Conversable Economist

Jeff Horwich serves as interlocutor in “Interview with Ariel Pakes: To understand the broader economy, unlock the inner workings of product markets” (Federal Reserve Bank of Minneapolis, September 25, 2026). Pakes is an industrial organization economist, so he is congenitally suspicious of macroeconomics aggregates and broad statements about “how markets work.” Instead, he wants to examine actual markets.

One of his most prominent papers, written with Steven Berry and James Levinsohn, was called “Automobile Prices in Market Equilibrium,” and published in Econometrica in 1995 (63(4), 841–890). The problem the authors faced was that if you want to think about how demand for one kind of car will be affected by a change in price, you also need to think about how demand for all other cars will be affected: for example, are there lots of close substitutes for the car where the price is rising, or not very many? But with 200 models of cars, this problem quickly becomes unwieldy. Here’s Pakes:

As an IO [industrial organization] economist, I have to analyze a market. In the auto market, there were 200 models. If I was to do a demand system—let’s say, how many people want to buy an Oldsmobile Ciera—you’d need the prices of all the goods. Just for the demand for the Olds Ciera, you’d need 200 price coefficients at least, even if you just did it linearly. That’s what we call, in IO, “product space.” The only thing you know is the products and their prices. And that’s how demand systems used to be analyzed. If you did that, I’d have Olds Ciera quantity on one side, 200 prices on the other, and a similar equation for every one of these 200 products. That’s 40,000 coefficients. There’s no data that could effectively estimate 40,000 coefficients.

What we did in the demand systems—the BLP [Berry, Levinsohn, Pakes] group—is we said, If we know the characteristics of the products, and the distribution of preferences over those characteristics—like you have a preference for car size that depends on your family size and price that depends on your income—then all I really need is the distribution of people’s preferences, some of which I can get from age and things like that. If that distribution is, say, normal, and there are 10 characteristics, there’s something like 50 covariance terms. And from that, I can give you 40,000 cross-price elasticities. I just determine what everybody would do if a price changed. I need to sum over households. But with modern computers, that’s easy even for a hundred million people. And then I say, What would happen if I changed the price? I see what you would buy if the prices changed, or what everybody else would buy, and I find out the price elasticity.

Working in “characteristic space” like this has two advantages. One, it allows you to estimate demand in a market with many products. Two, it allows you to do the second question in IO, which is, What products do I want to develop? If all I have is past price and quantity, I can’t tell you what would happen if I put out a new product. But if I know the characteristics of a new product, then I get some idea of what the demand for a new product would be. That’s the incentive for the development of new products.

Pakes had a follow-up story about this paper, when General Motors heard about it, that I had not previously heard:

There was a guy named Mustafa Mohatarem in charge of research who had a Ph.D. in economics and was a very smart guy. He asked us to come to General Motors headquarters and present on the basic stuff. It didn’t have any micro data, just aggregate characteristics and quantities demanded. Afterwards he said to us, “I know exactly the markups on every product, up to the options that are on the product. You guys got it almost right. What would happen if I gave you real data?” So, he gave us their micro data. They had, at the time, an annual 64,000-person sample from new purchases, new registrations across the U.S. And then they had another 64,000, asking the people what they liked and what they didn’t like and what their second choice was. So, we got an old copy of that.

Why I brought up the Olds was because one of the questions he asked us is, What would happen if we killed the Olds department of GM? This is because they had Pontiac, they had Chevy, they had many midsized cars or family cars, and they thought they were just cannibalizing each other. They were not expanding. They killed Olds before we got back to them because it took us a while to figure out how to use the micro data. But after we did this, we got a phone call from them asking for the programs. We put the programs together in a way they could use them, sent them to GM, and they never talked to us again about this. They just wanted the programs. I hope they ended up using them.

This idea of thinking about the product market in terms of characteristics of products, not just the products themselves, turns out to have wide applications, and Pakes touches on a number of them in this survey. For example, when calculating inflation, there is a problem with looking at the prices of products where technology is evolving rapidly: for example, the specific laptop computer you bought last year may no be on the shelf, because this year’s model has different characteristics. When calculating price change, one has to make a “hedonic” adjustment for the change in characteristics. Pakes was instrumental in getting the US government to use hedonic measurement for certain areas in the Consumer Price Index measure of inflation.

Or think about the official poverty line, which is a certain dollar amount, adjusted for family size, that rises by the rate of inflation each year. But the characteristics of what a low-income family can actually purchase have changed over time, so that a poverty line based on characteristics of goods and services might look quite a bit different over a few decades from one based on adjusting a dollar amount.

Or think about antitrust. If two large companies want to merge, how should government regulators evaluate the threat to competition? Rather than that looking at meaures like share of total industry sales, perhaps the more relevant measure is about the characteristics of goods that would be produced by the merged firm and the characteristics of goods produced by other firms in the market. Indeed, this approach is now being used by many antitrust regulators.

At the end of the interview, Pakes offers this thought on the interplay of data and theory:

I think it’s the philosophy of science, that you have to put together data and theory to understand the world. Data by itself cannot prove anything. You need the theory that generated the data to have an answer. The theory is not perfect either, but you’ve got to start somewhere. The issue isn’t that we get the answers right. The issue is that we get the answers better than the next best person could get them. The world is too complicated to get them exactly.

