The 'Time Price' of Education
Hoosiers are spending less than ever on education, and it shows . . .
In 1967, economist William Baumol published a paper entitled “Macroeconomics of Unbalanced Growth” in which he laid out a model of an economy dominated by two sectors. One sector had rapid productivity growth, while the other had slow productivity growth. However, as intra-national data make clear, overall wages are set by the high productivity sector. That’s why highly productive manufacturing workers in Mexico can be hired for $5 an hour, while low productivity factory workers in the USA still cost over $20 per hour.
This paper has been heralded as the exposition of Baumol’s Cost Disease. Here’s how I explain it.
I have a delightful picture of my grandfather in about 1898. He’s in an elementary school class picture with a couple of dozen classmates. My dad attended the old Perry County Courthouse school in Rome, Indiana. In it are classrooms with a couple dozen desk chairs, of the old fashioned type (I found my dad’s initials carved into one of them). I attended Somerset Elementary in Bowie, Maryland, in a class with about two dozen classmates. This is a contemporary photograph (not my class).
My kids all attended elementary school with roughly two dozen kids in each class. Each class had a teacher, whose salary grew over the roughly 110 years between my grandfather’s third grade experience and that of my kids. Yet, if the output is measured as the number of kids who move through third grade, there was no measurable productivity growth.
In other words, when it comes to simply measuring output in terms of physical products (kids), there’s been no change for more than a century. Of course, that’s not really true, 3rd grade is more complex than in 1898. Still, when it comes to paying for teachers we have to match the national labor market for their skills. So, in dollar terms, our spending on education is growing substantially.
In contrast, we have enormous productivity gains in industrial production. This graphic shows the inflation adjusted dollar value of industrial production in the United States and employment from 1946 to 2021. That is the first post-WW2 year through the most recent full year of data.
Over that time, the value of goods produced in the USA have risen by more than 750% in inflation adjusted terms. The number of people working in factories has declined by close to 10 percent.
So, manufacturing is the fast growing sector, while education is the slow growing sector. The resulting prediction from Baumol’ s work is the “cost disease” in which the cost of these low productivity sectors (education, healthcare, etc.) will grow, while we’ll have cheaper and cheaper goods and services elsewhere.
So, we sit in 2022 continuing to hear complaints about the high cost of buying services in these low productivity sectors (like higher education, K-12, and healthcare generally. [Note: Indiana has particular monopoly problems, which I discuss here].
But, does the ‘unbalanced growth’ that Baumol wrote about really mean that the low productivity sectors are getting more expensive? Sure, in dollars they are, but how would we measure inflation if it affects industries so differently, causes deep consumer shifts in spending and most importantly boosts all our wages?
The reason this is a tough issue is not that we cannot create price indices to measure inflation, but rather that all our wages are affected by productivity growth throughout the economy. So, teachers, professors, receptionists and ditch diggers all receive benefits in terms of wages from productivity growth in sectors outside our own. At the same time, goods get cheaper, so we get a ‘real’ boost in earnings from cheaper goods.
These sorts of questions are especially hard to answer when you jump technologies from a rotary phone to an iPhone, or when government provides some of the services. At least part of the answer was offered by William Nordhaus in his paper “Do real-output and real-wage measures capture reality?” It’s subtitle, was “The History of Lighting Suggests Not.”
In this paper, Nordhaus decided to evaluate price indices by examining the cost of an important measure of income; the availability of light. We use light to work in the dark, illuminate our workspaces and the road ahead of us. For most of human existence, light was very, very expensive.
The great whaling fleets of the USA were in search of the Sperm Whale, whose oil illuminated parlors and studies across affluent parts of America. And, for most of recorded human history, the night began when the sun set, and day at the rising of the sun. Only the very rich could afford lighting to extend activities past the setting of the sun.
In order to figure out how to measure the cost of these different technologies over time, Nordhaus’s simply converted the cost of light into the number of hours the average worker would have to toil to obtain a single measure of light.
This innovation dispensed with the technology shifts and other quality measures. In the following figure, he shows two measures, a traditional price index and the labor adjusted price index.
This graphic offers a truly innovative method of dealing with price changes over time, albeit applicable to a small suite of products. It suggests that the price indices aren’t really giving us a clear picture of the cost of things these days. They understate the abundance of our modern economy substantially, and overstate how much we are spending for things like lightbulbs. More importantly, it means the price indices and spending on many items may be overstating what we spend on many, if not all the products we consume.
Earlier this year, Marian Tupy and Gale Pooley published a book entitled Superabundance, subtitled the story of population growth, innovation and human flourishing on an infinitely bountiful planet. This is a superb, superb book that extends Nordhaus’s “Time Prices” to hundreds of commodities. Everyone interested in public policy should read this book. I cannot possibly recommend it enough.
The big thesis of this book is that growth comes almost solely from human capital — knowledge, innovation, creativity and market exchange. And, unlike doomsayers the world over, the fact is that on average, human beings produce more abundance than they consume. It is a ruthlessly empirical rejection of the limits to growth. For that reason alone, it should grace every home library in the United States. But, that’s not why I draw upon that work.
The “Time Price” of goods is a uniquely clear way of conveying the cost of goods or services against wages. What if we considered the “Time Price” of education in the same way we consider the “Time Price” of other goods and services?
