Category: Data Source
Spain’s second wave of Covid-19
Did this happen? Were Spain’s hardest hit provinces in the spring spared in the second wave?
To get a quick sense of the answers to those questions I plotted the cumulative number of cases per 100,000 population in the Spanish provinces since June 15 against the proportion of the population in the provinces that tested positive for antibodies after the first wave. If herd immunity were playing a large role in suppressing cases in the second wave, we would expect to see a negative relationship between provinces with high levels of antibodies in the population at the end of May and total case counts since that time…
Instead of a negative correlation, there is a positive, although weak, correlation between having higher prevalence of antibodies in the population and having a higher case rate in the second wave.
…Take Madrid for example, if roughly 13% of the population had antibodies after the first wave, at least one of the low-HIT models estimates the Rₑ would be approximately 60% lower than than the unmitigated reproductive rate (R₀). If population immunity were reducing transmission in the Madrid area by 60% below unmitigated levels, it seems unlikely Madrid would again have one of the highest rates of infection in the second wave [yet it does].
…Ultimately, the strongest conclusion that can be drawn from this look at infection rates is that there is not clear evidence herd immunity is playing a significant role, yet.
Also take a look at a deeper dive looking for herd immunity in Sweden (spoiler alert: no signs of it yet).
It is fine to call this inconclusive, but still the pattern predicted by standard herd immunity claims is not yet showing up. Here is the whole piece from Kbenes, very useful.
And elsewhere, this was not supposed to happen, as New York Orthodox Jews also have been cited as a “herd immunity” community:
Officials this week released statistics showing that the positivity rate in some Orthodox Jewish neighborhoods [in NYC] had grown to anywhere from 3 percent to 6 percent, significantly more than the city’s overall rate of between 1 percent and 2 percent. Officials are especially worried about the positivity rates in the Brooklyn neighborhoods of Borough Park, Midwood and Gravesend, which they have referred to as the “Ocean Parkway Cluster.”
Here is that full story (NYT).
European paintings show a rise in trustworthiness
Building on recent advances in social cognition, we design an algorithm to automatically generate trustworthiness evaluations for the facial action units (smile, eye brows, etc.) of European portraits in large historical databases. Our results show that trustworthiness in portraits increased over the period 1500–2000 paralleling the decline of interpersonal violence and the rise of democratic values observed in Western Europe. Further analyses suggest that this rise of trustworthiness displays is associated with increased living standards.
That is from a new paper by Lou Safra, et.al. in Nature. Scroll down a few pages for some good photos, basically the people in the paintings look less pissed off over time. Via Lionel Page.
Gangs, Labor Mobility and Development
We study how two of the world’s largest gangs—MS-13 and 18th Street—affect economic development in El Salvador. We exploit the fact that the emergence of these gangs was the consequence of an exogenous shift in American immigration policy that led to the deportation of gang leaders from the United States to El Salvador. Using a spatial regression discontinuity design, we find that individuals living under gang control have significantly less education, material wellbeing, and income than individuals living only 50 meters away but outside of gang territory. None of these discontinuities existed before the emergence of the gangs. The results are confirmed by a difference-in-differences analysis: after the gangs’ arrival, locations under their control started experiencing lower growth in nighttime light density compared to areas without gang presence. A key mechanism behind the results is that, in order to maintain territorial control, gangs restrict individuals’ freedom of movement, affecting their labor market options. The results are not determined by exposure to violence or selective migration from gang locations. We also find no differences in public goods provision.
That is from a new NBER working paper by Nikita Melnikov, Carlos Schmidt-Padilla, and Maria Micaela Sviatschi.
Unintended Consequences of Criminalizing Sex Work
We examine the impact of criminalizing sex work, exploiting an event in which local officials unexpectedly criminalized sex work in one district in East Java, Indonesia, but not in neighboring districts. We collect data from female sex workers and their clients before and after the change. We find that criminalization increases sexually transmitted infections among female sex workers by 58 percent, measured by biological tests. This is driven by decreased condom access and use. We also find evidence that criminalization decreases earnings among women who left sex work due to criminalization, and decreases their ability to meet their children’s school expenses while increasing the likelihood that children begin working to supplement household income. While criminalization has the potential to improve population STI outcomes if the market shrinks permanently, we show that five years post-criminalization the market has rebounded and the probability of STI transmission within the general population is likely to have increased.