Many of us are inconsistent in our economic desires. I would like the price of my home to rise, because it increases my wealth, but I would also like the prices of all other homes to fall, so that I could sell my house and buy an even nicer house. Also, I would like to have the value of my house rise, but also pay lower property taxes. David Schleicher digs into the second issue in “The Great American Property Tax Freak Out” (posted online at SSRN on September 1, 2026). He writes:

In the last three years, a number of states have substantially reformed their property tax systems, providing huge tax benefits to owner-occupied homes and shifting the burden of paying for services like schools and police to commercial property owners (including rental apartment buildings), to other local taxes, and to state funding, which itself largely comes from sources like sales taxes and income taxes. A few, including Florida, Ohio, North Dakota, and Texas, have considered going further, either completely abandoning property taxation for owner-occupied housing or even abandoning property taxation entirely. … One might think that homeowners would be happy that their largest asset has appreciated substantially in value, as homes did, particularly in suburbs, in the post-COVID period. But because property taxes are wealth taxes, property owners have responded to their increasing wealth with political anger, pushing shifts in property tax policy, often leading to increasing rates on commercial property owners who have seen declining values during the same period. …

[T]hese reforms shift the property tax from a tool homeowners use collectively to provide for locally-wanted services towards a more standard form of redistributive taxation, charging commercial property owners to pay for services for homeowners. Further, they will lead to greater state authority over local governments, less stable funding for local governments (but also less tax foreclosure in recessions), and stricter zoning controls (but more building of homes where it is legal). Perhaps most notably, these property tax reforms will lead to higher housing costs, and thus constitute a substantial transfer of wealth towards people who already own homes and have already seen their net worth increase substantially in the post COVID housing market.

A few other facts and insights seem worth appending here.

1) there’s a lot of local variation across public finance in the United States. But as a broad statement, property taxes are the major own-source of revenue for local US governments, and in turn the main source of financing for local schools. Many local governments have traditionally relied on property taxes to pay for police as well. The political dynamic of reducing property taxes often involves state-level actions that limit or block property taxes at the local level. It is not clear that voters for lower property taxes have drawn a connection in their mind to the likelihood of reduced local public services.

2) the property tax is a form of wealth tax, and it has the standard problems of a wealth tax: specifically, you can have more wealth–say, the value of your home, your business, or your retirement account went up–without having immediate income to pay a higher level of taxes. Thus, older voters who tend to have higher accumulated wealth, partly in the form of home equity, but also lower current income, are a politically powerful and sympathetic group in the push to reduce or eliminate property taxes.

3) If property taxes are reduced or abolished, the cost of owning a home falls. (For example, imagine that you could guarantee that the roof, exterior, driveway, plumbing and HVAC systems in your house would never need repair or replacement, which would also lead to a fall in the cost of owning a home.) With lower costs of owning a home, the market value of the home rises. Thus, the ironic outcome is that higher housing prices cause voters to want to limit or repeal property taxes, which would then lead to even higher housing prices. Moreover, the current homeowners who have already benefited from higher home values will benefit further, while homeownership will look even costlier for potential future homebuyers.

When economists think about how much utility is gained from additional economic resources the basic assumption is that an additional amount of income or wealth increases utility, but by a diminishing amount–because additional income or wealth has diminishing marginal returns. However, there’s also a long-standing tradition in economics of thinking about why this relationship might have an S-shape. Benoît de Courson, Willem E. Frankenhuis, and Daniel Nettle discuss this past tradition and offer new thoughts in “Explaining the paradoxical effects of poverty on risk taking: The Desperation Threshold Model” (Behavioral and Brain Sciences, published online February 23, 2026, not yet typeset).

Here’s a basic diagram, showing the relationship between resources (like income or wealth) and utility. Focus first on the upper-right portion. This standard shape shows that greater resources lead to additional utility, but the marginal gains to utility are diminishing, and so the curve flattens out. Imagine some one who is considering a risky choice that could result in a gain or loss of an equal amount of income. The marginal utility of the gain in income is smaller than the marginal utility of the loss, so the person will only make this risky choice if the probability of gain is high enough, certainly greater than 50:50

Now consider the lower-left portion of the curve. The underlying assumption here is that people need a basic level of resources to survive. For those who have resources substantially below that level, additional income doesn’t increase their utility, because they still don’t have the resources to survive. However, imagine someone who has resources slightly below the basic survival level, who is considering a risky choice that could gain or lose an equivalent amount. For that person, the utility loss from this risky choice is low, because the person is already below the basic survival level, and so being further below that level has a minimal loss in utility. However, if the risky choice can propel this person to a place above the basic survival level, then the marginal utility gains are large.

