Stuff I found interesting in December

1) How often does correlation equal causation? One neat example is Wilde and Hollister (2002):

In four of eleven non-experimental studies, the sign is the opposite of the true effect.

This seems like an extremely important topic. Suppose you come across some econ study that shows a clear correlation between X and Y and includes all the controls you can think of, other than bad controls. Imagine you thought that the chance that ‘X causes Y’ was 35% prior to seeing the study (just a guess/hunch, not based on much research). What should your new probability be?

2) On inequality:

Research by a trio of French economists—Thomas Piketty, Emmanuel Saez and Gabriel Zucman—has popularised the notion that American income inequality is soaring … Others have cast doubt on the trio’s findings, however—notably Gerald Auten of the Treasury Department and David Splinter of the Joint Committee on Taxation, a nonpartisan group in Congress. We first analysed their work in 2019, as part of a cover story. It modifies the French trio’s methodology and comes to a very different conclusion: American post-tax income inequality has hardly risen at all since the 1960s.

3) On a culturally unaware colleague:

He's 23. He doesn't use TikTok, Instagram, or Twitter. His favourite drink is Bud Light and was confused when an old dude at work asked him why he still supports them. He's dating a 20 year old college junior which isn't much of an age gap to most people but you already know that bothers a very loud subsection of the population. He saw someone wearing a sweater that says "From the River to the Sea" and asked which ocean the sweater was referring to; he's a huge cartography nerd. When he saw a manager put he/him pronouns in a sent email, he was confused and asked me "what does that mean? I thought he was already a man?". We had a bit of a coworker get together one day and played drunk truth or dare. Someone asked him what is body count was. His response was "I played MW3 a lot when I was 10 so it's definitely somewhere in the hundreds"

4) A reply from Nonlinear

5) Anti-immigration voters just care more:

I find that compared to pro-immigration voters, anti-immigration voters feel stronger about the issue and are more likely to consider it as both personally and nationally important. This finding holds across virtually all observed countries, years, and alternative survey measures of immigration preferences and their importance. Overall, these results suggest that public attitudes toward immigration exhibit a substantial issue importance asymmetry that systematically advantages anti-immigration causes when the issue is more contextually salient.

This result, combined with the fact that more voters are anti-immigration than pro-immigration to begin with, suggests that if you are pro-immigration you should consider just shutting up. I think immigration levels should be higher, but if I were a politician I would certainly not talk about either numbers specifically or immigration more generally. Decrease salience, increase numbers.

6) Congratulations to 8-year-old chess prodigy Bodhana Sivanandan for her title at the European championships. Here is a lovely video of her game against an 80 year old former British chess champion:

7) I haven’t yet watched it all, but the clips I’ve seen from the Lex Bezos interview were interesting. Here’s the full video:

8) I did not know very much about the leader of Germany’s populist radical right party. This profile was interesting:

Ms Weidel comes from a well-to-do family in a small west German town. She is armed with a phd in economics, and tends to prefer the controlled arena of the boardroom or studio to flesh-pumping through crowds. Her career trajectory before politics was rocket-fuelled. She worked for Goldman Sachs, a global investment bank, as well as Allianz, an insurance giant, before starting a private consultancy. She spent several years in China, but heeded warnings that it could be a career mistake to be branded a “China hand”.

Openly gay, Ms Weidel lives primarily in Switzerland. She and her partner, a Swiss filmmaker with Sri Lankan origins, are raising two sons, aged seven and ten. Ms Weidel says that although her partner holds “very, very liberal” views—and despite the German media’s intrusions into their privacy—she has been strongly supportive of her political career.

9) On betting with your children:

Over time, they've learned that my being willing to bet, especially at large odds, is pretty informative, and often all I need to do is offer. Lily was having a rough morning, crying by herself about a project not working out. I suggested some things that might be fun to do together, and she rejected them angrily. I told her that often when people are feeling that way, going outside can help a lot, and when she didn't seem to believe me I offered to bet. Once she heard the 10:1 odds I was offering her I think she just started expecting that I was right, and she decided we should go ride bikes. (She didn't actually cheer up when we got outside: she cheered up as soon as she made this decision.)

10) Life in a hospital prison for an actual genius:

Doctors deemed Kurtaj unfit to stand trial due to his acute autism so the jury was asked to determine whether or not he committed the alleged acts - not if he did so with criminal intent.

A mental health assessment used as part of the sentencing hearing said he "continued to express the intent to return to cyber-crime as soon as possible. He is highly motivated."