That is from a new NBER working paper by Lisa Cameron, Jennifer Seager, and Manisha Shah.
America fact of the day
…according to recent polls from Quinnipiac and Monmouth, 38 percent of registered Hispanic voters in 10 battleground states may be ambivalent about even voting…
Progressives commonly categorize Latinos as people of color, no doubt partly because progressive Latinos see the group that way and encourage others to do so as well. Certainly, we both once took that perspective for granted. Yet in our survey, only one in four Hispanics saw the group as people of color.
Here is more from Ian Haney López and
Claims about antibodies and T-cells and Sweden
Buggert’s study in Sweden seems to support this position. Investigating close family members of patients with confirmed covid-19, he found T cell responses in those who were seronegative or asymptomatic. While around 60% of family members produced antibodies, 90% had T cell responses. (Other studies have reported similar results.) “So many people got infected and didn’t create antibodies,” concludes Buggert.
That is from Peter Doshi, mostly a survey on pre-existing immunity, interesting and useful and properly agnostic throughout. Here is a version of the Buggert piece, also with a link to the published version.
Note two things. First, “the kooks” saw this possibility first, and insisted on its relevance, to their credit. Second, many of “the kooks” are overly dogmatic, not always to be trusted, and they commonly shift the goalposts (when predictions about cases are falsified, they switch to pretending those were predictions about deaths). Often the non-kooks do that too of course.
For a sobering worry, here are some recent numbers from Spain.
The key to interpreting the literature is to focus on the data, and to keep an open frame of mind, rather than digging in to a particular position. Right now I am focused on observing what kind of second wave London is going to have, and how mild or bad it will be, as that is most likely to induce me to update my positions, in one direction or another.
For the pointer I thank E. Ward.
American distress (average is over)
The proportion of the US population in extreme distress rose from 3.6% in 1993 to 6.4% in 2019. Among low-education midlife White persons, the percentage more than doubled, from 4.8% to 11.5%. Regression analysis revealed that (1) at the personal level, the strongest statistical predictor of extreme distress was “I am unable to work,” and (2) at the state level, a decline in the share of manufacturing jobs was a predictor of greater distress.
As for the definition, exceptional distress is the percentage who reported major mental and emotional problems in all 30 of the last 30 days.
That is from Blanchower and Oswald, via the excellent Kevin Lewis.
What is new in behavioral genetics?
Here is an excellent conceptual survey article by K. Paige Harden, it goes well beyond the usual. Hard to summarize, but here are two good bits:
An early study using a DNA-based method estimated the heritability of height to be∼80%,and it noted that this result was “consistent with results from independent twin and family studies but using an entirely separate source of information” (Visscher et al. 2006). However, although the results from DNA-based method of estimating heritability scale with the estimates from twin and family studies, the former are typically smaller (Young et al. 2019). This discrepancy between heritability as estimated from classical twin and family studies and heritability as accounted for by measured DNA was labeled the missing heritability problem (Manolio et al. 2009). Recent work has suggested that some of the missing heritability is actually “hiding” in rare variants that are not typically measured and in the heterogeneity of genetic effects across populations (Tropfet al. 2017, Wainschtein et al. 2019, Young 2019). Whether missing or hiding, the continued gap between DNA-based estimates of heritability and estimates from twin/family studies means that the latter might still be overestimating heritability due to faulty assumptions. But it is no longer reasonable, contra some predictions, to expect that advances in human genomics will reveal that the heritability of psychological phenotypes is entirely illusory.