The S-shaped graph thus seeks to explain why people with low income levels may be more prone to risky decisions, like crime or gambling. The authors are careful to call this a “mid-level theory,” by which they mean that it suggests a pattern but then “raises many further detailed questions: about its explanatory scope; its cognitive and neural implementation; and its ability to predict and explain in different socioecological contexts and at different scales.” That is, the purpose of the theory is not to assert that all people act in a certain way, but to offer a basis for additional discussion and appropriately hedged real-world insights. One obvious question, for example, is how people in different times, places, ages, and economic settings may perceive the basic level of resources. Perhaps some middle-class people (or upper-income people, for that matter) who are well above the poverty line do not perceive themselves as having what the regard as “basic” resources. The model also suggests that people who are just above the “basic” level (however defined) might avoid risk and act cautiously, while those just below the “basic” level might be risk-takers.

De Courson, Frankenhuis, and Nettle point out that the idea has entered economic theory in a number of situations. For example, why do many middle-class people both buy insurance (risk-reducing) and gamble (risk-increasing?). The authors write:

Friedman and Savage (1948) proposed a utility function with two inflection points, to explain why the same individual can sometimes both buy insurance and gamble. Their model predicts a mixture of risk taking and risk aversion, but not for reasons to do with basic needs. Rather, they predict maximal risk taking among people in the middle of the resource distribution. People in such a position were hypothesized to have strong incentives to move up into an elite, which they can only do through big gains.

What about investment managers? “Roy (1952) analysed the case of a portfolio manager trying not to maximize expected gain, but to minimize the risk of catastrophe (he cites ‘death, bankruptcy, and a prison sentence’ as examples, p. 433). He framed the problem as a step-like utility function.” Similarly, managers of banks that are insolvent have an incentive to “gamble for resurrection”: that is, if a high-risk strategy succeeds, they get to keep their position at the bank; if it fails, the bank was already going broke and the regulators were going to step in.

What about subsistence farmers in low-income countries? “Since Chayanov (1926), it had been commonplace to analyse peasant behaviour as aimed at guaranteeing a minimal harvest necessary for subsistence. The implications of this motivation for risk taking were formalised by multiple authors in the 1970s. Roumasset (1976) and Kunreuther & Wright (1974) proposed hierarchical goal models: the producer first tries to ensure that they do not fall below a ‘disaster level’, and only if sufficiently confident of succeeding, try to maximize profit. Masson (1974) represented this idea using utility functions with ‘jump discontinuities’. These authors had both the S-shaped utility function, and a concept of basic needs.”

There are other applications as well. For example, apparently some biologists have a “risk-sensitive foraging theory”: “When starvation is imminent, animals should take any level of risk to gain food (desperation prediction), whereas once it has been averted for the time being, they should avoid too much risk (caution prediction).” Once the S-shaped curve has entered your mind, you can see it in many places.

For some people, their pulse beats faster when their favorite sports team wins, or their favorite musician drops a new album. For others, pulse and respiration accelerate when confronted with “Spending, Taxes, and Deficits: A Book of Charts.” If you (like me) are in this second category, Jessica Riedl has got your back (Brookings Institution, April 2026). This is a book with a lot of pictures and not too many words.

Here’s the basic picture. Federal debt as a share of GDP is near the highest level in US history, which happened after the US financed the fighting of World War II with borrowed money. National debt is now about $235,000 per US household. The pattern of the last two decades shows that we fought the Great Recession of 2008-09 and the pandemic recession of 2020 with borrowed money, too. Assuming no future giant recessions, wars, or pandemics, past tax cuts expiring when they are scheduled by law to do so, and interest rates not rising further, this is the path we’re on.

Those of you with long memories may recall that back in 2000, the US government actually had budget surpluses for a couple of years. What has changed since then? The short answer is tax cuts, higher spending on Social Security and Medicare, and higher spending on interest payments.

I sometimes talk to people who believe that the defense budget is the biggest part of federal spending. I heard someone refer to the “military-industrial complex” the other day, a phrase form President Eisenhower’s 1961 farewell address. If you hold this belief, time to update your priors. At the time of Eisenhower’s speech, defense was about half of all federal spending, while Social Security/Medicare and anti-poverty programs wer about 16% of federal spending. Those categories have now (roughly) switched places. Social Security/Medicare and antipoverty progrmas ar now 53% of total federal spending, while defense is about 13%.

One grim sign, when you are in debt, is the vicious circle where your interest payments are so high that you can’t reduce your current borrowing, and so you sinkhole of debt keeps getting deeper and deeper. On its current trajectory, interest payments in the US federal budget are headed down this path. The purple bars show the rise in federal interest costs since 2018; the green bars show the path the budget is currently on. On this path, about 30% of all federal revenues will be going to interest payments on past debt by 2036–and interest payments would be a higher cost than Social Security.

Reidl uses the charts to make a case that addressing the long-standing long-term insolvency of Social Security and Medicare would go a long way to addressing these issues. US politicians have been ducking this topic for decades, and continue to do so. Based on selection of charts and labels, Riedl is clearly less enthusiastic about tax increases, but in my own mind, going back in the direction of the tax burdens from back in 2000s can be a reasonable step as well. For example, I would favor moving the estate tax back toward its 2000 levels, and I’d at least be willing to talk about eliminating all income tax reductions–which at this point only apply to about 10% of all taxpayers who tend to have higher incomes. But Riedl offers the useful reminder that the more generous welfare states of Europe are not in fact financed by taxing high incomes and wealth at conspicuously above US levels, but instead by value-added taxes (think a national sales tax) and higher payroll taxes. In short, they provide additional middle-class benefits by taxing the middle class.

