Increasing Texas’ Human Capital Stock through Higher Education
To translate the productivity-response function into an estimate of the economic effects of higher education on the Texas economy requires an estimate of the annual contribution of Texas’ higher education system to the educational base of the state’s employed labor force.
The top portion of Table 4.1 presents the enrollment figures in Texas’ public higher educational institutions from 1996 to 1998. On average, the Texas higher education system generated 599,000 person-years of education each year. Yet, not all of this added education could be counted as staying in Texas. To adjust for migration of students, it is assumed that all out-of-state students would return to their place of origin and all in-state students would remain in Texas. Accordingly, it is estimated that the state’s higher education system annually pumps 549,000 person-years of education into the Texas population over age 18. Since Texas’ relatively strong economy allows it to retain at least some of the out-of-state students currently studying in the state, if anything, this assumption may be overly conservative.
Because Black and Lynch’s productivity-response relationships measure the effects of increasing education on the employed workforce, only that amount of the additional educational improvement provided to the employed workforce should be considered to affect productivity. According to national estimates from the Current Population Survey, approximately 85 percent of the civilian population with at least some college education is employed so the Texas higher education systems injects approximately 466,500 person-years of additional education into the employed workforce each year.
The middle section of Table 4.1 presents the Census Bureau’s estimated Texas resident population by age group for 1998 and the average number of school years attained by these groups as estimated from the Texas portion of the national Current Population Survey. Combining these two data items indicates that the Texas population 18 years and older represents a combined 170 million person-years of education. However, not all are employed. Adjusting for these considerations indicates that the educational base of the employed workforce in Texas is 122 million person-years of education. As such, the 466,500 person-years of education produced by the state’s higher education system results in increasing the educational base of the employed workforce by slightly less than 0.4 percent annually.
Applying this percentage growth to the adjusted productivity-response function based on Black and Lynch’s work indicates that the Texas higher education system increases productivity by 0.19 percent in manufacturing and by 0.22 percent in non-manufacturing each year. This increase in productivity applied to the Texas gross state product generated in manufacturing and non-manufacturing indicates that higher education adds $1.4 billion to the state’s economy through productivity gains or $3,100 per full-time equivalent student (bottom of Table 4.1).
This gain, by itself, does not include two important considerations. First, what is the public and private cost to the economy of procuring this gain? And second is the recognition that this gain, like other capital investments, does not represent a one-time phenomena, but generates a lifetime of returns.
Estimated Impact of Higher Education on Texas Productivity
Higher Education Enrollment (FTE) Fiscal Years 1996 1997 1998 3-Yr Avg In-State (%) Community & Technical Colleges 236,619 241,157 246,743 241,506 94.7% Public Universities 342,835 340,759 342,467 342,020 89.6% Health-Related Institutions 14,642 15,378 15,333 15,118 90.5% Total Public Higher Education 594,096 597,294 604,543 598,644 548,838 Total Percent Employed 85% Total Number 466,512
Population, School Years and Workforce, 1998 (thousands) Resident School Years Workforce Age Population Per Person Total Percent School Yrs 18 to 24 Years 2,038 11.8 24,048 74.9% 18,012 25 to 44 Years 6,035 12.5 75,438 70.2% 52,957 45 to 64 Years 3,989 12.4 49,464 71.6% 35,416 65 Yrs and Older 1,994 10.6 21,136 72.3% 15,282 Total Population 18 Yrs and Older 14,056 12.1 170,086 71.5% 121,667
Black and Lynch Texas GSP Gains
School Year Gain/Year (thous) Employed Higher Education Students 467 Employed Population 121,667 Schooling Gain per Year 0.38%
Gross State Product, 1988
Productivity Gain/10% Schooling Low High GSP (Millions) Percent Manufacturing 4.9% 8.5% $106,460 16.2% Nonmanufacturing 5.9% 12.7% $551,130 83.8% Total Economy 5.7% 12.0% $657,590 100.0% Estimated GSP Gains (Millions) Low High Manufacturing $200 $347 Nonmanufacturing $1,247 $2,684 Total Economy $1,447 $3,031 Per FTE ($) $3,101 $6,497
Sources: Carole Keeton Rylander, Texas Comptroller of Public Accounts, Texas Higher Education Coordinating Board and US Bureau of the Census, Current Population Survey.