And this one:
In contrast to what is seen for educational attainment, most studies find a minimal effect of shared environmental factors on cognitive abilities, particularly when measured in adulthood. It has been suggested,however,that this near-zero main effect of the family-level environment masks the heterogeneity of the effects of the shared environment across the SES spectrum.An early paper by Turkheimer et al. (2003) analyzed data from a sample of twins with an unusual overrepresentation of children in poverty and found substantial effects of the shared environment on cognitive ability at age 7. Subsequent research on the genotype×SES interaction effect yielded mixed results, with several studies finding null effects or even effects in the opposite direction. However, a meta-analysis of this literature (Tucker-Drob & Bates 2016) found evidence of a significant interaction effect (albeit with a smaller effect size than estimated by Turkheimer and colleagues, an example of the winner’s curse), particularly in the United States.
The importance of the shared environment for cognitive ability has also been demonstrated us-ing adoption studies. In particular, population-wide data from Sweden allowed researchers to estimate the impact of the family environment using a unique sample of male-male sibling pairs where one brother was adopted while the other brother was raised by his biological parents (Kendler et al.2015). The IQ score of the adopted brother was, on average,∼4 points higher, an increase that varied with the education level of the adopting parents.
Recommended, interesting throughout, and worth a reread as well. I have forgotten who sent it to me, if indeed anyone did, but I thank you.
Which economic methods are in practice statistically more honest than others?
…our results suggest that the [instrumental variables] and, to a lesser extent, [difference-in-difference] research bodies have substantially more p-hacking and/or selective publication than those based on [randomized controlled trials] and [regression-discontinuity]… (p.3)
And:
We find no evidence that: (1) Papers published in the ‘Top 5’ journals are different to others; (2) The journal ‘revise and resubmit’ process mitigates the problem; (3) Things are improving through time.
That is from this forthcoming AER paper by Brodeur, Cook, and Hayes.
In contrast, this blog post argues that:
I have proposed here that we should not infer that literatures with more bunching just past .05 are less trustworthy, and that visually striking comparisons of ‘expected’ and observed test results can be quite misleading due to incorrect assumptions about the expected line.
The authors respond here. I do not yet have an opinion on this dispute, but everyone is talking about it right now, so I thought I would at least send along the basic documents to you all.
Determinants of life expectancy
Life expectancy in the US increased 3.3 years between 1990 and 2015, but the drivers of this increase are not well understood. We used vital statistics data and cause-deletion analysis to identify the conditions most responsible for changing life expectancy and quantified how public health, pharmaceuticals, other (nonpharmaceutical) medical care, and other/unknown factors contributed to the improvement. We found that twelve conditions most responsible for changing life expectancy explained 2.9 years of net improvement (85 percent of the total). Ischemic heart disease was the largest positive contributor to life expectancy, and accidental poisoning or drug overdose was the largest negative contributor. Forty-four percent of improved life expectancy was attributable to public health, 35 percent was attributable to pharmaceuticals, 13 percent was attributable to other medical care, and −7 percent was attributable to other/unknown factors. Our findings emphasize the crucial role of public health advances, as well as pharmaceutical innovation, in explaining improving life expectancy.
That is from a new paper by Jason D. Buxbaum, Michael E. Chernew, A. Mark Fendrick, and David M. Cutler. Via the excellent Kevin Lewis.
Who is my favorite public intellectual?
Over the weekend I sat in on Anna Gát’s Interintellect Salon, which I enjoyed. Many of the participants were asked who is their favorite public intellectual. My answer was something like:
Alex Tabarrok, he’d better be, I’ve been working with him for thirty years! There would be something wrong if he wasn’t. And I always look forward to reading what he writes.
So there you go. None of the other answers, worthy though they were, had equivalent support in demonstrated preference.
The fragility of herd immunity
Trouble in the Madrid region is brewing again, even though earlier seroprevalance had clocked in at about 20 percent:
Good for New York of course, here is a thread discussing the comparison, to me the conclusions seem premature. The important point in any case is that Covid-protected time periods need not last forever, and you can end up in multiple rounds of “let it rip.” As far as I know, this is the first established case of a major “second wave” in a previously hard-hit area.
