Category: Economics

Revisiting Alan Greenspun’s 2007 book

by philg

Alan Greenspan died this week at 100. Let’s look back at Alan Greenspan Explains Modern Economics, my 2009 review of his fall 2007 book (i.e., he wrote it just as the U.S. subprime collapse was beginning).

Here’s a cautionary for folks like me with portfolios that are 100% invested in SpaceX:

One interesting story sheds light on the limitations of government economic forecasts. A booming economy and stock market swelled federal tax collections so much that there were unheard-of federal budget surpluses during the final years of the Clinton Administration. The non-partisan Office of Management and Budget, in the first half of 2001, predicted that federal budget surpluses would grow year after year. Everyone was trying to figure out what the Feds would do once they’d paid off 100 percent of U.S. long-term debt. Would there be massive tax cuts? Would the U.S. government start buying hard assets in other countries, the way that sovereign wealth funds from China and the Arab countries do now? Everyone in the government, including Greenspan, was shocked when the surpluses evaporated almost overnight. The forecasters hadn’t figured out that a sagging stock market would mean an end to collecting capital gains tax.

It looks as though Greenspan was prescient regarding the issue that has propelled socialists and “super-progressives” into political power here in the U.S. recently.

Greenspan sprinkles the book with discussions about income inequality. Greenspan says that as an economy becomes more productive, the returns to having good skills and being smart will increase (Gregory Clark has some statistics in Farewell to Alms showing the opposite; the returns to skilled labor in England fell and unskilled laborers were the biggest beneficiaries of economic growth). He thinks that the minimum skill level necessary to be productive in the U.S. is now far above what the graduates of our pathetic public school systems are capable of. He thinks it would be politically infeasible to turn our schools from unionized employee paradises into centers of educational excellence. With only dumb young Americans as a labor source, the U.S. economy will stagnate. His solution to continued economic growth is therefore a massive expansion of immigration of smart, well-educated, highly skilled workers from other countries. (Note that Chinese schools on average don’t have to be better than U.S. skills; we just need to attract immigrants from among the millions of Chinese who are better educated than the U.S. average.) Greenspan opposes our current immigration system, which does not give much weight to an immigrant’s potential as a worker.

It seems that Alan Greenspan didn’t realize that Somali immigrants had built Minneapolis and Boston, as their respective mayors now inform us. We can also look at this part of the book as an example of how little effect on policy even the most powerful Washington insiders can be. Greenspun was Chairman of the Fed. He had access to every member of Congress and four presidents. He presumably did tell these lawmakers “Hey, you should really put in IQ, education, working age, health, and income requirements on every immigrant and eliminate the U.S. asylum system so that you don’t just import needy humans into a massive welfare state.” What was the effect? The politicians doubled down and and then tripled down on low-skill immigration, oftentimes of people too old to work and guaranteed to need taxpayer-funded everything (e.g., automatic green cards for 75-year-old parents of new U.S. citizens, who might themselves be aged 50 or 60).

Perhaps we can give Greenspan credit for predicting our Age of AI and Robots (the perfect time to be importing low-skill humans!)?

Various portions of the book are sprinkled with Greenspan’s enthusiasm about technology and what it can do for productivity growth. He is basically optimistic about the future because humans will figure out how to do more with less. Like any good economist, he hedges his predictions of a prosperous 2030 here in the U.S. The main risks that he sees are Islamic terrorism and a resurgence of protectionism that would undo the benefits of globalization (you won’t find Greenspan showing up to protest a WTO meeting!). The main challenge that he sees is funding Medicare and Social Security, which are currently pay-as-we-go (i.e., Ponzi schemes). Despite increased immigration, taxes will rise to crushing levels and benefits will fall. The Europeans will be in even worse shape because they don’t have as much immigration. Greenspan does not address the issue of why a group of citizens would wish to pack their country with double the number of people in order to pay for their retirements. He puts no value on living in an uncrowded place with reasonable real estate prices and traffic.

(I don’t give him credit for predicting the impending insolvency of Social Security, which was obvious from them paying out benefits to the very first recipient that were 1000X what she’d paid in via taxes. Maybe I can give myself credit for predicting that immigration between 2009 and now would doom Americans to UNreasonable “real state prices and traffic”?)

Speaking of real estate prices, New York Times today (example of “When the market gives you an answer that you don’t like, declare market failure”):

Excerpts:

“We’re in a full-blown housing crisis,” Senator Elizabeth Warren of Massachusetts, the top Democrat on the Banking, Housing and Urban Affairs Committee, said in an interview. “Home prices are sky high, rent is through the roof. The median age of a first-time home buyer is at an all-time high. So the pressure to move was almost irresistible. This bill got through because it is big.”

Chief among the sticking points was a provision to check institutional investors, which had been crafted in negotiations among White House officials, Senator Tim Scott, the South Carolina Republican who leads the Banking Committee, and Ms. Warren.

