You know what we don't like? Bad social science.

And, sadly, the Harvard Business Review has just given us an almost canonical example of it. "When Populists Rise, Economies Usually Fall," the article, written by Roberto Stefan Foa and Rachel Kleinfeld, blares.

The article, which is clearly intended for a business audience, is meant to explain that when populists win elections, the business environment gets worse. More specifically, the article is particularly keen to explain that right-wing populists can be just as bad for business as left-wing populists.

And the way it goes about "demonstrating" this is, frankly, laughable, in a way that we would think serious people could no longer tolerate these days. And certainly, we wish that people who produce work of this nature should no longer find employment at any institution that claims to be engaged in something of a scientific, or just serious, nature.

What is the paper's methodology?

Here it is: "To account for the forward-looking nature of market pricing, we collected data starting six months before a new government was elected and ending either 1) six months before an election loss or 2) through June 2024 (for those who remained in office). To ensure fair comparison, we tracked relative rather than absolute returns — that is, how a country index performed vis-a-vis equivalent markets over an identical period of time, using either a global index for developed economies or a specific index for emerging markets as appropriate."

That's it.

That's the paper.

When populists are elected, the stock market doesn't do as well relative to some peer groups.

The first and obvious thing to note is that public equity markets hardly reflect either the entire business community or the economy of a country.

The second, obvious thing to note is that populism is a response to elite failure. Populists arise (let alone win elections) when things have become so bad that large swathes, even majorities, of the electorate are willing to bet on some strange outsider who says strange and radical things, which doesn't happen when times are good. Therefore, even if the article's thesis was true, you would expect a selection effect where mismanagement by elites causes both bad economic numbers and the rise of populists.

An obvious contemporary example is Argentina's Javier Milei, who definitely qualifies as a "populist". Milei, a firebrand corporate economist and TV pundit with an awkward demeanor and extreme anarcho-capitalist views, would never have had a chance of being elected if Argentina hadn't been in the midst of one of its worst recession and hyperinflation crises in decades. The jury is still out on Milei's performance, but even if he pursued the best economic policies imaginable (whatever we might think those might be) we would still expect the economy to do poorly for at least several years into his term.

This is the classic problem of correlational studies (although calling this study correlational is generous), which is that there is always the possibility of a selection effect that makes it impossible to prove the cause-and-effect relationship you're trying to prove.

But what gets our spidey sense truly tingling is the article's arbitrary criteria. The study's dataset includes data "starting six months before a new government" and ending "six months before an election loss" (except for those who remained in office). Why six months before? "To account for the forward-looking nature of market pricing." Ah. Of course, you might argue that you might start the clock six months after the government takes office and after it loses reelection, to take time for the new government to get set up and implement new policies. And why six months, for that matter? Why not four? Or eight, or nine, or twelve? Obviously there is no "scientific" way to answer that question, even though picking any of these numbers might produce a very different result.

Also, the relevant comparison for a developed country is "a global index for developed economies." Because there is absolutely nothing, nothing at all, that might account for how the British (say) stockmarket does and how a global index of all developed economies behave, than the policies of the British government. Why not a regional index? Why not a synthetic index of countries with similar economic and/or demographic profiles? The answer, it seems to be, is: ¯\_(ツ)_/¯

It's pretty clear that what we are dealing with, here, is not anything remotely representing "science", but haruspicy.

At best.

One possibility for why the authors went with this arbitrary set of criteria rather than any other would be because they ran the numbers on several of them and picked the one that gives the result they wanted. The technical term for this is, well, fraud. To be clear, we have no way of knowing if that is what is going on in this case, and we are not alleging it. Of course, the fact that this type of fraud is practically impossible to detect is a good reason to engage in it. In one survey, over half of academics surveyed admitted to engaging in this type of behavior.

Regardless, everyone involved with producing this study, bringing it into the world, and publicizing it, should be ashamed.