The basic patterns of US manufacturing jobs are fairly well-known. Here’s a figure showing total US manufacturing jobs over the last half-century or so: more-or-less flat from the 1970s up through 2000, then a decline from 2000 to about 2010, and more-or-less flat since then. Part of what happened in the early 2000s was a surge of imports from China, after China became a member of the World Trade Organization. Another part was that US workers who might otherwise have been in manufacturing jobs shifted to house-building during that boom-and-bust of US housing in the lead-up to the Great Recession. Part of the reason was US manufacturing firms making much more widespread use of robotics.

Putting manufacturing jobs in the context of total jobs in the US economy gives something of a different impression. This figure shows that even when total manufacturing jobs seemed to be more-or-less holding their own from 1970 up to about 2000, other jobs in the US economy were actually growing. The result is that manufacturing jobs were about 26% of the US total in 1970, but the percentage had fallen by half to 13% of all US jobs by 2000. From this perspective, the decline of the role of manufacturing jobs in US employment is a long run process–a little faster in the decade after 2000 than in the decades before, but not much faster.

Vittoria Dicandia  provides a closer look at some long-term patterns in US manufacturing jobs in “Manufacturing in the United States” (Economic Commentary: Federal Reserve Bank of Cleveland, September 21, 2026).

1) The share of US manufacturing workers with no college has fallen substantially, and the share with a college degree has risen substantially.

2) College-educated manufacturing workers have a modest wage premium; non-college manufacturing workers do not. A “wage premium” is “defined as the average wage difference between workers in a given industry and those with similar observable characteristics, such as age, education,and occupation, in the rest of the economy.” Dicandia calculates:

3) The traditional story about US manufacturing jobs from 1970 up through about 2000 was that the number of manufacturing workers wasn’t going up, but the productivity of US manufacturing workers was rising substantially. That story no longer seems to be true. In the 1990s and even into the 2000s, productivity in manufacturing rose faster than the rest of the economy; since 2007, productivity growth in manufacturing has actually been slightly negative.

4) Over time, the US workforce has adapted to a smaller share of manufacturing jobs not so much by an outflow of existing manufacturing workers, but by a lack of inflow of younger workers into the sector. A result is that current US manufacturers are concerned that they would like to increase production, but lack the workers to do so. (Of course, the economist in me suggests that increasing the wage premiums would help them in attracting more workers.) Dicandia writes:

Alongside this persistent decline in employment share in the sector, manufacturers themselves report significant workforce challenges. The 2025:Q3 Quarterly Survey of Plant Capacity Utilization (US Census Bureau) reports that more than one-fifth of all manufacturers indicate that they were not able to work at full production capacity because of an insufficient supply of labor, while the National Association of Manufacturers survey in 2025:Q4 reports that more than half of manufacturers identify attracting and retaining qualified employees as a relevant challenge for their business.

Worker flows into and out of manufacturing have evolved over time but do not differ markedly from patterns observed in other sectors of the economy (results not shown, author’s analysis of US Census Bureau LED data). This suggests that worker mobility involving manufacturing largely reflects economy-wide labor market dynamics rather than sector-specific turnover. Instead, the key driver of manufacturing’s employment decline has been the lack of younger workers entering the industry (results not shown, author’s analysis of CPS data). Specifically, between 1960 and 2009, each successive cohort of workers has entered manufacturing at progressively lower rates. 

Amidst all the policy proposals, claims, and counterclaims about energy use and carbon emissions, it’s useful to touch base from time to time with basic current facts. One of my starting points is the Statistical Review of World Energy, published annually (free registration needed to download the report). The report is full of figures and tables, with some text sprinkled in. Here, I’ll just pass along a couple of big-picture fact patterns.

This figure shows sources of energy for regions around the world, as well as energy supply by region. Energy is measured by joules, and global energy supply is about 600 exajoules (that is, quintillion joules). The vertical axis shows total energy supply in each region, with the height of the region proportional to energy supply. The horizontal bars then show the share of energy for that region from the main primary sources. (Electricity is not included here because it isn’t a “source” of energy, but rather a way of transmitting and storing energy produced by primary sources.)

For advocates of net-zero carbon emissions, figures like this are a cold slap of reality. The shades of gray are fossil fuels. If the goal is literally to reduce the gray bars to zero, there is a very long way to go. If you look up energy supply per capita in the tables, the challenge of reducing carbon emissions looks even bigger. Countries of Africa, for example, average 14 gigajoules of energy supply per person. India averages 26 gigajoules of energy per person. In contrast, the nations of Europe average 106 gigajoules per person and the United States has 270 gigajoules per person. As lower-income countries pursue economic development, they will want to expand energy consumption quite substantially–like tripling or more.