The measure prohibits corporate entities from owning more than 350 existing single-family homes, although it does not require them to sell homes purchased before the measure became law. A stricter proposal that would have required investors to sell single-family homes built explicitly as rentals after seven years was dropped; it had prompted a backlash by home builders and affordable housing advocates, who feared it would discourage new home construction.

I can’t figure out how this is Constitutional. Most of what the federal government does is allowed, despite contradicting the Framers’ intent, because of the Commerce Clause:

[The Congress shall have Power] To regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes;

Suppose that a corporate entity in the Mamdani Caliphate buys 351 houses in Buffalo, NY in order to rent them out via Section 8 to noble asylum-seekers from the world’s various dysfunctional societies. How is that an example of interstate commerce “among the several states”? Let’s assume that the corporation is a New York corporation based in New York and that 100% of its employees are New York State residents. The houses are all in New York State. It would be physically impossible for one of these houses to be sold for use in, e.g., the Islamic Republic of Michigan. Regardless of the merits of the law, how can the federal government do this? Why doesn’t it have to be a state-by-state decision?

Full post, including comments

If the 1950s were a “rat race” for men, what is the correct term for the 2020s?

by philg

White-collar men in the 1950s often characterized their world as a “rat race”. If college-educated, they competed with only a small subset of Americans for high-paid desk jobs. Men in the 50s did not compete with immigrants because substantial importation of humans into the U.S. stopped in 1924, not to be restarted until President Johnson signed the Hart-Celler Act in 1965 (see below). A house in a safe suburb with good schools and A/C could be purchased, at the end of the 1950s, for about one quarter the cost today (in real dollars; see $112/month to live in a brand-new house in Bowie, Maryland). Relative to income, a house cost about 1.7X annual salary vs. over 5X today (ChatGPT table below). Partly due to this low cost for housing in a safe suburban neighborhood with decent schools ($1+ million today?), a man’s income was generally sufficient to support a wife and 2-3 children as well as himself. Sex outside of marriage was discouraged both legally and socially and, therefore, the man would usually be married before age 25. No-fault divorce (“unilateral” in research parlance) did not exist and, therefore, if the man wasn’t behaving outrageously (beating the wife, drinking heavily, failing to work, having affairs), the wife couldn’t profit via a divorce lawsuit (a divorce might be arranged by mutual agreement, of course). In addition to marital security, the 1950s man often enjoyed a lot of job security from (1) the lack of competition in the labor market, and (2) the tendency of large companies to provide lifetime jobs, which today is limited to government work.

What’s the correct term for what similar men face today? They inhabit a world in which you can’t spit in the street without hitting a college graduate. Men must compete with women for jobs and, despite women being more likely to earn college degrees, be passed over for hiring or promotion when a company decides that “diversity” is its strength. If a female or favored minority human competitor doesn’t take the white-collar man’s job, Claude is ready to replace him. The companies that once offered native-born Americans jobs for life are now home to platoons of H-1B “non-immigrant” immigrants.

A house in a neighborhood with low crime, an orderly familiar culture, and good schools, is about 10X the median college graduate’s income (5X for houses overall, but the typical suburb is no longer a white picket fence idyll). A college education for the kids, so that they can get into the “rat race” that the parents ran, is now 5X more expensive state colleges and 9X more expensive at elite Queers for Palestine-type schools . (ChatGPT on the history of federal government programs to make college more affordable: “GI Bill for veterans in 1944, first general federal student loans in 1958, major modern federal aid framework in 1965, and Pell-style direct grants in 1972/1973”)

Where in the 1950s he likely partnered with a virgin aged 20 (ChatGPT says 10-25% of 1950s brides might have had sex with someone other than their fiancé/husband), today he’s with a 30-year-old veteran of the sexual revolution. If he is persuaded to marry her, she can sue him for divorce a day later for any reason or for no reason. For men who strayed in the 1950s and got sued for a “fault divorce”, the resulting financial drain was primarily alimony and it lasted only a few years because the plaintiff would remarry and that shut down the alimony revenue stream. The risk of losing his role as a father was controllable due to the requirement that a plaintiff find a “fault” ground, such as infidelity. If a man gets sued today because the wife found someone she likes better, the man can lose his “father” role, and access to the young people who used to be his children, due to factors entirely beyond his control. The man’s biggest financial exposure in a divorce lawsuit is typically “child support” (paid to an adult female to spend on whatever she wants, not to a “child”), which can last for 23 years (Massachusetts) or 21 years (New York) even if the plaintiff has married her lover and that lover earns far more than the defendant and even if the lover is the biological father of the child (nytimes: “I pay child support to a biologically intact family, a father and mother, married, who live with their own child.”). (In the cases where alimony is the primary profit from a divorce lawsuit, the defendant might be paying for 50 years because there is no longer any social pressure for the plaintiff to remarry. She can have sex with 100 men and write a magazine article about the “single MILF” lifestyle and this has no impact on her cash entitlement.)

