Belyanin on capital in the 21st century

https://notes-from-the-golden-horde.blogspot.com/2019/03/a-review-of-capital-in-twenty-first.html

Can privatizing universities stunt economic growth?

In transition economies, governments have sold off universities to raise money and reduce the public footprint―notably in Kazakhstan, the largest of the post-Soviet economies in Central Asia. In that country, the educational share of government spending has usually been lower than the world average since 2004, according to World Bank data. In general, poor countries may emphasize education more, relative to other products of the government, than rich countries do. Education’s share of government spending is larger in highly-indebted poor countries than in the world on average, which in turn has a larger share than do high-income countries.  These trends raise the possibility that governments view education as a tool of limited value in developing the economy.  Advocates of privatization may believe that especially in higher-income countries, turning over colleges to the market will enhance efficiency.

In reality, privatizing universities may be lucrative in the short run but costly in the long. This post explores why selling colleges to the private sector might stunt economic growth.  The paper will look at a case study of this argument―how the United States government in the last half of the 19^th century stimulated growth that continues today by publicly financing universities.

A morsel of theory

Advocates of privatization argue that the market can run universities better than the government can. In reality, colleges are more complex than privatizers concede.

To see why, consider two services that a university provides:

Practical education, in such fields as accounting and marketing, which enables the graduate to earn more money.  This is a private service that earns a rate of return on the labor market, since it enables the graduate to earn more.  An owner of the university can claim part of this return for herself, so she has reason to increase the net value of practical education. The market may be able to provide practical education efficiently, in the sense that it produces all units of such education that has value net of costs.

Research and social education.  By “social,” I mean that a university prepares the graduate to be a better citizen through courses in such areas as political science, history, and the liberal arts. An owner of a university who maximizes net revenue may not provide these classes because they do not increase the graduate’s wages and thus the owner’s own rate of return. Similarly, basic research has social value but does not directly create profits. Breakthroughs in the theory of molecular biology may not command a price on the medical market, although they may lead to treatments of cancer someday.   

In general, research and social education create “public goods.” The layman defines a public good as anything produced by the government; but in the view of economists, pure public goods are products with two characteristics that prevent markets from furnishing as much of them as we would like: 

Nonexcludability. By definition, no one can be physically constrained from consuming public goods, whether or not she pays for them. Because nonbuyers cannot be prevented from enjoying public goods, no market producer will provide enough of them, since she cannot charge all beneficiaries a price. For example, no one can be prevented from breathing, so everyone will benefit from clean air whether she pays for it or not. Therefore she won’t pay―and so no profit maximizer will volunteer to purge the air of particulates.

Another example: In colleges, training in political science creates better citizens who help others by making more informed choices at the polls. Because the college cannot collect payments from all citizens who benefit from informed voting, it may not offer enough education and research in the field if it is trying to maximize net revenues.

Nonrivalry.  One person’s consumption of a public good will not reduce the amount available to someone else.  If Smith studies Jones’s history of the English language, Smith will not diminish the degree to which someone else can understand that history by studying the book. In contrast, an apple is rivalrous: If I consume it, you can’t.

A nonrivalrous good is not scarce, so it cannot earn a price based on scarcity.  The market won’t supply it, because a producer cannot profit from a zero price.

In short, if a profit maximizer buys a university, he may produce little of such services as basic research and social education, because he cannot profit directly from them.  Ironically, eliminating these services will diminish a university’s reputation and thus reduce the market rate of return to its practical education. In principle, a profit maximizer may furnish research to maintain the university’s reputation and thus its high rate of return; but such a producer must be unusually far-sighted. 

Education also suffers from a more mundane constraint in markets: The impossibility of collateral, as Friedman pointed out.  If Smith wants a loan for purchasing a home, he can put up the house itself as collateral.  But he cannot do this if he wants a loan for education, since providing human capital as collateral amounts to slavery.  For that reason, poor households cannot get many educational loans without guarantees from third-party financiers such as the government.

These barricades to education may thwart economic growth, since the amount of education acquired accounts for a fifth of the differences in labor productivity across countries, as estimated by Hall and Jones. One might think that the gain in productivity will fully show up in wages and thus induce the potential worker to pay high tuition while in college. But even productivity increases that are not due to public goods may not be completely reflected in wages. In a competitive labor market, wages will not fully capture the average amount produced by a worker, because they are determined by the least valuable worker hired. Consider a bicycle shop with two mechanics: One repairs five bikes a day; the other, three bikes. Customers pay $50 per repair. The value of the first mechanic is $250 per day; of the second, $150. If the shop pays the same wage to both mechanics, its wage will not exceed $150, since otherwise it will lose money on the second mechanic. But a $150 wage understates the value of the first mechanic.

