Mid-December 2025 links
Is Instacart price discriminating? If so, is this necessarily bad for consumers?
Agreed on this point, yet gun control does indeed seem to work (from ChatGPT 5; I posed this question a few months ago). An underrated complication in social science is simultaneous interaction effects: both statements are not mutually exclusive. Gun control will reduce homicides, yet locking up sufficiently more violent recidivists for sufficiently longer will achieve an equivalent effect in the absence of gun control. Note that gun control tends to be conflated with a blanket ban on ownership, which is unnecessary; the British approach is excessively restrictive, yet the open-carry laws of red states are ludicrous (even from a libertarian perspective, private ownership of land means that landowners can refuse open carry of guns on their territory). Tort law is an underrated remedy. Also (though this may just be the Brit in me speaking) I fail to understand why the second amendment is necessary to prevent tyranny. Other democratic nations lack this provision, guns will not solve the free-speech crisis in Britain (if you care about some slippery slope), and in any case violent revolution is almost never justified. I back a loosening of gun ownership laws here simply for the benefit of benign hobbyists - that argument is enough for anyone concerned with individual liberty.
Government becoming a larger proportion of UK GDP is likely a major factor in our stagnation. This trend even held up throughout the austerity years.
The age of transhumanism, what Emil Kirkegaard refers to as the biosingularity?
As I've said, AGI is already here. The only obstacles to its widespread deployment are the challenges of incorporating this into production (and consumption!) processes on the human end. Most people don't have a clue about what is already here, and what is about to hit them!
National conservatism, Groyperism, postliberalism: they're all mirror images of DEI, for white males.
Gibbard-Satterthwaite: college admissions edition. Also applies to disability accomodations.
So a deteriorating labour market in Britain is not just my anecdotal observation?
My fear of dogs seems to not be an irrational phobia after all…
One lesson we can draw is that even with rare variants, there is still a large gap between the estimates of twin vs molecular studies. People are still unsure whether twin studies inflate the estimates. This debate is becoming more confusing each day!
Surge pricing means that quantity supplied is greater, and shortages (in this case Uber delays) are eradicated. The most basic of supply and demand!
An erosion of free speech imposes far more (tangible) costs than simply silencing racists or whatever.
Via twin studies, we can derive causal estimates of the socioeconomic effects of cannabis use on the individual. No clear effect on alcohol abuse (on a macro scale in America, where most states have legalised the plant, it seems to be gradually replacing alcohol as society's drug of choice) or criminality, yet arguably confirmation of the gateway hypothesis (though to relatively harmless drugs like psychedelics or ecstasy, or the harder drugs like cocaine or heroin?). It does in fact seem to improve mood, despite the notion that it adversely impacts mental health (again a result of selection), and (despite the conventional view) the link to psychosis is not causal either. More concerning however is that getting high depletes your human capital, which likely explains its costs on income and employment. Therefore I somewhat disagree with Lasker’s framing, although the correct answer comes intuitively to me as an economist: simply compute your Langrangian of happiness vs costs and solve for the FOC! Nonetheless, this does actually update my prior sharply against cannabis: the cost to human capital bears large negative externalities as a tax on economic growth (although it's still superior to alcohol on many of these dimensions). Moreover, selection effects ARE much of the problem with drugs: you associate with other drug users, correlated with loserdom, which exerts negative peer effects on yourself, and most ‘environment’ in behavioural genetics is peer effects. As for physical health, it contains many of the same chemicals as tobacco does, so the health implications of regular daily use are likely tantamount to smoking cigarettes (in any case joints are mixed with tobacco). It's ironic that it took generations to defeat smoking, only to then progress to smoking weed. Another, somewhat Cowenesque argument against weed, is that whilst on the munchies food tastes much better, you become proportionality less discriminating when it comes to quality. I discovered that myself on a rather expensive trip to a mediocre restaurant…
Age-specific incidence of dementia is on the decline. Fed up of the negativity of the mainstream media: subscribe today for good news!
"Interventions that move children to a more favorable neighborhood have large effects but lose impact when they are scaled up because parents' equilibrium responses push against successful integration". Environment matters too, alongside heritability. Important not to lose sight of that.
Many mental illnesses, such as schizophrenia or bipolar, appear to have a genetic link. This relates to my previous point; that cannabis does not seem to cause psychosis (except perhaps in those already predisposed to such).
The UK government should drop the policy gimmick of nudging.
Drinking more than 10 drinks per week (presumably beers and wines of standard 4-5% ABV, so 20-25 units, admittedly the graph is not explicit on this) can double your risk of dementia. 9 drinks (18-22 units) can increase risk by 50%. Stay within the UK guidelines - 14 units per week is actually a pretty good threshold!
Incarceration does not reduce earnings, nor even exert a long-term effect on likelihood of employment. Fascinating.
More on marijuana: its potency is up. The same results (using police seizure data) can be found for cocaine quality in Britain. Illicit drug markets are surprisingly competitive and innovative.
It's obvious that this suboptimal NE escalation, by raising the costs of parenting, is a large factor behind South Korea’s poor fertility.
Collapse in trust in science and academia is essentially just a Republican phenomenon. An obvious effect of education polarisation, albeit wokeness has undoubtedly contributed to this too.
A reminder that our knowledge of black holes is still largely in its infancy.
UBI has no bearing on violent crime, yet may reduce property crime. Substance abuse rapidly increases. I remember Shaun Bailey, the former Conservative candidate for Mayor of London, got mocked for making precisely this point…
“Beyond amino acids, sugars, and nucleobases for RNA, scientists also found on asteroid Bennu a large, disordered network of organic molecules far more complex and chaotic than proteins, with structure and isotopic ratios not found on earth”.
Representative agents and complete markets are sufficient yet unnecessary for uniqueness, which you can get from heterogeneity. Multiple equilibria is not a function of heterogeneity (as can be demonstrated with SMD): we have multiple models where such arises.
Yes there are many intangible features of human communication, hence agglomeration effects, that cannot easily be substituted by the web. The evidence on that front appears to be accumulating.
China does not have a weak safety net relative to similarly rich nations. It is more socialist than you think.
“Stablecoins are indirectly instruments of US debt monetization. The overwhelming majority of them are backed by US government bonds.” They also exert increased pressure on fragile currencies.
The so-called paradox of choice always seemed like a dubious idea to me. To accept such a paradox would be an attack on the notion that humans are capable of rationality. Glad to see empirical confirmation.
Cowen spoke recently about a brain drain from Britain. Very high marginal tax rates on our talent pool relative to other rich economies is likely a significant factor.