The authors seem to be particularly obsessed with Viktor Orban's Hungary, making much of a single example of a media company whose owner was allegedly pressured into selling to political allies of Orban.

Apparently stock market returns haven't been great under Orban (we take the authors' word for it). Does that mean that Hungary is a bad place to do business in?

Maybe, maybe not. Perhaps it's relevant that growth in Hungary has averaged around 5% every year in the past ten years (excepting 2020); that the country has one of the lowest unemployment rates in Europe; a 7% trade surplus; the lowest corporate income tax rate in the EU at 9% (!) and 13% payroll taxes…

Maybe there are other reasons why Hungary is a bad country to do business. Maybe it is as corrupt and has as little rule of law as Viktor Orban's political enemies allege. The point is that simply looking at stock market returns vs a "global index for developed economies" to ascertain whether a country is a good place to do business or whether its government has made it a worse place to do business is laughable.

Do better, folks. Do better.

Policy News You Need To Know

#AI #Transportation — The biggest news this week, on any front, is probably Tesla's announcement of Robotaxi, its system of driverless taxis. There is a car, which everyone expected, and also a gorgeous, art deco-looking bus. Making self-driving cars a reality is the first kind of technology that might have the kind of economic and social impact commensurate with the car itself, which ended up creating suburban culture and everything downstream from that. Autonomous driving would be much more efficient, saving millions of lives from accidents, cutting commutes, making cities greener and more breathable, and so much more.

#AI — Speaking of, over 30% of U.S. workers could see at least 50% of their occupation’s tasks disrupted by generative AI, while some 85% of workers could see at least 10% of their work tasks impacted, according to a new Brookings study.

#AI — Speaking of, the DOJ is suing RealPage, an AI startup that helps landlords set rents. The Competitive Enterprise Institute has a good overview of the lawsuit and, especially, its implications for antitrust policy. The idea, which sounds spurious, is of "algorithmic price collusion." The way these startups work is that they use AI and algorithms to help landlords and other companies set the profit-maximizing price. The idea is that if enough companies use this software, they will be rising prices in lockstep, so that, even without any collusion, they would de facto act as a price-fixing cartel. The problem with this idea is that, of course, AI doesn't supersede the law of supply and demand. If the algorithm gets all companies to set a profit-maximizing price that is too high, some competitor will come along and lower the price. The reason why price-fixing cartels are bad is not because they increase prices as such, but because they prevent new entrants from undermining the too-high price and letting the market find the right price. There doesn't seem to be any reason to believe that AI does that.

That being said, there is another problem here, which is that, as we wrote previously, dynamic pricing (as this is called) can invite a political backlash nevertheless. Here's what we wrote when we first wrote about this: "The idea behind dynamic pricing is simple. Let's say you sell a widget for $10. Let's say Alice, your customer, would actually be willing to buy your widget for $15. If you sell it for $10, you're leaving money on the table. But selling for $15 may not be a viable solution, because Bob, who buys your widgets for $10, is not willing to pay more and would stop buying if you raised the price. Suppose corporations were able to charge each individual customer the highest price he is willing to pay? By definition, they would be maximizing profits to the utmost. Businessmen have been trying to solve this problem for generations. It is called capturing the customer surplus. Airlines have gotten famously good at it, with algorithms setting prices that change constantly depending on time of day, class, and so on. Another famous example is Microsoft, who will sell (say) “Home,” “Small Business,” and “Enterprise” versions of essentially the same software in order to capture that customer surplus. Many corporations (and many consultants serving those corporations) believe AI could help them finally and completely crack the customer surplus problem. But one problem with that idea, as the airlines example shows, is that customers hate it. It offends their basic sense of fairness. Everyone knows the airline charges a slightly different price to everyone, and it makes us hate the airline because we never know when we buy a ticket whether we've been gouged. […] The point is this: dynamic pricing offends people's basic sense of fairness, and if corporations go too far with it, they will face a political backlash. Caveat venditor..."

