What I have been reading: October 2025 update
Helsinki has gone one whole year without a traffic fatality, due to a combination of low speed limits, the focus on public transportation, and technological improvements like E-scooters. I would suggest this as a case for the marginal inconvenience of 20 mph zones, yet I wonder to what extent driverless cars will change the calculation? We have all seen the data published by Waymo, and whilst you may think “of course a profit-seeking company will publish results good for its bottom line”, there is good reason to believe they are on the modest end. In particular, as driverless cars become popular, drink-driving fatalities will plummet.
AI seems to be changing how we communicate in everyday life. I admit, I’m using the phrase “bottom line” a lot more often.
Séb Krier on how he imagines AGI will arrive in practice. Wait and see is my approach: there is a high possibility that our current conceptions of future AI will be far different to ex-post reality. Just consider the predictions of 20th century science fiction vs today, or how Victorians imagined the future whilst the Industrial Revolution was taking off. Of course, this should inform our debates on AI alignment and existential risk. The implications: attach a wide confidence interval to any p(doom) value that you assign.
The gender divide: movies edition. It has been noted that where there is the greatest disparities in average rankings, women tend to select more recent films, and males the older films. I have more to say regarding how patterns of feminisation have influenced the wider culture (wars), yet this is a future post in itself. By the way, as a man, I should say that the Harry Potter series (mostly the fault of the books) is overrated; did the franchise really need to stretch to seven books? Conciseness is a virtue, including in fiction.
Personalised reading technology has arrived. Hopefully, this will make reading much more fun and rewarding. Let's be honest: not many enjoy reading for its own sake. Even I, a relatively voracious reader (certainly in the top 10%), skim read much of the content I come across. This ties in well with my primary criticism of the Substack ecosphere - the equilibrium has converged around essays, yet how many want (or have the time and attention) to consume the full piece? This informs my writing too. I blog with the assumption that hardly anyone will read the full post, and at best will skim read, start then lose attention, or read only the parts that interest them. Moral of the story is to keep your blogs short! Essays are an elaborate academic signalling device. Most good arguments only take a few pages, and if they require an essay, then the number of premises taken to reach the conclusion are such that at least one is highly likely to be uncertain.
In perhaps one of the most modest tweets I have ever encountered, entire sections of the human genome can now be edited. Of all avenues for progress in my lifetime, the possibilities for human enhancement are up there with what excites me most. Transhumanism has arrived! Biotechnology will radically save and change millions of lives for the better.
The long-standing minimum wage debate has reignited again. Yes, I have linked to my own response as a subtle self-promotion, yet read the entire conversation. This is one of those issues where there are strong, empirically-driven arguments on both sides. This fusion of theory and empirics is where economics is at its strongest. I should add that many of the studies purporting no changes to employment (such as the famous Card and Krueger study) consider number of jobs only. In the standard competitive labour market model (anti minimum-wage), L is the x-intercept.
Eigenvectors in economic theory. Strictly speaking not “exotic” (every economics student is taught them), yet the use of them within the field is scarce.
Although undoubtedly mass state subsidies into infrastructure are misallocating resources (roads to nowhere anyone?), one must at least appreciate the aesthetics of this impressive engineering feat. For reference, this is twice as high as the Eiffel tower.
A proposal for dealing with infinite values in expected utility theory. Also see the quote tweets for some objections.
Math in economics is, contrary to the popular view, underrated. A convenient tool to screen out bullshit.
The impending bailout of Argentina is not a failure of libertarian ideas. If anything, Milei was too timid. He should have moved faster towards a floating rate.
Want to make easy money? Sell your privacy today. I don't think I’ll be engaging in this transaction, but each to their own.
On partisanship and mental health. The results are endogenous to measurement. I must admit, I was skeptical of the claims that liberals (particularly young liberal females) are suffering more than anyone else, or that conservatives are uniquely thriving for that matter. Wasn't an entire academic research agenda dedicated to establishing some link between right-wing populism vs the “left-behinds” after all? In general, it's difficult to encounter objective measurements of affect. Whether someone suffers from depression or anxiety depends to a large extent on how one defines it. If my mother dies, and I'm melancholy for months, is that depression (which implies some mental disorder), or just a natural response to tragedy? If I'm shy and socially awkward, is that GAD, or just introversion?
Republicans are twice as likely to trust podcasts over other media than Democrats, yet this ecosystem is highly dominated by the likes of Tucker, Candace, and Rogan. As a right-winger (libertarian, not conservative) myself, most right-wing influencers should not be trusted to hold open the door, let alone be relied upon as a source of information! X aside, I tend to stick to centrist or centre-left leaning outlets (the FT, BBC, and The Economist). This is perhaps, aside from ideological bias, why I may be vulnerable to excessively focusing on liberals’ flaws, despite the right being epistemologically worse. I consume a lot more left-leaning than right-leaning news sources.