This is not to say that American in the 1950s was better overall, of course. We had been starved of enrichment via immigrants since 1924 and, therefore, weren’t as strong under the “diversity is our strength” axiom. We didn’t have Internet or LLMs for personal use. A 1950s car, though beautiful in our museums today, came out of the factory as a junk heap compared to a 3-year-old Toyota today. We had three TV channels to watch on a 21″ CRT. But in terms of career security and personal life security, the 2020s are inferior to the 1950s. So, returning to the title question… if the 1950s were a “rat race” for white-collar men, how would we characterize the situation today?

Loosely related…

A post on X from an offensively titled username so I’ll just copy the text:

White Americans and Europeans are the ONLY people worldwide that are EXPECTED to compete with the ENTIRE world for jobs.

50+ years ago White men with STEM degrees got good jobs. Things like engineering or applied mathematics guaranteed a good career.

Now we are required to compete against not just our own people, but the brown and black hordes worldwide that are willing to work for pennies.

It was an ECONOMIC CRIME committed against our people.

(I post this not for the truth or falsehood of what the author writes, but for the expression of a feeling of insecurity and, therefore, pressure even worse than the rat race of 50 years ago.)

Full post, including comments

Nobel-grade do-gooding (microfinance) considered harmful

by philg

In the spirit of “Go To Statement Considered Harmful” by Edsger Dijkstra, renowned sourpuss…

“Hundreds of Billions in Loans Didn’t Make a Dent in Global Poverty” (Wall Street Journal):

Microfinance, loans issued in communities not served by traditional banks, would help poor people in developing countries start businesses and work their way toward prosperity. That was the goal of Muhammad Yunus, a U.S.-trained economist, who pioneered the practice in Bangladesh during the 1970s.

“In a poverty-free world, the only place you would be able to see poverty is in the poverty museums,” Yunus told his audience in Oslo in 2006 when he accepted the Nobel Peace Prize for his work.

Led by the adage of “doing good while doing well,” microfinance lenders have since advanced hundreds of billions of dollars to poor people in countries from Albania to Zimbabwe. Prominent voices including Hillary Clinton and Natalie Portman told inspiring tales of women entrepreneurs lifting the fortunes of their communities. Along with easing poverty, microfinance aimed to expand access to education and end gender inequality.

That was the dream, including for yours truly (I kicked in some money circa 2000 to a web-based microfinance portal). What has been the reality?

Academic studies, including randomized controlled trials, have found that microfinance doesn’t improve the economic conditions of most borrowers. Economists found excessive microfinance lending has set off repayment crises for borrowers in half a dozen countries, including Bosnia, India and Cambodia.

High interest rates, which can top 100% in some Latin American countries, and pressure tactics by loan officers have been tied to suicides, homelessness and children pulled from school to work. Rather than using the loans to invest in small businesses, many borrowers spend the money on medical expenses and other necessities.

Does failure to achieve stated goals have an effect on nonprofit organizations? No.

The hardening evidence of microfinance’s failure to alleviate poverty should have led to a rethinking of its use as a development tool, said Rafe Meager, an associate professor at the University of New South Wales in Australia, who has studied the academic research on microfinance.

“There still hasn’t been this kind of reckoning in a serious way,” Meager said.

The average microfinance borrower in Cambodia owes more than $3,900, nearly three times the median annual per capita income. Average debt per borrower is more than $6,000 when including small loans from microfinance lenders that are now commercial banks also providing other financial services.

Microfinance’s breakneck expansion in Cambodia in the early 2010s coincided with a government push to formalize land ownership. Contrary to Yunus’s vision that debts shouldn’t be collateralized, most of Cambodian microfinance loans greater than $3,000 are secured by a borrower’s land, which is the main hard asset for most poor families.

What happens when we throw AI into this mixture? Some people were already poor because their skill levels were too low to compete in a globalized economy. Do they get a boost in value for a while, at least, because the Optimus-style robots won’t be ready until well after the AI brains are perfected? Or do already-poor people in poor countries become further devalued by AI because they’re being partly paid for the use of their brains? Or, on the third hand, do they get a boost in income because they’ll use AI to become much more productive?

Separately, now that Elon is well on his way to a second $trillion, why isn’t he loaning Bosnians, Indians, and Cambodians however much money they want?

Full post, including comments

Why we can’t simply limit oil and gas exports to 2025 levels

by philg

For the past couple of months I’ve been wondering here why U.S. consumers are paying more for gasoline than in 2025 (albeit still less than in 2022) if the Trump administration could simply limit exports of oil and gas to 2025 levels with a simple “it’s a war” explanation. This question is answered, to some extent, in “The World Can’t Get Enough U.S. Energy, Keeping Prices High for Americans” (WSJ, yesterday):

The Trump administration is trying to tamp down rising prices, including by waiving restrictions on trade between U.S. ports and releasing oil from strategic stockpiles. Trump said last week he supports suspending the federal gasoline tax. Gasoline prices nationally averaged $4.51 a gallon on Sunday and could keep climbing into Memorial Day weekend, the starting gun to the busy summer driving season.