To illustrate the economic value of a college education, consider a case study of the United States.

 Log-cabin colleges

American universities supply public goods partly because voters think that the schools should advance social welfare―an attitude that strengthened over the century following the American Revolution.  Land subsidies to American education date back to the early American republic. The states created public universities after the Revolutionary War ended in 1783; schools that had received land from government included several that are still successes, such as Harvard, William and Mary, Yale, Michigan and Dartmouth, noted Tolar. 

As proposed by Thomas Jefferson, the Northwest Ordinances of 1785 and 1787 set aside a section in each township to support public education. “Religion, morality, and knowledge,” said the 1787 ordinance, “being necessary to good government and the happiness of mankind, schools and the means of education shall forever be encouraged.” Congress also encouraged new states to provide public education.  “On the value of diffusing knowledge through public schools there was substantial agreement,” wrote the historians Tyack, James and Benavot, according to Kaestle.  By the 1860s, free education was as mandatory in the Old Northwest as it had long been in New England. However, before 1862, most Americans had thought education “the responsibility of parents, churches, and local town meetings,” wrote Kaestle.  This view would have circumscribed the public goods produced by universities, particularly because only the rich could afford college since students had to pay for their own lodgings, noted Cook and Ehrlich.

Northwest leaders embraced the new philosophy. “An early attention to the instruction of youth is of the greatest importance to a new settlement,” said Manhasset Cutler, who founded a major player in the Northwest, the Ohio Company. “It will lay the foundation for a well-regulated society. It is the only way to make subjects conform to its laws and regulation from principles of reason and custom rather than the fear of punishment.” (The quote is from Kaestle.)

Growing pains 

Though well-intentioned, the Northwest Ordinances did less to advance education than did a later law. Perhaps the Ordinances were hampered by the primitive conditions of the frontier.  In contrast, in its own words, the Morrill Act of 1862 sought “to promote the liberal and practical education of the industrial classes in the several pursuits and professions in life.” 

This motivation was decades old. “The agricultural origins of the Morrill Act go back as far as the American Revolution,” wrote Duemer.  The Act’s contribution was to create colleges through a national system.  It led to land-grant universities in each state to teach agriculture, engineering, and the arts of war. Before the Act, colleges had taught only a limited menu of fields.  In 1819, the surveyor general of New York, Simeon DeWitt, said colleges prepared students for "only three professions, law, physics, and divinity, the only professions recognized as ‘learned.’" (The quote is from Peterson).  This was in the European tradition.

Public acceptance of public universities varied from region to region.  Abraham Lincoln signed the Morrill Act in 1862, during the Civil War, when Southern congressmen weren’t around to object that the Constitution didn’t say that the national government could provide education. After losing the Civil War, Southerners changed their minds, and the 1890 Morrill Act set up agricultural colleges in their region. 

In any region, governmental wealth stimulated the creation of public universities. In the 19^th century, the Louisiana Purchase and smaller acquisitions left the federal government richer in land than in funds. So Congress often pursued economic development through land grants.  The 1862 Morrill Act granted 30,000 acres to the state for each of its Congressmen. Thus a state with two senators and three representatives received 150,000 acres from the federal government. Roughly speaking, the Act channeled more funds to more populous states.

The first Morrill Act created 57 land-grant universities, mainly in rural areas to serve farming, Liu noted.  The pace of creating schools picked up after 1887, when the Hatch Act threw more federal money at them. The second Morrill Act added 18 land-grant universities in the South, many of them devoted to African-American students because of segregation.

The land-grant colleges still account for most departments of agricultural education in the United States.  The success of the colleges “has shown the value of federal leadership and the advantage of bulletins on nature study, school gardening, and having uniform laws and finances creating an entire higher education system,” wrote historians Herren and Hillison. The land-grant schools also grew over time and took on more disciplines.

To the land-grant colleges, the Hatch Act added agricultural experiment stations. The stations supported such tasks as renewing soil, picking crops to grow, and canning.  “For example,” wrote Tolar, “a small farmer in eastern Oregon may contact the local experiment station and ask for advice on anything from which type of seed is best for the soil on his farm to advice on how to rid his barn of mice. His wife may ask for directions on how to preserve certain vegetables or what to feed chickens.”

Other variants on the Morrill Acts included the Smith-Lever Act of 1914, which expanded education of farmers and homemakers, according to Whalen and Tolar. And the 1944 GI Bill subsidized higher education for 8 million military veterans, noted Cook and Ehrlich.