The good news is that Madrid cases seem to have peaked, and furthermore the death rate is much lower the second time around, the latter being one good reason for postponing cases into later time periods rather than taking them all up front.
Note also that England has had months of open pubs, and a very quiet situation, but now cases there are doubling every six to seven days (FT). Don’t switch back to talk of deaths! The “simple” theory of herd immunity is surprised to see that new trend in cases. What I call semi-herd immunity suggests a high degree of protection for the current configuration of social relations, after some point. As those social relations change, some of that temporary herd immunity dissolves, as new infecting connections are being created and new superspreaders arise and do their thing. But that takes a while, possibly months.
The herd immunity theorists downplay the possible temporariness of the equilibrium they pinpoint. They instead prefer to focus on the (correct) point that most of the mainstream approaches did not forecast the collapse in deaths and hospitalizations found in England, Sweden, New York, and now parts of the American South. In reality, you need to put both sides of the picture together, and grasp both the insights and limitations of the herd immunity theorists.
So herd immunity does seem to be fragile, and if other developments (treatment, antivirals, steroids, masks and thus lower dosage) lower death rates, bravo, but case behavior still moves against the simple herd immunity theory, at least in Madrid. How fragile we still do not know, and I readily grant and indeed would emphasize that Madrid is the only major counterexample to date. Appreciate the limits of knowledge!
If you listen to Ivor Cummins, a darling of the herd immunity theorists, he doesn’t seem to grasp these problems of possible temporariness (he loves to switch to talk of deaths at just the wrong time), but rather treats herd immunity as “it’s over,” with a few vague qualifiers tossed in at the very end. We will see.
The Washington Consensus Works: Causal Effects of Reform, 1970-2015
Sustained economic reform significantly raises real GDP per capita over a 5- to 10-year horizon.
Despite the unpopularity of the Washington Consensus, its policies reliably raise average incomes.
Countries that had sustained reform were 16% richer 10 years later.
As for the method:
In this paper, we define generalized reform as a discrete, sustained jump in an index of economic freedom, whose components map well onto the points of the old consensus. We identify 49 cases of generalized reform in our dataset that spans 141 countries from 1970 to 2015. The average treatment effect associated with these reforms is positive, sizeable, and significant over 5- and 10- year windows. The result is robust to different thresholds for defining reform and different estimation methods.
There are dozens of books trying to tell you this is not true, but…it is true, at least as best we know.
That is all from the new paper by Kevin and Robin Grier, did you know by the way that I helped to fix them up, leading to their later marriage and also coauthorships?
Evidence from 27 Thousand Economics Journal Articles on Africa
The first two decades of the 21st century have seen an increasing number of peer-reviewed journal articles on the 54 countries of Africa by both African and non-African economists. I document that the distribution of research across African countries is highly uneven: 45% of all economics journal articles and 65% of articles in the top five economics journals are about five countries accounting for just 16% of the continent’s population. I show that 91% of the variation in the number of articles across countries can be explained by a peacefulness index, the number of international tourist arrivals, having English as an official language, and population. The majority of research is context-specific, so the continued lack of research on many African countries means that the evidence base for local policy-makers is much smaller in these countries.
Here is the article by Obie Porteus, via David Evans.
New results on the housing boom and bust
We build a model of the US economy with multiple aggregate shocks that generate fluctuations in equilibrium house prices. Through counterfactual experiments, we study the housing boom-bust around the Great Recession, with three main results. First, the main driver of movements in house prices and rents was a shift in beliefs, not a change in credit conditions. Second, the boom-bust in house prices explains half of the corresponding swings in nondurable expenditures through a wealth effect. Third, a large-scale debt forgiveness program would have done little to temper the collapse of house prices and expenditures but would have dramatically reduced foreclosures and induced a small, but persistent, increase in consumption during the recovery.
That is from a recently published article by Greg Kaplan, Kurt Mitman, and Giovanni L. Violante in the Journal of Political Economy. Here are some less gated versions.