The administration has said it wouldn’t impose a ban on energy exports. Energy Secretary Chris Wright said on CNBC last week that the U.S.’s economic future depends on selling its energy abroad and that this was a top item on the Trump agenda.

“We can’t be a major energy exporter to the world if we decide sometimes to stop exporting our energy,” he said.

In other words, the Trump administration is allowing Democrats, previously climate change alarmists who wanted fossil fuel prices to be higher, to harp on lower-than-2022-but-higher-than-2025 gasoline prices, possibly resulting in dramatic losses of Congressional seats in November 2026, in order to preserve the U.S.’s long-term market position.

What’s the scale?

The ports of New York, Philadelphia and Albany, N.Y., exported 174,000 barrels a day of gasoline, diesel and other petroleum products last month, according to Kpler. That is 10 times the volumes they shipped over the same period last year. Halfway through May, the pace of exports is even higher, well over 200,000 barrels a day—the highest monthly pace on Kpler’s records since 2017.

These barrels so far this month are predominantly heading to Europe, including France, Belgium, the Netherlands and the U.K., Kpler’s Smith said. Analysts say that is a sign that a shortage of refined products has spread from Asia to Europe.

The U.S. exported 2.7 million barrels of U.S. diesel, gasoline and other refined products to Australia in March, according to Kpler. Before the war broke out, exports there had been sporadic. An additional 1.8 million barrels headed to Australia in April.

I wonder if the Trump administration’s policy makes sense even for those who have a long-term perspective. If Democrats can take control of Congress maybe they will obstruct the U.S. fossil fuel industry in some other ways, e.g., with a long-dreamt-of carbon tax.

Separately, why isn’t there a lot more production in response to the higher price? The current price of oil is about 15% lower than it was in 2022 (chart below), but still much higher than it was in 2025:

Maybe it is because the market is predicting a sag down to $89/barrel by October 2026 and a further sag to $75/barrel by October 2027?

The lower chart is curious. Investors have changed their opinion of the likely cost of oil in October 2027, up from about $60 to $75. Are they expecting that we’ll still be at war? That inflation will go back to the raging 2022 levels?

Full post, including comments

Is it bad that Florida is no longer affordable for the middle class?

by philg

Recent Wall Street Journal article, “Florida’s Population Boom Fizzles as High Costs Drive Away Middle Class”:

Florida’s migration patterns are changing dramatically. Residents in their prime working years are heading to other states, often citing affordability concerns. At the same time, the stream of people arriving from other states is shrinking.

Meanwhile, an influx of wealthy people from other states—turbocharged during the pandemic—has helped drive up home prices. Inflation in parts of Florida outpaced the national average over the past decade and home-insurance rates soared.

These side-by-side trends could spell trouble for a state whose economy relies on continued population growth and real-estate development.

“The affordability picture has changed in Florida almost more than anywhere else in the country,” said Eric Finnigan, vice president of demographics research at John Burns Research & Consulting.

First, note the assumption that underlies almost all American politics: infinite growth should be the goal. (Never mind that growth without limit in an organism, and without regard to available resources, is known as “cancer”.)

Second, the WSJ implicitly assumes that a place that is affordable is better than a place that is unaffordable for median-income residents.

Third, the WSJ lumps all of “Florida” together. Florida is about the same size as all of New England. The WSJ wouldn’t lump together Boston and western Maskachusetts, much less Bridgeport, Connecticut and Houlton, Maine. (It’s still possible to get a brand-new single-family house in central Florida for less than $300,000, though the same can’t be said for coastal Florida; the house will be about 1500 square feet, which is the size of the house I grew up in (family of five) and with the added advantage that Floridians don’t need as much indoor space.) The most convenient housing for a SpaceX or Blue Origin engineer is in Titusville, where a decent (not new) house can be purchased for $300,000 (relocation guide).

Fourth, the WSJ assumes that the market is full of stupid people who bid up the prices of houses in places that aren’t desirable. Single-family home prices are $10.15 million in Palm Beach and $212,000 in Dearborn Heights, Michigan, where Ayman Ghazali mostly peacefully lived. From this we can infer that living among Iraqi and Lebanese immigrants in Dearborn Heights is better than living among Manhattan immigrants in Palm Beach (perhaps not an unreasonable inference!).