Looking at world carbon emissions from energy for 2025, the US accounts for 13.3% of the world total, although US carbon emissions have been falling at an average annual rate of 0.8% for the previous 10 years. The nations of Europe as a group account for 9.9% of the world total carbon emissions from energy, with the total falling at an annual average rate of 1.7% in the last decade. In contrast, the Asia Pacific region as a whole accounts for 53.4% of global carbon emissions from energy, and for this region, the total has been growing at an average annual rate of 2% per year in the last decades. This region includes China, which accounts for 31.3% of all global carbon emissions, growing at 2.1% per year over the last decade, and India, at 8.3% of global carbon emissions and growing 3.4% per year over the last decade.

If the goal is to reduce future global carbon emissions, steps taken in the US and Europe will have only a modest direct effect. However, the US and Europe can play a meaningful role in developing the technologies and finding the capital for a vast expansion of non-carbon and low-carbon energy sources around the world–especially in the low-income countries and the Asia Pacific region.

Adam Smith’s 1776 book, An Inquiry into the Nature and Causes of the Wealth of Nations, is often described in casual conversation (at least by noneconomists) as a manifesto for the benefits and power of free markets. That’s a part of it, to be sure. But Smith was a profound thinker, not a pamphleteer, and he digs down into a much wider discussion of why the people of certain nations become wealthy. Barry R. Weingast offers his own thematic emphasis emphasis in “What Every PhD Student Should Know About Adam Smith” (Annual Review of Economics 2026, pp. 647-668).

Weingast runs through the Smithian topics that are familiar to most economists: division of labor, supply and demand, wages and apprenticeships, the role of money, poor laws, land rents, free trade, and so on. But Weingast’s emphasizes Smith’s discussion of how the laws and institutions supporting these factors arose. Where the lords ruled, those who worked the land called “slaves” by Smith. They had no right to move to another area, and if the land was transferred from one lord to another, the farmers also transferred automatically with the land. In this situation, Smith explains, no farmer under a lord had any incentive to increase production, because as Smith wrote, “[M]en in this defenseless state naturally content themselves with their necessary subsistence, because to acquire more might only tempt the injustice of their oppressors.” Smith also offers data that free men, in contrast to the “slaves” under control of feudal lords, produced about three times as much. Weingast calls this the no-growth “feudal equilbrium.”

Weingast encapsulates the Smithian argument about the transition away from feudalism in this way. In this feudal times in England involved three primary groups: the king, the great lords, and people in towns. However, commerce begins to travel up and down rivers, and at key points along those rivers, towns start building up to handld and manage the flow of commercial traffic. Weingast describes the power shift in this way:

The trading towns of Northern Europe managed to escape the feudal equilibrium, creating liberty, the rule of law, and long-term economic growth—at least for the merchant elite. How did the towns make this escape? … Smith began with the townsmen as part of the feudal equilibrium. He reports that the king and the lords attacked these unfortunate souls without “pity or remorse.” Over time, however, the king and the townsmen came to see that they had a common enemy in the barons and lords. Indeed, the townsmen were the enemies of his enemies. The king and towns therefore could benefit from a political exchange in which the
towns gave the king taxes, fixed for all time, in exchange for various rights. These included the right to build walls around their cities and to create their own militias for security, to have their own magistrates and judicial systems, and to create their own laws and mechanisms to enforce them. These changes provided the critical political and legal infrastructure to support markets, allowing the towns to get rich over time.

The towns thus created three separate and non-incremental revolutions—in security, liberty, and the economy. The towns controlled all of these revolutions. Moreover, these exchanges were incentive-compatible. The kings valued the revenue, giving them advantages over the lords, and the towns valued their rights, greater security, and the income and profits from long-distance trade. Both parties to the exchange therefore had incentives to maintain the deal, and both had an interest in seeing the lords lose power relative to their own power. Over time, the towns grew in income and stature relative to the stagnant agriculture in the countryside. Indeed, as the towns grew rich, they were ahead of the countryside by two centuries from the standpoint of political and economic development. Smith summarizes this process: “Order and good government, and along with them the liberty and security of individuals, were, in this manner, established in cities at a time when the occupiers of land in the country were exposed to every sort of violence” …

The towns also created an important transformation of the local countrysides as the towns brought these areas under their (local) security umbrella and with it a semblance of property rights, contract enforcement, and other aspects of the foundations of liberty and the rule of law. This allowed the former slaves in these areas to become specialists in the production of food and raw materials sold to merchants in the town for local consumption or for export in long-distance trade. Per Smith’s division of labor, their incomes were sure to rise above
the subsistence agriculture they had previously suffered for generations. Of course, this transformation occurred largely in the vicinity of the towns at first, spreading into the countryside far more slowly. Nonetheless, much of the agrarian hinterland remained trapped in the feudal equilibrium.

The towns used these rights from the king to take part in long-distance trade. I argue that, in many areas, resources became scarce over time, especially wood for building and for warmth (e.g., firewood). Every year this problem worsened as the scarcity increased. Holding constant for the risks, this scarcity slowly raised the value of taking part in long-distance trade. These merchants became enormously rich as they supplied the local economy with various goods that were scarce in their own towns while at the same time brought goods made or prepared locally that were scarce in other towns, often hundreds or even thousands of miles away. Again, we see here Smith’s division of labor at work.