To reflect the first consideration, Table 4.2 notes the two main components of the cost of education. First, are the earnings lost while in school, which from 1990 to 1997 averaged $21,412 per year. Table 4.2 also notes the average annual college costs from the 1999-2000 of $11,871. These costs include both those paid by the student and by state and local governments, based on direct educational all-fund appropriations in the 1998-99 General Appropriation Act and the county community college levy.
Texas Discounted Productivity Gains
Real 1998 $
Lost Wages of High School Graduate or Equivalent Year Males Females Average 1990 $26,906 $15,187 $21,047 1991 $26,316 $15,385 $20,851 1992 $26,088 $15,092 $20,590 1993 $26,927 $15,434 $21,181 1994 $27,014 $16,061 $21,538 1995 $27,336 $16,091 $21,714 1996 $27,426 $16,478 $21,952 1997 $28,166 $16,683 $22,425 Average $27,022 $15,801 $21,412
Average College Costs (1) Year Tuition
Transportation Personal Total Expenses 1999-2000 $8,771 $715 $1,547 $377 $461 $11,871
Internal Rate of Return and Present Values Year Cost/Return 0 -$33,283 1 $3,101 2 $3,194 3 $3,290 4 $3,389 5 $3,491 40 $9,822 41 $10,117 42 $10,420 43 $10,733 44 $11,055 45 $11,386 Rate of Return 12.11% NPV (6%) $39,334
Total Economic Gain $/FTE Students Gain (Mil of $) Present Value $39,334 466,512 $18,350
(1) Room and board, transportation, and personal expenses included for only non-commuting bachelorŐs-degree students.
(2) State and local higher education all-funds appropriation, including community college levy, per full-time equivalent student.
Source: Carole Keeton Rylander, Texas Comptroller of Public Accounts.
Together these costs of almost $33,300 in the first year more than offset the net productivity gain of $3,100 per student. But, since the productivity gains of increased education continue to produce economic gains throughout the student’s 45-year working lifetime, these productivity gains produce a stream of income gains throughout the years that rises with their real value to the economy (assumed to be three percent annually). Discounting this stream of income gains over time at a rate of six percent indicates a net present value of approximately $39,300 per student for a year of education. Considering the potential Texas workforce of 466,500 working students taught by the Texas higher education institutions, this amounts to a total net gain to the Texas economy of $18.4 billion.
This attempt to measure the impact of Texas’ higher education system on the state’s economy has considered the effects of higher education on the productivity of the Texas workforce. Increased productivity effectively increases the supply of labor available to the economy, which allows the economy to expand. This effect relies on the ability of the educational system to augment the supply of human capital available to the economy.
But this analysis largely ignores the knowledge capital function that Texas’ higher education system also performs. To the degree that the effects of increasing the stock of knowledge capital in the state is not included, this analysis of the economic effects of higher education on the state’s economy underestimates the true contribution.
Clearly the knowledge function of Texas’ higher educational system—basic and applied research—is important to the economy. There are numerous examples of discoveries, patents and technology transfers through Texas universities and colleges that have helped companies grow in the state and helped start new firms or even new industries. In a way, the efforts to account for research dollars flowing into the state captures part of the knowledge function. Also, to the extent that this knowledge function is truly a joint product with the human capital function, measuring one probably captures most of the effects of both. Nonetheless, to the degree that the knowledge effects on the state’s economy are missed by both of these two measurement efforts, this analysis underestimates the true impact of Texas’ higher educational system on the state’s economy.