Maybe in a country with a shared language and culture it would make sense to try to find an inexpensive place to live. However, in a country that is jammed with low-skill migrants from all of the world’s most violent and dysfunctional societies (our asylum-based immigration system ensures that someone from Switzerland or Japan goes to the back of the line), isn’t it actually an advantage from a typical native-born perspective that a place is out of reach for the median present-day American? Google AI: “Newport Beach has lower racial diversity and worse racial disparity across various indicators compared to the average for California cities.” Given the stratospheric real estate prices, it seems that a lot of people are willing to pay for low racial diversity and “worse racial disparity”. As of 2021, the town was supposedly 85 percent white (source):

The Dallas metro area is more affordable than most parts of the US with jobs, which has enabled a mostly-immigrant community of 130,000 Muslims to set up more than 60 mosques and lay out EPIC City, “a master-planned Islamic community-centered residential development project”. Non-Muslim Americans who don’t want to hear the muezzin calling five times per day might prefer to spend more on a house that is in an area that is “unaffordable” to immigrants from Syria, Egypt, Afghanistan, and Somalia.

We could take this to an extreme. Aspen, Colorado is absurdly unaffordable for the median worker. My friend doesn’t like Aspen (see An actual skier goes to Aspen to ski), but apparently a lot of people do like it. Would we say that Dearborn Heights, Michigan is a better place to live than Aspen? That Aspen is bad because the population isn’t growing 3% per year like Gaza’s or Somalia’s? (Maybe Gaza and the West Bank are the ultimate examples of affordability. US and EU taxpayers pay for all of the basics, e.g., shelter, food, health care, education, etc. Nobody needs to work. Hamas-ruled Gaza is a model society by Ivy League standards, but wouldn’t the typical American rather be in St. Barts, Aspen, or Nantucket (all of which rank near the bottom for affordability on a median income)?) We could also consider a massive public housing project in Chicago or New York City. They’re “affordable” by definition since no tenant is charged more than 30% of his/her/zir/their income (often 30% of $0 since the tenants aren’t stupid!). Would a typical American prefer to live in the 6000-person Queensbridge Houses (“well known for its contributions to hip hop and rap music”; “a problem with drug dealers and drug users”) or in Atherton, California (population 7,000; home to Larry Ellison before he spent $450 million to escape to Florida)?

In short, given the continued flood of low-skill migrants (70 million since 1976) maybe “affordability” shouldn’t be the goal for any city or state that seeks to maintain a pleasant environment.

Full post, including comments

Schrödinger’s job market: “strong” and “slow” at the same time (NYT)

by philg

Home page of the New York Times today, the job market is “strong” (top story) and “slow” (just below):

We were informed that Donald Trump’s border closure would destroy the U.S. economy (see U.S. economy defies Science and Immigrants expand our economy, but millions of immigrants exiting the U.S. don’t shrink our economy) and that native-born Americans aren’t willing to work. Yet the number of workers in the U.S. keeps growing even as the number of migrants shrinks and also as the number of federal government jobs shrinks.

Details:

The labor market put in a strong showing in March, as wintry weather receded, strikes concluded and businesses started looking beyond the significant uncertainty of President Trump’s first year in office.

[We had certainty under the capable steady hands of Joe Biden and Kamala Harris. Now we have frightening uncertainty]

  • Health care dominance: Even factoring in the addition of 31,000 jobs from workers ending a strike in California, the sector continued to lead gains, adding 76,000 positions. Manufacturing, which has been trending down for three years, added 15,000 jobs and construction grew by 26,000.
  • Federal government still contracting: The federal government shed another 18,000 jobs in March and is down a total of 355,000 positions, or 11.8 percent, since reaching a peak in October 2024.

Separately, we’ve been told that Donald Trump’s “without any plan” war against Iran would destroy both the world economy and the U.S. economy. Do investors agree with the wise prophets at the New York Times and CNN? Compared to the no-war situation a year ago, U.S. stocks are up 22% in nominal terms (19% in real dollars if we use official BLS CPI):

Full post, including comments

False Dawn: FDR and the Great Depression

by philg

False Dawn: The New Deal and the Promise of Recovery, 1933–1947 is an academic economist’s look at all of the academic economists’ explanations for the Great Depression and how, with muscular government intervention led by a now-revered FDR, it could have lasted so much longer than economic downturns that happened in the U.S.’s free market period (1630-1930). (I introduced this book a few months ago with Philip’s Book Club: False Dawn.)

One popular theory is that World War II brought the U.S. out of the Depression after about a decade of failed federal government interventions. The book points out that calculating GNP during World War II is somewhat arbitrary:

But there is a deeper sense in which the wartime recovery, however and whenever it started, was no recovery at all. “In the crucial respect of waste of economic resources,” Broadus Mitchell (1947, 396) observes, “the war was, particularly for the United States, a deepening of the depression.” Tens of millions who had been either unemployed or employed in peacetime activities now took part in activities that, however crucial, continued to reduce instead of enhancing their own and the world’s living standards. To label such a state of affairs “full employment” was, Mitchell says, but “a flattering unction”: the employment thus generated was “for purposes which, by very definition, could have no place in a normal economy.” In short, the war was but a temporary solution to the problem of economic depression, and the more temporary the better. The point may seem trite. But it’s a necessary response to those—and there are many—who declare that World War II ended the Depression and just leave it at that.