The towns also had considerable violence potential. Given the violence of the lords, the violence potential of the towns had to be even higher. Indeed, this prowess in violence seemed a necessary and existential condition for the town’s survival. Had this condition failed, the lords would have continued to plunder the towns without pity or remorse.

In short, Smith (and Weingast) argued that the underlying conditions for division of labor, supply and demand, trade, and the other aspects of market forces were a balance of violence that allowed for public order, and then for contracts and markets to function. Weingast writes:

Indeed, in 1793, Dugald Stewart gave a eulogy for Smith [Stewart 1982 (1793)]. He reported a 1755 paper by Smith then in his possession (now lost), which said, “Little else is requisite to carry a state to the highest degree of opulence from the lowest barbarism, but peace, easy taxes, and a tolerable administration of justice; all the rest being brought about by the natural course of things.” To a neoclassical scholar of development, this sentence makes development sound formulaic. The reason, as I mention in the introduction, is that these economists assume the first and the third conditions when studying the developed world. But, as I suggest, Smith argued that creating these conditions is the task of development. To create a rule-of-law judicial system rather than a corrupt one based on bribery and privilege represents one of the central problems of modern development.

Of course, no country eliminated bribery and privilege altogether. There is always an element of “you-scratch-my-back, I’ll-scratch yours” and “who you know matters as much as what you know.” There are always interest groups lobbying politicians, and politicians seeking contributions. There are always people looking to build their own power. Adam Smith was a consummate realist, and in the quotation above refers to “a tolerable administration of justice,” rather than perfect justice. But even in a high-income country, pushing back hard against bribery, corruption, and privilege matters, because those factors erode the foundations on which the wealth of a nation is built.

Aquaculture refers to farming the water: that is, growing fish or shrimp or other aquatic animals in a controlled space, rather than capturing them in open water. I’ve tended to think of aquaculture as an interesting sideline to standard capture-style fishing, but apparently that view is so 1995. The UN Food and Agriculture Organization (FAO) has published The State of World Fisheries and Aquaculture 2026, which shows that the quantity produced by aquaculture has now surpassed conventional fisheries.

Indeed, this figure has two substantial lessons. One is that the total catch from capture fisheries has not increased in the last 40 years or so, since the 1980s. I have seen this pattern described both as a good thing, because it represents better management and quotat systems to protect fisheries, and also as a bad thing, because it represent the fragility of ecosystems that have been and are being overfished. Of course, both explanations can be true, and either way, substantial future growth in output from traditional fisheries seems unlikely. But in this report, I was especially struck by the rise in aquaculture, which has had larger output than traditional fisheries since 2013.

This shift to aquaculture has been concentrated in China and other countries of Asia, although there is a smaller but noticable rise in countries of Africa as well.

You can dig into the report for details of what aquatic animals are being farmed, along with seaweed and other output like shells and and pearls. For myself, I had not previously read about the distinction between fed and non-fed aquaculture: “Non-fed aquaculture of aquatic animals refers to the cultivation of species that obtain nutrients from the surrounding environment, such as plankton and organic matter in the water, without artificial feeding. This category typically includes filter-feeding species, such as certain carps and bivalve molluscs, and is generally considered more environmentally sustainable due to its low input requirements, minimal waste generation and water filtering properties.”

it’s natural to wonder about whether aquaculture is, at least in some cases, causing environmental problems of its own. The report discusses some high-tech example, like large ponds that are constructed with solar panels overhead, which both provide power to clean and filter the water and also provide shade so the water temperature doesn’t rise too much in the heat of the day. A company that builds such a facility has a strong incentive to prevent it from becoming a source of pollution. In other cases, aquaculture refers to cage-grown fish in a water environment where the cost of waste may be imposed on others who are nearby. But broadly speaking, it seems easier to me to address environmental harms of aquaculture than the environmental harms–like species extinction–that can arise from traditional capture overfishing.

The price of candy has risen considerably, and when a price goes up, there are always two school of thought. One argument is that greed and “big candy” has jacked up the price in pursuit of higher profits. The challenge for this explanation is to explain why big greedy companied had not already raised prices last year, or the year before. It seems implausible that their level of greed suddenly increased, and if or when candy prices fall again, will it mean that they deserve praise for diminishing their greed? The alternative explanation is rooted in factors affecting candy supply. Natalie Ho provides details in “Why is candy so expensive these days?” (Monthly Labor Review: US Bureau of Labor Statistics, September 9. 2026).

Here’s the basic pattern: the orange line shows the overall inflation rate as calculated by the Consumer Price Index; the blu line show the inflation rate for the “candy and chewing gum” subcategory.

Ho offers some basic facts. For example, about two-thirds of total annual US candy sales are linked to four holidays: Halloween, Christmas, Valentine’s Day, and Easter. Also, “[f]or the U.S confectionery industry, approximately 65 percent of ingredient costs come from three material inputs: cocoa beans (28.4 percent), sugar (8.7 percent), and corn syrup (28.1 percent).” Thus, events that drive up prices of key inputs–especially cocoa and sweeteners– will drive up the output price of candy.