#FreeSpeech #Media — You may be aware that the famous writer Ta-Nehisi Coates has written a book which is, in part, an avowedly one-sided anti-Israel screed. When Coates went on CBS to promote this book, the journalist Tony Dokoupil respectfully challenged some of the claims in the book (and, to be clear, Coates acquitted himself well in his answers). Since then, he has been the subject of an internal Maoist struggle session at CBS. Smart people should no longer trust corporate media, that's for sure. Puck News reports that from now on journalists at CBS must have their questions vetted by racial commissars.

SEE ALSO: Our analysis of the reception of Coates's book and why it shows wokeness is hardly dead →

#VotingRights — An investigation by Maine Wire shows votes cast under names of non-citizens in multiple elections in Maine since 2016. Given its size, Maine gets less attention than places like Pennsylvania and Arizona, but it is a swing, or at least purple, state.

#Trade — AEI's Michael Strain is one of the smartest economic analysts out there. He recently wrote a paper for the Aspen Economic Strategy Group looking at trade and protectionism. One thing that Strain, a free trade enthusiast, notes is that the US manufacturing employment declined steadily both before and after the famous "China shock."

#Trade — Meanwhile, American Compass's Mark A. DiPlacido is at Newsweek calling for an end to PNTR status for China.

#Immigration — Pennsylvania school districts have seen a 40% increase in the number of non-English-speaking students since 2021, Megan Brock and Jason Hopkins of the Daily Caller News Foundation report.

#Veterans — RAND produces excellent work on veteran's issues, and this new study on America's military and veteran caregivers is no exception. The study notably finds that among caregivers to service members and veterans 60 and under, 42% met criteria for depression, 36% wanted mental health treatment in the past year but didn't receive it, 70% had trouble paying bills, and 40% met criteria for food insecurity.

#TaxPolicy — From ATR: Walz Endorses Individual Mandate Tax, a Violation of Kamala’s $400k Tax Pledge

Friday Essays

If you have been a Catholic interested in theology on the internet in the past 10 years, you probably ran across Greg Hillis, most recently director of the Aquinas Center at Emory University, his work, and his sunny personality. Your correspondent only seldom interacted with Hillis but he remembers his intelligence and gentleness. For several months, friends and acquaintances have known that Hillis had been diagnosed with stage IV cholangiocarcinoma, a rare form of cancer, which deteriorated quickly and left him with a diagnosis measured in months. Hillis went to his eternal reward last week. He leaves behind a wife and three young boys. For the Catholic magazine The Lamp, the author Thomas Casey has written a beautiful, profound remembrance. "I can only describe Greg’s peaceful demeanor as he confronted the end of his life as heroic. He was never bitter. He was never angry or self-pitying." We wish we had known him better. May he rest in peace.

New York Times writer Shawn McCreesh spent several days in Ohio and Kentucky with JD Vance's mother, Beverly Aikins, and wrote a beautiful profile. Ms Aikins' history of struggling with addiction is, of course, a big part of Vance's personal story that became the book, which became the film, which became the springboard for one of the most astonishing political rises in recent memory. Ms Aikins has been clean and sober for almost ten years and works in the field of addiction recovery. The essay is a beautiful look at a mother who clearly very deeply loves and admires her son. She now uses her accidental fame to help others who struggle with addiction. It's just a beautiful and moving story and, if you need a policy angle, it will make you even more angry and determined to attack the horrible addiction crisis affecting America.

Chris Rufo and Christina Buttons have conducted an in-depth investigation into precisely how a nexus of government-funded NGOs has been flooding struggling American towns like Charleroi, PA, with Haitian migrants, seemingly from nowhere and seemingly overnight. It's very important journalism. Connected to this, you should read Mike Needham's essay for American Compass on the crisis in Charleroi, about how elite self-dealing leads directly to crises like this.

Santi Ruiz, who writes the excellent newsletter Statecraft, has written a deep dive essay on How to Fix Defense Procurement. It's self-recommending if you have any interest in these types of issues.