… the most serious shortcoming of wartime output measures, namely, their tendency to overstate, perhaps dramatically, both the nominal value of war matériel and the extent to which it should be considered part of national output at all. As Higgs (1992, 45–47) reports, Kuznets, whose wartime and postwar deflators are among those that have been called into question by Friedman and Schwartz and others, had his own grave misgivings about the Commerce Department’s valuation of wartime output. “A major war,” he observed, “magnifies” the usual challenges involved in estimating national income, because war matériel isn’t sold at anything resembling “market” prices and also because wars blur “the distinction between intermediate and final products” (45).

Such considerations persuaded Kuznets to come up with several alternative measures of wartime and postwar GNP, all of which imply a less impressive wartime boom, or no boom at all, and no postwar slump. For example, according to Higgs (1992, 46), “whereas the Commerce Department’s latest estimate of real GNP drops precipitously in 1946 and remains at that low level for the rest of the decade,” Kuznets’s “wartime” estimate “increases in 1946 by about 8 percent, then rises slightly higher during the next three years.” Another Kuznets GNP estimate—what he called “peacetime” GNP—revises the record still further by valuing goods produced for military use at their nonmilitary surplus values only. According to that estimate, between 1945 and 1947 real output rose by almost 18 percent!

The above quote illustrates the principal flaw of the book for a non-academic reader. The author spends most of his ink summarizing and referencing the work of other academics. He’ll lay out four or five theories and deal with each one in turn, forcing the reader to tease out the author’s personal point of view. The book is more accurately characterized as a survey of 100 years of academic thought regarding the Great Depression than as an explanation of the Great Depression and subsequent recovery.

Another popular theory is that the New Deal was an example of Macro Economics 101 Keynesian deficit spending during a downturn. Two chapters are devoted to “The Keynesian Myth”. The author points out that Roosevelt was committed to a balanced budget and that Keynes himself wrote critically of American economic policy. The New Deal was a project to increase the power of the federal government in regulating business, not a deficit spending plan. Some of this project was abandoned when the federal government needed private industry to be its partner for World War II weapons production, but the effect was to stifle business investment and reduce personal consumption.

“Conventional wisdom has it,” Gary Best (1991, 222) observes, “that the massive government spending of World War II finally brought a Keynesian recovery from the depression.” However, Best continues, the fact that the government was no longer at war with business, as it had been during the original New Deal, deserves more credit. “That,” Best says, “and not the emphasis on spending alone, is the lesson that needs to be learned.”

On the extent to which there was any Keynesian borrowing:

New Deal deficits were less impressive than New Deal spending because the Roosevelt administration went to considerable lengths to boost tax revenues and did so even when it meant relying heavily on taxes that mostly burdened low-income Americans. For that reason, the administration chose not only to retain and then repeatedly extend most of the excise taxes imposed as part of Hoover’s 1932 budget—taxes Herbert Stein (1966, 210) considers “the purest act of pre-revolutionary fiscal policy”—but also to increase taxes on gasoline and tobacco, revive the liquor tax upon the repeal of Prohibition, and introduce its AAA tax on food processors. Because they fell on consumers, either directly or indirectly, excise taxes, which eventually funded 60 percent of the government’s “ordinary” revenues (Leff 1984, 147), tended to be deflationary even when fully offset by government spending. Such taxes therefore had little to recommend them from a countercyclical fiscal policy perspective (Brown 1956, 868; Leff 1984, 39). But because excise taxes were revenue workhorses, to an administration not much less determined to limit deficits than Hoover’s had been, they made perfect sense. At the height of the New Deal, Mark Leff (1984, 38) points out, the tax on food processors alone “accounted for one-eighth of total tax revenues,” which was more than the yield from either the personal income tax or the corporate income tax. For this reason, after the tax on food processors was struck down, Roosevelt “continued to suggest processing taxes to balance the budget and to fund farm subsidies” (Leff 1984, 44). What was true of excise taxes was truer still of the Social Security payroll tax that the government began collecting in January 1937. According to Leff (1984, 45), when Roosevelt first came up with his plan for funding Social Security, his advisers warned that because it would draw purchasing power from consumers for the purpose of establishing a $47 billion reserve fund without making any like disbursements from that fund for many years, the plan would be dangerously deflationary. Still, Roosevelt insisted on it, saying that it would assist in balancing the budget while projecting “an image of fiscal responsibility” (47). According to Sherwood Fine (1944, 114), the regressive Social Security tax diverted “more than a billion dollars of purchasing power . . . away from an industrial establishment sensitively attuned to consumer demand” in the midst of a severe economic downturn. “Running along, as we are, on a low level,” Alvin Hansen (1939b, 283) wrote afterward, when various amendments to the Social Security Act were under consideration, “we cannot afford . . . the luxury of a Social Security Program which turns out in effect to be essentially a compulsory savings program, and which thereby seriously curtails the volume of consumption expenditures.”