For cocoa beans: “Global production dropped more than 13 percent in the 2023–24 season compared with the 2022–23 season, falling from approximately 5.044 million metric tons to 4.365 million metric tons. At the same time, global cocoa demand was estimated to decrease only 5 percent, dropping from 5.058 million metric tons to 4.810 million metric tons. … The decrease in global production stemmed primarily from Côte d’Ivoire and Ghana, the world’s largest cocoa producers. Together, they generally account for 50 to 60 percent of total global production. Both countries have experienced poorer weather conditions since 2021. In 2022, Côte d’Ivoire and Ghana had hotter and drier weather, reducing soil moisture and the size of beans. In 2023, Côte d’Ivoire received the highest total rainfall in its cocoa-growing regions in the previous 20 years. The excessive rain led to cacao swollen shoot virus and black pod disease, killing cocoa trees and destroying beans. The torrential rain also delayed deliveries to ports. Subject to similar problems, Ghana’s 2023–24 crop yield reached a 23-year low. … Further, speculation in the cocoa futures market also pushed prices higher. Hedge funds and other traders joined the cocoa market in droves starting at the end of 2023. By early 2024, traders purchased a record $8.7 billion worth of cocoa futures contracts between the London and New York markets.  Although likely not the major cause of the rise in prices, these hedge funds contributed to the increase, with prices more than doubling over the previous year by 2024. With chocolate companies also attempting to secure supplies, the cocoa futures market reached prices not seen in over 40 years, as the futures markets were traditionally only used to hedge risk. From October 2022 onward, prices climbed sharply. They peaked initially in April 2024 at just over $11,000 per metric ton. … Later in the year though, prices resumed their ascent, climbing another 10 percent, reaching a record $12,565 per metric ton in December 2024. By comparison, from 2000 to the third quarter of 2022, cocoa future prices stayed between $1,000 and $3,500 per metric ton. This overall price increase did not, however, initially directly translate into higher prices for farmers, as contracts were signed 8 to 12 months prior to the 2023–24 season when cocoa futures were less than half of what they were in November 2023.”

On sugar: “Sugar in the United States is relatively expensive compared to the rest of the world, largely because of the United States sugar program that restricts sugar imports. Mexico has generally provided nearly half of United States sugar imports, with a policy beginning in 2008 that made Mexican sugar duty-free and quota-free. However, in 2014, the two countries suspended the quota- and duty-free policies and agreed to implement minimum price and quantity restrictions on imports of Mexican sugar, which led to an increase in sugar prices. … Many of the major sugarcane-producing states in central and southern Mexico experienced droughts in recent years, including a major drought in marketing year (MY) 2019–20 and a less severe drought in MY 2022–23. The 2023–24 season’s exceptional dryness reduced Mexican sugar production to its lowest volume in 24 years and its lowest export volume to the United States in 17 years. … In addition, global sugar prices increased as El Niño caused unusually dry weather in India and Thailand (the second and third largest sugar exporters globally), which resulted in reduced production in those countries in MY 2023–24. Port bottlenecks, fire, and competition from other crops for shipping services also affected exports from Brazil (the largest global sugar exporter). Further complicating the issue was competing uses for sugar, as 51 percent of Brazil’s 2023 sugarcane crop was used for their domestic production of ethanol.”

Of course, the supply shocks to input prices will affect the rest of the market. The quantity of nonseasonal chocolate purchased in the US has been declining since 2022. “Shrinkflation” happens when candymakers reduce the size of their products. But big price movements are almost always about changes in underlying condidtions of supply and/or demand, not fluctuations in greediness of sellers.

Each year, the White House Council of Economic Advisers publishes the Economic Report of the President. The CEA is led by politically appointed academics, and thus you need to take it as given that the annual report will be supportive of the presidential administration. On the other side, the academic economists are there for relatively short-term appointments (typically a few years), and will typically be returning to academia, so their reputation for making arguments based on evidence and theory is (at least somewhat) at stake. In the 2026 Economic Report published earlier this year, Chapter 9 is titled “Work Means More Than Making a Living: Labor, Challenges, and Opportunity,” and it starts with a discussion of what it calles the “latent” value of work, focused on dimensions beyond quantities produced and compensation paid. The report argues:

Work is a center of meaning, a means of identity and purpose, and a powerful node of human relationship and connection (Durkheim 1897; Jahoda 1982). Understanding these “latent” benefits of work is essential for crafting policies that truly support human welfare. … [F]our striking facts are documented: work improves health and extends life by years; work increases happiness and reduces depression; work helps to develop the knowledge and skill of individuals; and work provides a webbing of social connection and community.

On the link from work to improved health:

If work were merely a source of stress to be escaped, one would expect job separation to improve health outcomes and reduce mortality. The evidence shows precisely the opposite. Workers displaced through mass layoffs experience 50–100 percent higher mortality rates in subsequent years, shortening life expectancy by 1 to 1.5 years on average (Sullivan and von Wachter 2009). Given standard estimates of the value of a statistical life from the U.S. Department of Health and Human Services (DHS 2025), the reduction in life expectancy costs about $300,000–$500,000. These are not workers who lost jobs due to poor health or performance; they were displaced through plant closures and downsizing that had nothing to do with individual circumstances. The excess deaths span multiple causes—strokes, heart attacks, accidents, suicide, and complications from substance abuse—suggesting that job loss triggers a cascade of physical and mental health deterioration (Eliason and Storrie 2009; Browning and Heinesen 2012). These studies, which are from Sweden and Denmark—countries with government healthcare—demonstrate that the health consequences of job loss operate through psychological and social pathways rather than simply through the loss of medical access.

