Economist Writing Every Day

Are resources becoming scarcer as world population increases and per capita consumption increases? Are basic goods becoming more expensive relative to wages in the face of potential resource shortages? These are some of the main questions that are addressed in the just released book Superabundance by Marian Tupy and Gale Pooley. The authors were kind enough to provide me with an advance copy, which is why I’m already able to review this book on its release date (I’m not really that fast of a reader).

The author take a very optimistic view of the issues surrounding those opening questions. Properly measured (one of the key tasks of their work), resources are becoming more abundant, not more scarce. And properly measured, almost all consumer goods are becoming cheaper relative to wages.

The authors use the approach of “time prices” throughout the book. They are not the first to use this approach. Julian Simon (their inspiration for this project) used it in various places in his work. William Nordhaus famously used it is in paper on the history of the price of lighting. And Michael Cox and Richard Alm have used the time-price approach in many of their writings, from the 1997 Dallas Fed annual report, to a full-length book a few years later, as well as updates to the original 1997 report. And if you follow me on Twitter, I like to use this approach too.

In short, “time prices” tell us how many hours of work it takes to purchase a given good or service at different points in time. How many hours would you have to work to buy a pound of ground beef? A square foot of housing? An hour of college tuition? It’s the superior method when you are looking at the price of a particular good or service over time, compared with a naïve inflation adjustment, which only tells you if the price of that good/service rose faster or slower than goods or services in general, not if it’s become more affordable. Inflation adjustments are really only useful when you are trying to compare income or wages to all prices, to see if and how much incomes have increased over time. Of course, which wage series you choose is important (and you need to have a consistent series over time, or at least the end points), but as the authors point out (which they learned from me!), if you looking at wages after 1973, the wage series you use doesn’t matter much. Median wages, average wages, wages of the “unskilled” — these all give you the same trend since 1973. We don’t have all of these back earlier (especially median wages), but there’s not much reason to believe they’ve diverged that much. And the authors also present their data using multiple wage series in many of the charts and tables.

What do the authors find?

Whether you start in 1980, 1960, 1900, or 1850, commodities have gotten “cheaper” in time-price terms, and thus more abundant. Or if you look at the price of various foods in the US starting in 1919. Or if you look at a broader set of consumer goods (appliances, clothing, etc.) starting in 1979. Whether you look at the US only, or at groups of country spanning the income range from India and China up to the US and Norway, you see the same trends. You can find exceptions here are there. From 1960 to 2018, gold and crude oil got slightly less abundant compared with US wages, but that’s not true if you start in 1850, 1900, or 1980. But the broad picture is: as measured by time prices, resources are getting more abundant. If we were really running out of a resource in an important physical way, the price would rise dramatically. We just don’t see that in the data over long periods of time. And whether we start in 1919 or 1979, the goods that consumers by are almost all getting significantly cheaper in terms of “time prices.” There is no great stagnation!

That much is true for goods. And I believe this book really puts to rest the idea that we are running out of resources, either natural resources or agricultural land to grow food, in any important way. Julian Simon spent a good part of his career trying to show this in various ways, but Tupy and Pooley have really shown us all the data, from multiple directions, in a way that is clear to understand. Resources aren’t only becoming more abundant, they are becoming superabundant. To just take a few examples, since 1960 the global “time price” o rubber, tea, tobacco, palm oil, coffee, and cotton all fell by 90% or more! (They use global GDP per hour worked as the “wage” in this example). And on average commodities fell in “time price” by 83%, meaning that the typical worker could double their consumption of commodities every 22 years. World population has grown from 3 billion to almost 8 billion since 1960, yet resources became massively cheaper.

One caveat: most of their data stops in 2019 (the hazards of writing a big book that takes a lot of time to produce), and commodity and food prices have risen a lot since 2019 to today. But these rises are for very specific reasons, not an overall crisis of abundance. No doubt though, the voices of resource exhaustion are already started to raise their voices again right now. Tupy and Pooley would say: focus on the long-term trends.

But, Services and Housing…

But… that’s only for goods. What about services? What about a very specific kind of good: housing? Services constitution about 2/3 of all consumer spending in the US today. And that’s not a recent change: they have been over half of consumer spending since 1970. And depending on how you count it, housing expenditures are somewhere between 35 and 40% of consumer spending. We can’t ignore these! I take this issue so seriously on Twitter, I’ve even been turned into a meme.

This is the part of the review where I present some criticisms, or really some suggestions for future work and research. Because the book, at about 400 pages, spends very little on these topics. That makes sense, which is why this isn’t a criticism per se, since the authors are mostly concerned with resource abundance, the question that drove Julian Simon, not living standards directly (though they certainly talk about it a lot). To know whether the average American has seen their living standards improve since the 1970s, we need to look not just at ground beef and TVs, but also health care and housing. Inflation adjustments come in handy here, since that’s exactly what they are designed to capture. But on these items, as well as other services such as college tuition, it’s well known that the standard inflation adjustments are imperfect.

In the book, there is some time spent on health care, though they only look at elective cosmetic procedures. Why? They contend that medical and hospital bills “would be lower if health care in the United States were subjected to some of the competitive forces that operate in the market for cosmetic procedures.” And perhaps they are right, but… we can’t really know for sure. And elective medical services probably have very different demand curves than, say, the demand curve for open heart surgery or insulin, so I don’t think they are directly comparable. Still, there exercise is interesting for cosmetic procedures, as these have all fallen in “time price” since 1998. I expect you could do a similar analysis for prescription drugs once they are off patent.

There is a short section (Box 9.1) on housing, where they do some good adjustments for square footage, etc. But the topic of housing deserves more than 1 page. Heck, it might even deserve a full book! And there is also a clear regulatory story to be told with housing, since the supply is restricted legally in some many ways. And yet, by many measures housing has become more affordable relative to wages in certain cities and states. But not everywhere. I would love to see the kind of analysis they did for 50 different commodities over time done for 50 different states or 500 cities on housing prices, along with some quality adjustments for square foot, etc., and current interest rates.

Education is discussed as an outcome of the Great Enrichment in Chapter 8, though there is not discussion of the cost of higher education or anything else. This is also an area that is tough to measure: thousands of college and universities, little transparency on actual prices paid, many third-party payments and other government subsidies. It’s not as simple as looking up the price of tin or crude oil in two different time periods! But it’s an important question, one that is driving a lot of public policy right now, and I think one where there actually is a case for optimism, if looked at closely and properly measured.

The authors the issue of measuring services, such as healthcare, and suggest this is an area for future research. I fully agree! And if you read this blog or follow me or Twitter, you will often run across my attempts to make time-price comparisons for services and housing. Perhaps someday we’ll have enough material for a book. In the meantime, read this book by Tupy and Pooley.