Land-grant colleges were so successful that Congress extended the concept to marine research (1966), urban research (1985), space research (1988), and research into sustainable energy (2003―the “sun grant” colleges).

Apples and curve correction

The Morrill Acts seemed designed to provide public goods.  The state could build the colleges on the land―or sell the land to fund the colleges. But it could not spend the money on buildings. This is consistent with the interpretation that the Morrill Acts intended to finance public goods that―unlike buildings―would not be supplied by markets.

The laws also banned the state from using the land grants to pay for managing the investments. The state would have to shoulder that cost, wrote Whalen. This too jibes with the idea that the Morrill Acts were meant primarily to furnish nonmarket goods.

Finally, the Morrill Act “also specified that the state had to invest the trust so as to yield at least 5 percent annually, make that income available for the purposes of the grant, and make up any deficiency,” noted Whalen. Thus the law did not regard the trust as a money-making proposition. Perhaps one of its purposes was to provide education and research that markets wouldn’t produce.  “Land-grant institutions were asked to deploy a liberal and liberating education for a broad swath of society, an education that had practical value.”

To say that the education was practical is not to say that the market would necessarily produce it. For example, the Morrill Acts expanded the supply of engineers in the US. Although the practical worth of engineering may seem obvious today, before the Morrill Acts most engineers were products of West Point, and they numbered only 300 in 1866. By 1911, under the Morrill Acts, 3,000 engineers were graduating each year in the US.

The public services produced by a university can strengthen markets.  For example, a land-grant school, Washington State University, discovered that the Japanese were not buying American apples because these were oblong rather than round.  So it developed perfectly round apples, said Tolar.  Yes, we can view such activities as private goods that the market can support: The export of round apples yields profits.  But the underlying knowledge of such innovations may be a public good if its dissemination cannot be contained; and giving such practical education to poor farmers makes more equal the distribution of income.  

A final example of public goods that a university can bestow upon us: Basic research for defense services.  We have already seen that the market will not produce enough of such a public good. And yet its value is evident.  In the United States, “the successful research underpinning the production of the atomic bomb convinced the public that fundamental research was worthwhile, even though its application might not be immediately apparent,” said Peterson. So perhaps the government should create it by financing research universities.

Disembodied know-how

All told, the number of universities per capita in the United States grew rapidly after passage of the first Morrill Act, igniting economic growth, noted Cook and Ehrlich.  In their econometrics, they devised a quasi-experiment in which the Morrill Act was a treatment of the United States, with the United Kingdom (and sometimes other Western countries) as the control group.  They find that after the Act passed, 78 more universities per 100,000 residents were created in the US than in the UK.  They controlled for factors that might also have affected the creation of universities, notably real GDP per capita.

Cook and Enrlich conclude that the Act “launched the public higher education movement in the US…[which] spearheaded a higher long-term rate of growth in per capita income in the US relative to the UK and other major European countries.” For much of the 19^th century, the US had been “relatively poor” (p. 1).  But it established a lead in educational attainment over Europe in the 20^th century, largely due to the growth of high schools from 1915 to 1940.  This triggered economic growth: “Higher educational attainment was perhaps a major, if not the major, instrument through which the US overtook Europe as the economic superpower in the 20th century.”  In 1998, of those aged 25-64, a larger share had gone to four-year colleges in the US (27%) than in any of the five largest economies of Europe.

Higher education produces technological advances through “creative knowledge, which flows from the minds of scholars, scientists, inventors, and entrepreneurs and increases their capacity to accumulate new knowledge.” Such disembodied knowledge may have increased the growth rate of per capita income in the US.  Over the period 1871–2012, the per-capita annual real GDP growth rate in the US versus the UK was 1.8% versus 1.4%.  “Our basic thesis,” wrote Cook and Ehrlich, “is that the differences in the long-term per-capita income growth stem primarily not from differences in physical stocks, including land or other natural resources, but from differences in the rates of growth of human capital.” This resulted partly from the institutional carrots in the US for accumulating knowledge, such as high rates of return to schooling. 

In general, Cook and Ehrlich thought their evidence “remarkably consistent with the view that human capital formation—especially through the channel of public higher education—was the ‘secret weapon’ through which the US was able to achieve its robust long-term rate of persistent, self-sustaining growth in productivity and per capita income” (emphasis in the original passage).

Some macroeconomists dispute that going to school for more years will spur economic growth more rapidly than other causes can. One view, analyzed by Spence, is that people may attend prestigious colleges not to learn but to signal a prospective employer that they deserve a high wage. In another view, developed by Hausmann and others, growth is more likely to arise from the increasing complexity of the economy, since this combines old and new knowledge in innovative ways.