The most compelling evidence that these effects transcend income loss comes from retirement. Fitzpatrick and Moore (2018) exploit the sharp discontinuity in retirement timing at age 62, when Social Security eligibility begins. Roughly one-third of Americans retire precisely at this threshold. This age-based threshold creates a natural experiment, since retirement timing is driven by policy rather than individual health status. Despite having their income partially replaced through Social Security, these new retirees experience animmediate 2 percent increase in mortality risk—implying that retirement itself increases mortality risk by about 6 percent for those who choose it. The effect is particularly pronounced for men, who may especially derive their identity from work.

On how work increases happiness and reduces depression:

Beyond its effects on physical health, work shapes human emotional well-being in ways that contradict people’s stated preferences. People tell themselves that they work for the weekend, yet the evidence suggests that they find more fulfillment in Monday morning than Saturday afternoon. Csikszentmihalyi and LeFevre’s (1989) pioneering study revealed this paradox elegantly. Using experience sampling methodology—giving subjects beepers that prompted them to record their activities and feelings at random moments—they found that people reported higher levels of happiness, engagement, and energy while working than during their leisure time. This finding challenges classical economic theory, which treats labor as “disutility”—a cost paid in lost leisure to obtain income. Despite the common cultural narrative that frames work as drudgery, people appear to derive deep satisfaction from productive effort and the structure it provides. …

Research on unemployment duration reinforces this pattern. Bayer and Juessen (2015) demonstrate that short unemployment spells have minimal effects on well-being, while persistent joblessness drives significant unhappiness. This suggests that the psychological benefits of work operate through multiple channels—not just the act of working, but also the security of knowing you will have work. Krueger and Mueller (2012) reinforce this interpretation, documenting how laid-off workers experience not just depression but also an explicit sense of meaninglessness.

On how work helps to develop knowledge and skill:

The modern workplace demands constant adaptation—new software, evolving best practices, shifting market demands. Workers meet these challenges not usually through formal training but by daily problem-solving alongside colleagues. This learning-by-doing, which is nearly invisible when it occurs, becomes painfully apparent only in its absence (Arrow 1962).

Bessen (2016) documents how workers master new technologies, primarily through on-the-job experimentation rather than formal education. Similarly, Jarosch, Oberfield, and Rossi-Hansberg (2021) show that workers learn substantially from their peers, with an increase in coworker quality raising own wages by 2 to 3 percent permanently of 1 standard deviation. These knowledge spillovers happen through informal channels—watching a colleague handle a difficult client, debugging code together, sharing Excel shortcuts across workstations. The cumulative returns from this workplace learning are substantial and persistent. Consider a young worker in his 20s considering whether to take a gap year after completing his training. One additional year of work experience increases his annual earnings by 3.0 to 3.6 percent throughout the rest of his career (Altonji and Williams 1992). For an average worker earning $60,000 at age 25, this translates into about $86,400 in additional lifetime earnings before retirement—all from a single additional year of early career experience. The workplace functions as an unstructured but irreplaceable quasi-university where learning happens through osmosis and the mother of all: necessity.

On how work provides a webbing of social connection and community:

[T]he benefits of work extend beyond the individual psyche and skill set to the social fabric that binds people together. Work is where people form friendships, find mentors, and build the weak ties that sociologists have long recognized as essential for opportunity and belonging. When work disappears, so does this essential social infrastructure. … [J]ob separation triggers sustained social isolation. … Unlike depression, which spikes immediately after a job loss, loneliness emerges gradually and persists. Ten years after separation from work, both unemployed and retired workers report significantly higher loneliness than when they were working. …

These individual costs ripple outward through communities. When unemployment rises locally, even those who keep their jobs experience decreased well-being—Helliwell and Huang (2014) find the effects equivalent to a 4 percent decline in household income. This spillover reflects the breakdown of social networks: when your colleague loses their job, you lose a colleague. When enough people lose work, the entire community fabric begins to fray. The social consequences of joblessness extend into people’s most intimate relationships. Work shapes not just friendships but also family formation. Autor, Dorn, and Hanson (2019) demonstrate that regions hit by trade-induced job losses experience collapsing marriage markets, particularly for noncollege men whose diminished employment prospects make them less attractive as partners. The results are fewer marriages, more single-parent households, and declining fertility—a cascade of social disconnections that begin with economic displacement.

The broad takeaway from these arguments is that work is more than a contractual relationship between an employer and an employee. Work involves human beings, with all their human issues of health, happiness, skills, and social connectedness. The US economy benefits considerably from its ability to reallocate labor between employers, industries, and places. But the costs of that reallocation are much lower when workers voluntarily leave one job to take another one, rather than when workers lose a job involuntarily.