France was the only significant (imagine that there was a time when France was significant!) foreign nation inspired to follow our example:

France was one of the few countries and the only major one to take longer to recover from the Great Depression than the United States. France was also the only country that resorted to policies closely resembling—indeed inspired by—the New Deal, including NRA-style codes. And it was the only country that did not experience a substantial improvement in output after devaluing its currency. As Barry Eichengreen (1992a, 349) observes, France’s example shows, even more clearly than that of the United States, that “devaluation was necessary but not sufficient for economic recovery.” France’s first stab at New Deal–style industrial planning consisted of the so-called Flandrin experiment, an attempt by the conservative ministry of Pierre-Etienne Flandrin (November 1934–May 1935), directly inspired by the NRA, to cartelize French industries and reduce workers’ working hours. Flandrin’s experiment went no further, and his government fell after six months. But several weeks after decisively winning France’s May 3, 1936, parliamentary election, the Popular Front—an alliance of French radicals, socialists, and communists—implemented an NRA-inspired plan of its own. That plan was so aggressive that recent scholars have dubbed it “a sort of NIRA on steroids” (Cohen-Setton, Hausman, and Wieland 2017). As Barry Eichengreen (1992a, 375–76) explains, “Employers were compelled to sign the Accord de Matignon granting trade union recognition, collective bargaining privileges, and wage increases. . . . [T]he work week was shortened again, but this time without any corresponding reduction in pay. The government legislated an annual paid vacation and a 40-hour week. Wages were raised by 7 percent for high-paid workers and by up to 15 percent for the lower paid. . . . Other elements of the French ‘New Deal’ raised the school-leaving age and nationalized the armaments industry.” The Matignon Agreements’ mandatory wage rate increases went into effect at once, raising nominal labor costs by between 7 and 15 percent (Cohen-Setton, Hausman, and Wieland 2017, 279). The rest of the Popular Front’s plan, including its forty-hour week provision, was phased in industry by industry between then and the end of the year.

Noting how after 1936 France’s industrial output was persistently 30 percent below its long-run trend, Paul Beaudry and Franck Portier (2002) consider various possible explanations, including technological stagnation, only to conclude that the best is the simplest: French output fell 30 percent because between them the Blum government’s labor market legislation and strikes caused total hours worked to fall 25 percent. A decline in the ratio of investment to output, itself traceable to France’s New Deal legislation, accounts for the remaining five percentage points by which output fell.

What did cause the early-1930s collapse, then? The author points to an agriculture boom during World War I that led inevitably to a bust after WWI that dragged down most of America’s banks.

Bank lending to farmers itself doubled between the start of the war and 1920. After the war, both crop prices and US farm exports fell as sharply as they’d risen during it, triggering a farm crisis that was to ruin many farmers over the course of the next decade, often bringing their banks down with them (Belongia and Gilbert 1985). In 1921 alone more than 500 banks failed, topping the previous record established during the Panic of 1893. The 1921 failures coincided with the general economic depression of that year. But while most other industries recovered quickly from the downturn, and did so with little help from either the Federal Reserve or the Treasury, agriculture and banking didn’t. Instead, bank failures mounted.

Although thousands of US banks managed to survive the 1920s, many were in no condition to withstand any further shocks. So when commodity and security prices sagged after the onset of the Depression, bank failures became even more frequent. Rural banks were still the main casualties, although now instead of being concentrated in the western grain-growing states, bank failures were especially frequent in the South and the Midwest, where collapsing cotton, tobacco, and livestock prices combined with reduced cotton and wheat yields—a result of what the Weather Bureau described as “the most severe drought in the climatological history of the United States”—proved to be the last straw (Hamilton 1985, 602).

How much can we rely on expert analysis and wisdom? The author points out that all of the best minds of economics were in agreement that there would be a depression following World War II due to soldiers returning and finding themselves unemployed and the government ceasing purchases of weapons. Business

Full post, including comments

AI economy terminates as petrostate?

by philg

A friend owns a three-unit building in San Francisco (rents carefully controlled by qualified government officials!) and is concerned about the value of his place if AI eliminates most programming jobs. My response was to consider the extreme case that 100 percent of GDP is generated by AI and robots. In that case, the economy becomes like a petrostate, e.g., Kuwait. The efforts of human residents aren’t significant economically compared to oil flowing or NVIDIA chips cogitating. In petrostates, however, the rulers don’t expel most or all of the citizens (like Bhutan did!), but instead use whatever money isn’t stolen by elites to house and feed everyone. Thus, in the typical petrostate, real estate still has substantial value. “Maybe everyone in San Francisco will be on Section 8 and the rent will be paid by the government instead of individuals,” I said, “but you’ll still get rent.”

How will the government get revenue? Petrostates often nationalize the oil industry, as Venezuela did in 1976 and 2007 (Hugo Chavez for the win!). If most of the wealth and income of the U.S. ends up in the hands of the owners of NVIDIA, Anthropic, et al., the government can simply nationalize the top 20 most successful AI-related companies. (We can see a half measure of this right now with Bernie Sanders and Ro Khanna proposing to harvest 5 percent of billionaire wealth every year.)