To do this, we’d need only to know a few key bits of data. One is the amount Hoosiers spend on education (college and K-12), which I obtain from the National Association of State Business Officers, who compile this from states. I’d need to know state GDP and state employment, which I obtain from the Bureau of Economic Analysis at the Department of Commerce. Then, I’d have to obtain an estimate of the number of hours worked each year, which come from the U.S. as a whole. I divide this by the number of workers in the U.S economy to obtain a rough number of hours worked each year. I assume Indiana is similar to the nation as a whole.
From this, I can measure the hours worked by Hoosiers to provide higher education, and K-12 education. I can then divide this figure by the number of workers to understand what the average "(mean") worker spends each year to fund Kindergarten through 12th grade in the state. Because this is a mean, it overstates what the middle or ‘median’ taxpayer pays for education. This is profoundly illuminating.
I start with K-12 education, since this is the biggest of the state budget items. I limit this sample to the period 2010 through the present (FY 2021), since the state law passed in 2008 (HEA1001) took over all operational spending of local schools. This is state spending only, and doesn’t include Federal or local spending (property taxes). The property tax share is roughly 40% of this, and the Federal share would be closer to 10 percent of this (Indiana is a net recipient of Federal Funds).
From 2010 to today, the amount of hours worked by the average Hoosier to fund K-12 education dropped by 11.2 percent, from 43.8 hours per year to 39.3 hours per year. Importantly, enrollment grew during this time period, so there are more students in the K-12 classroom (public and private) than in 2010.
The story for higher education was even larger. From 1991 to 2021, the time spent by the average taxpayer supporting public higher education in Indiana declined by more than 80 percent, from 15.4 hours per week to 8.5 hours per week. This is only state dollars, and doesn’t include Federal spending on higher education, or private spending through tuition.
These offer a compelling antidote to the idea of cost increases in these activities. If anything, Hoosiers are spending substantially less than they used to for education across the board. To place this in context, we can compare the American Time Use Survey administered by the Bureau of Labor Statistics for comparison.
Fortunately, we can also compare the cost of attending Higher Education in Indiana, through tuition in the state’s schools. Of course, as most know, tuition is the sticker price, and many students, especially those with very high academic achievement, and those with very low incomes, do not pay full price. But, if you did, the graphic below shows the time cost of paying for a year of tuition at Indiana’s large public 4-year universities.
Since the 2011-202 academic year, the number of hours it takes to pay for tuition at Indiana’s large public universities has declined an average of 8.4 hours. The biggest declines were for Purdue, with a 15.3 hour drop. The smallest decline was at Indiana State (which still has the lowest tuition), but which declined by 5.2 percent.
Indiana’s most expensive public university tuition and fees in the last year was at Indiana University Bloomington, which required 188 hours of work for the average employee in the state to afford. Importantly, that buys roughly 420 hours of classroom instruction for an undergraduate taking 15 hours per semester.
What are the major takeaways of this way for accounting for cost of education and other products. I think there are two.
The Superabundance book lays out a masterful argument that economic growth is caused by human innovation and ingenuity. I believe it is axiomatic that most of that innovation will come from highly educated people — particularly in free countries. The ability to innovate and sell ideas is only really nurtured in free places. There are also pretty strong local spillovers of much of this innovation. Particularly as it effects labor markets. As I’ve argued elsewhere, human capital is the driver of local growth. The rest is smoke and mirrors.
Second, Indiana is at deep risk of failing to rise to the challenge of educating our population. As Indiana’s Commission on Higher Education lays out in this report, our college going rate is in the midst of an epic collapse. We are already well behind in the most critical elements of economic growth, and the current evidence suggests we are entering a period of declining educational attainment. This graphic from the ICHE report makes clear the depth of our challenge.
Not all of these declines are caused by inadequate funding. COVID, tight labor markets for unskilled workers and frankly incorrect claims about the benefits of higher education all play a role in this problem. But, funding is almost surely the prime cause of this. As evidence, I point out that enrollment by Hoosier kids at private schools and out of state schools remains fairly stable. Tuition for private school students and out of state students are higher than in-state tuition. The huge declines come in those funded by Hoosier taxpayers.
Quite simply, the problem is funding. The downward pressure on tuition and the declining state support for higher education restricts he available resources for student support through scholarship and aid. This is why poor students, particularly African-American students, attend college at much lower rates than white students.
Why does any of this matter? Can’t a kid make a good living without going to college? Well, it matters because the single factor that matters more to regional prosperity than anything else, is the share of adults who’ve completed college. Tax rates barely matter, capital investment levels barely matter, a nice climate barely matters, mountains barely matter. It is education.
This graph comes from my recent study on Indiana’s Poor Economic Recovery, 2010-2019 in which I lay out the case for deep reconsideration of state policy.
This figure compares incomes and bachelor’s degree share. This is among the strongest, and most repeated relationships in social science research over the past century. Indiana is way behind the average state, and moving quickly towards the bottom. Oh, and yes a kid make make a good life in the trades, just make sure you take that skill to a community with lots of college graduates. That’s where the largest wage premium is located.
If Hoosier leaders wish Indiana to be part of the Superabundance of the future, they must more fully fund both K-12 education and higher education. Hoosiers have never given less of their time to supporting higher education and K-12 education than they do today. That is a policy mistake that has already caused significant declines in college attendance. Unless it is corrected, we should anticipate decades of poor economic performance as a consequence.
Note: This post was edited on October 27th, to correct a typo.