How many times do we have this discussion? History seems to repeat itself at least once a year. I remember dealing with these sorts of criticisms of the economics profession back before the pandemic!
A nice summary of the economic benefits of migration. A stark reminder of what is at stake should Reform gain power.
A reason for Latin America’s underperformance relative to the rest of the West?
No. You cannot state with absolute certainty that interest rates hikes will necessarily reduce inflation.
In part, yes, but I think the fertility crisis is a lot more complicated than that. Feminisation has to interact with the total opportunity costs of childbirth. Yes, this includes the motherhood penalty, but also high housing costs, young unemployment and nonparticipation in the labour-force due to education pressures (it was young adulthood when individuals a few decades ago were starting families after all), and so on. As this fertility decline occurs in red-states, with relatively conservative societies (such as Italy or East Asia) seeing some of the greatest reported declines in TFR, should we really expect that a return to the traditional “stay at home mothers” will improve matters? The causes are poorly understood by everyone, and we must be honest about that.
Will the changes to H1-B harm the GOP’s electoral prospects?
Should economists consider political constraints when formulating their policy prescriptions? Read the entire exchange. I agree with Cochrane. Without integrity, what use is contributing to our stock of knowledge and understanding?
“The FCC under Presidents Kennedy and Johnson used the Fairness Doctrine to target right-wing radio hosts in the 1960s.” Antitrust is bad for a free society. As a general rule, all regulatory bodies can be weaponised for political purposes. Get rid of them all and return regulation back to our democratically-elected legislatures. Hayek warned about precisely this in his famous book…
Excessive credentialism leads universities to reject child prodigies. Is the university model of academia really the best we can hope for with respect to innovation or scientific progress? Of course, bureaucratisation and extensive government subsidies lock in this intransigence.
Investor disagreement correlates with abnormal trading volume and average future returns. Risk should be correlated with disagreement, risk is correlated with return, so no surprises. The main contribution of this paper, in my view, is to demonstrate how AI can enhance our research.
This paper finds no evidence of harmful algorithmic collusion. If anything, algorithmic pricing seems to benefit consumers.
"R&D has become more effective", but "may also render rivals’ technologies obsolete, making innovations more transient. Because of obsolescence, rising R&D does not necessarily mean rising aggregate productivity growth."
Marx wrote his works not long after economic growth actually became a real thing. Previously, Malthusian conditions were the norm, and may well have continued to be the case if not for capitalism.
The latest statistics on cancer survival. This is one of the best stories of this century so far.
Labour and the Brahmin left, again.
FDA submissions are private, which "gives a structural advantage to large pharmaceutical incumbents who have amassed regulatory filing archives over small biotechs".
An exchange between Brooks and Setser. Luckily inflation and deficits are relatively low, so Japan can stomach a depreciation (although far from an ideal situation). In addition, much of GPIF's portfolio is denominated in USD, so a Yen depreciation is not necessarily bad for Japan's fiscal position.
NIH grantmaking has been fully restored. Conditional on having institutional setups that find science in the first place (which I have argued against), this is a good thing.
One step closer to simulating economies. A potential revolution in economic theory and forecasting may be underway (how easy it would be to attract funding for artificial AI economies is a different story altogether though).
Of course, right-wingers have faced discrimination in academia (e.g. via the use of DEI statements for hiring). Yet is discrimination the primary cause of these results? Or is this mainly the result of assortative matching given education polarisation?
Sex reversal, where outward appearance does not match genetic sex as determined by chromosomes, in birds. "It's the genes carried on those chromosomes—not the chromosomes—that are the main players. The SRY gene on the Y chromosome, for example, kick-starts male development in mammals. Anyone missing this key gene will end up developing as female". In some species, this can give rise to gynandromorphs - a fusion of the male and the female.
Reduced access to abortion “leads to significant economic hardship, reflected in lower labor force participation, rising debt, widening income inequality, and heightened housing insecurity. This financial strain translates into higher rates of financially motivated crime, such as theft and burglary, with no significant effect on violent crime.” The right to choose is fundamental to a free society.
A model of ESG. Catering to the political preferences of employees increases profits but reduces wages.
The best evidence yet of former life on Mars. On a related subject, I do believe in the existence of aliens (defined as intelligent life). A universe infinite in space makes such a near certainty. Far more controversial: do aliens exist within a light-year radius that could make contact with them ever possible?
No surge in politically motivated killings, which thankfully continue to remain at pretty much a rounding error from zero. Given this, any claim of whether the left or right engages in more of these acts will be incredibly unrobust. You get evil and deranged individuals in all walks of life.
A thread on how Islam has influenced European art. Why the hell do we want to close the door to Muslims? Yes, most of the complaints regarding asylum seekers are refugees from the Islamic world. Islamophobia is sadly becoming a pervasive force in European politics, even in relatively cosmopolitan Britain.