In other words, it doesn’t make sense to be an AI Doomer on economic grounds because being a citizen of a typical petrostate isn’t terrible (let’s ignore Venezuela for the moment!).

Let’s check in on a petrostate that has been shooting down U.S. fighter jets recently. Kuwait is rich, though not quite as rich per capita as it once was. It looks like they’ve grown the denominator via population growth and, thus, each individual’s share of the oil income has been reduced.

(Note that U.S. politicians, beginning with Lyndon Johnson in the mid-1960s, have been working desperately to grow U.S. population via immigration, exactly the opposite of what makes sense if our destiny is that most wealth comes from something that functions like an oil well.)

Full post, including comments

U.S. economy defies Science

by philg

We’ve been informed that low-skill migrants, as a matter of Scientific fact, are positively correlated with U.S. economic growth (at least aggregate growth if not per-capita!). Low-skill migrants have been departing the U.S. at an unprecedented rate since the Trump Dictatorship v2.0 began (CIS; NYT (covers a different time period than the CIS analysis)).

Toda we learn from the Wall Street Journal that the aggregate GDP is expanding even as the migrant population shrinks.

Maybe the GDP numbers are wrong? We can see for ourselves that valuable Somalis and Latinx are being kidnapped by ICE (should we try to fight ICElamophobia?). We know that these folks are worth $billions even though there is not another country on Planet Earth that is willing to take the migrants whom we deport (i.e., no other country wants to be enriched as we have been). If the GDP data are correct, could the apparent contradiction be explained by The Science being merely a projection of researchers’ love for migrants? “Why immigration research is probably biased” (Guenther, December 20, 2025):

All of these choices resulted in 1,261 submitted models; no two were identical. Notably, this heterogeneity arose even though the hypothesis and data were the same! Think how much freedom researchers have when they are allowed to choose the hypothesis and the data.

It is not necessarily problematic if researchers are more liberal than the general public, but it is problematic if these attitudes make them analyse data in a biased way, to arrive at conclusions that reinforce their prior attitudes. In that case, immigration research ceases to be research and transitions into propaganda, where only hypotheses are tested that one can anticipate to portray immigration positively, and the research design is chosen to obtain the desired conclusion.

Related:

Full post, including comments

Philip’s Book Club: False Dawn

by philg

Maybe some of you will join me in reading False Dawn: The Mirage of Recovery, an economist’s book about the Great Depression, which is when Americans came to accept the idea that every problem should be met by a larger federal government. FDR is almost a god for today’s Democrats (in a debate Ayatollah Mamdani identified FDR as the best modern-day U.S. President and then Florida Realtor of the Year 2020/2021 Andrew Cuomo said FDR would be his pick as well if FDR could be considered “modern”). If nothing else, False Dawn would make an awesome last-minute Christmas gift for anyone with insomnia (384-page work by an economist).

The Wall Street Journal selected this book as one of 2025’s ten best. Some excerpts from their review:

In 1932 Franklin Delano Roosevelt won the presidency with the promise to restore prosperity. But he and his advisers had no clear explanation for the collapse and his subsequent New Deal would amount to a series of experiments. FDR admitted to the nation that some of his proposals took the nation down “a new and untrod path.” If they failed to “produce the hoped-for results, I shall be the first to acknowledge it.”

George Selgin’s “False Dawn” asks if the New Deal’s varied experiments produced the promised recovery. In dispassionate, careful and finally devastating detail, “False Dawn” shows that, with a few exceptions, FDR’s experiments did not work. And he did not acknowledge it.

Based simply on raw numbers, the case for the New Deal is not strong. Although the economy did recover from its nadir when FDR took office in 1933, by 1939 the unemployment rate was still 17%. After six years of supposed recovery, the economy was in worse shape than in any other recession of that century or the following one.

Some might suppose that FDR used deficits rather than the Fed to juice the economy. But deficits as a percent of the economy were hardly different during Roosevelt’s time in office than they had been at the end of Herbert Hoover’s. While the New Deal spent more, it also imposed new taxes on food and payrolls. The result was a bigger federal government, but not one that relied on deficits as stimulus.

If not by increasing the amount of money or deficit spending, how did Roosevelt and his advisors hope to create recovery? The earliest solution they hit on—odd considering the rampant shortages—was to restrict production and thus raise prices. The National Industrial Recovery Act that passed in mid-1933 turned much of the American economy over to giant cartels. Industries colluded to raise prices and unions colluded to raise wages. The result was fewer goods on the market and an immediate economic collapse that would still be remembered today if it hadn’t been surrounded by so many others.

This could be an interesting update to The Forgotten Man, by Amity Shlaes, a Wall Street Journal reporter. I wrote a lot about that book shortly after its 2007 publication (what awesome timing by Schlaes and her publisher, given that the U.S. economy collapsed just a year later):

Full post, including comments