Admittedly, “systematic econometric studies have yet to verify the hypothesis that investment in schooling serves as an engine of long-term growth,” wrote Cook and Ehrlich.  Nevertheless, Liu gives supporting evidence. Liu studied whether land-grant universities created in the 1860s generated spillovers for the local economy.  Like Cook and Ehrlich, he took an experimental approach.  Presence of a Morrill land-grant university was the treatment for a county; control counties had no such university. The control counties were synthetic: Liu constructed them as weighted averages of potential control counties, choosing the weights so that each synthetic county resembled a treated county in traits that preceded the land-grant university. This eliminated differences between treated and control counties that might have influenced the results systematically. By comparing the performance of the matched counties, one can observe the impact of the university as time passes.  For example, in Knox County, Tennessee, population density over time was much like that of the synthetic control until the late 1860s, when East Tennessee College came into being (the University of Tennessee today). Population density then grew more rapidly in Knox than in the control.

Do land-grant schools generate spillovers? Liu’s answer: Yes, eventually.  On average, population density around the colleges that had been built through the 1862 Act, rose by 6% over 10 years and 45% over 80 years. This created scale economies, cutting production costs.  Also, proximity to the colleges boosted productivity in manufacturing by more than half in the long run (80 years) though not in the short.  These results are consistent with the view, common among regional economists, that the success of the two most famous high-tech clusters in the US, Silicon Valley and Route 128, owes much to their nearness to Stanford University and the Massachusetts Institute of Technology.  “Investment in higher education increases population density and enhances local productivity,” wrote Liu. “This partially justifies the continuous subsidy to post-secondary education.”

But though the Morrill schools targeted the practical arts, their impact on the share of manufacturing labor in the population was statistically insignificant. “This yields potential implications for policy makers who seek to develop an industrial city by investing in higher education.”  

Public funding of universities can grow the economy over the long run if it emphasizes products that markets would under-provide, such as social education, basic research, and little-known but promising industries. Alternatively, one might ensure such efficiency for a privatized university by turning over its shares to a nonprofit entity. --Leon Taylor tayloralmaty@gmail.com

References

Cook, A., & Ehrlich, I. 2018. Was higher education a major channel through which the US became an economic superpower in the 20^th century?  (Asian Development Bank Institute Working Paper No. 820).  Retrieved from the Asian Development Bank website: https://www.adb.org   

Duemer, L. S. 2007.   The agricultural education origins of the Morrill Land Grant Act of 1862.  American Educational History Journal, 34, 135-146

Friedman, M. 1962. Capitalism and freedom. Chicago: University of Chicago Press.

Hausmann, R., Hidalgo, C. A., Bustos, S., Coscia, M., Chung, S., Jimenez, J., Simoes, A., & Yildirim, M. 2018. The atlas of economic complexity: mapping path to prosperity. Center for International Development, Harvard University; Harvard Kennedy School; Macro Connections, MIT Media Lab; Massachusetts Institute of Technology.  Retrieved from   https://atlas.media.mit.edu

Herren, R. V., & Hillison, J.  1996.  Agricultural education and the 1862 land-grant institutions: the rest of the story.  Journal of Agricultural Education, 37, 26-32. doi: 10.5032/jae.1996.03026

Hall, R., & Jones, C. 1999. Why do some countries produce so much more output per worker than others? Quarterly Journal of Economics, 114, 83–116.

Kaestle, C. F. 1988. Public education in the Old Northwest: “necessary to good government and the happiness of mankind.” Indiana Magazine of History, 84, 60-74.

Liu, S.  2014. Spillovers from universities: evidence from the land-grant program.  Unpublished manuscript, University of Southern California, Los Angeles.

Peterson, J. B.  1977. Origins of the land-grant philosophy and its influence on agronomic education.  Journal of Agronomic Education, 6, 25-29.

Spence, A. M. 1973. Job market signaling. Quarterly Journal of Economics, 87, 355-374.

Tolar, R. L.  2008.  The “land-grant model” in U.S. higher education.  Journal of Siberian Federal University. Humanities & Social Sciences 3, 408-412.

Tyack, D., James, T., & Benavot, A. 1987. Law and the shaping of public education, 1785-1954.  Madison: University of Wisconsin Press.

Whalen, M. L. 2002. A land-grant university.  Retrieved from https://dpb.cornell.edu  (Republished from the Cornell University 2001-02 financial plan, pp. 1-13, Ithaca, New York: Cornell University)

World Bank.  2018.  World Development Indicators.  Retrieved from http://www.worldbank.org