The jury was told that while he was on bail for hacking Nvidia and BT/EE and in police protection at a Travelodge hotel, he continued hacking and carried out his most infamous hack.

Despite having his laptop confiscated, Kurtaj managed to breach Rockstar, the company behind GTA, using an Amazon Firestick, his hotel TV and a mobile phone.

11) Prediction markets aren’t magic

12) The bogleheads forum during the financial crisis:

I have been retired for 10 years. I am one who has said over and over again. Stay the course. Look for the long term. Yeah, sure. That's fine until today. Today did it. I am just starting to be scared so that I won't tell my wife what happened today...stocks down...bonds down...I'm down. Our retirement funds are sucking down the drain. I lost today alone a year's worth of normal distributions for expenses. I keep thinking tomorrow will be a turn around. I have said that for 30 days.

13) Research Roundup 2023:

Homeopathy is a perfect ‘null field’. There is no plausible mechanism for it to work. Despite this, there is a wealth of published research, much of it in reputable journals. In effect, the field of homeopathy can act as a control to compare with other scientific fields.

This paper takes a sample of 50 randomised placebo-controlled trials of homeopathy and extracts features. Overall, they calculate an effect size (Hedge’s g) of 0.36 versus placebo - does this look familiar? Yes, it’s the same size as antidepressants over placebo!

To be clear, I do not think antidepressants are equivalent to homeopathy, however, it does show how far researcher bias can push an inactive treatment (homeopathy) over another inactive treatment (placebo).

14) I know I’m late to the party, but I’ve been watching Silicon Valley and it’s much better than I expected. Thought it would be Big Bang Theory 2.0, have been pleasantly surprised. Recommend if you enjoyed Pete Holmes’ Crashing. I guess I also recommend Crashing if you enjoyed Silicon Valley.

15) Are birds smarter than mathematicians? On pigeons attempting the Monty Hall Problem:

A series of experiments investigated whether pigeons (Columba livia), like most humans, would fail to maximize their expected winnings in a version of the MHD (Monty Hall Dilemma). Birds completed multiple trials of a standard MHD, with the three response keys in an operant chamber serving as the three doors and access to mixed grain as the prize. Across experiments, the probability of gaining reinforcement for switching and staying was manipulated, and birds adjusted their probability of switching and staying to approximate the optimal strategy. Replication of the procedure with human participants showed that humans failed to adopt optimal strategies, even with extensive training.

16) Fuck Nuance (h/t Matthew Barnett):

It would be foolish, not to say barely comprehensible, to argue against the very idea of nuance. That would be like arguing against the idea of yellow or the concept of ostriches. Nor does it make much sense to think of nuance as something we can add to or take away from theory just as we please. That is a bit like the author whom Mary McCarthy described as busily revising a short story in order to “put in the symbols” (Goodman 1978:58). What I call “Actually Existing Nuance” in sociological theory refers to a common and specific phenomenon, one most everyone working in sociology has witnessed, fallen victim to, or perpetrated at some time.

17) I like Sam Bowman and Stian Westlake’s website on what a centre-right plan for the UK ought to look like:

We need a government that invests in research and technology in a comparable way to other advanced economies. It must do so in a way that encourages radical, high-impact innovation, which will require a diversity of funding streams and room for experiment. It must be backed up with a regulatory system that creates space for new, tech-enabled insurgents to enter established markets, and a capability to persuade other countries to adopt our rules and standards.

I am not on the centre-right (at least, I don’t think that I am!), but there is much to agree with in this report.

18) On poverty and crime:

We use data on four Swedish lotteries matched to data on criminal convictions to gauge the causal effect of financial windfalls on player’s own crime and their children’s delinquency. We estimate a positive but statistically insignificant effect of lottery wealth on players’ own conviction risk. Our estimates allow us to rule out effects one fifth as large as the cross-sectional gradient between income and crime. We also estimate a less precise null effect of parental lottery wealth on child delinquency.

But on the other hand, another study:

We estimate the effect of losing Supplemental Security Income (SSI) benefits at age 18 on criminal justice and employment outcomes over the next two decades. To estimate this effect, we use a regression discontinuity design in the likelihood of being reviewed for SSI eligibility at age 18 created by the 1996 welfare reform law. We evaluate this natural experiment with Social Security Administration data linked to records from the Criminal Justice Administrative Records System. We find that SSI removal increases the number of criminal charges by a statistically significant 20% over the next two decades.

That’s all for this year, I hope you have a wonderful 2024!