Category: Data Source
Do walls work?
To be clear, I do not favor building the Trump Wall (at all), still I am willing to present relevant evidence when it appears. Here is the abstract of a new paper by Benjamin Feigenberg:
This paper estimates the impact of the US-Mexico border fence on US-Mexico migration by exploiting variation in the timing and location of US government investment in fence construction. Using Mexican survey data and data I collected on fence construction, I find that construction in a municipality reduces migration by 27 percent for municipality residents and 15 percent for residents of adjacent municipalities. In addition, construction reduces migration by up to 35 percent from non-border municipalities. I also find that construction induces migrants to substitute toward alternative crossing locations, disproportionately deters low-skilled migrants, and reduces the number of undocumented Mexicans in the United States.
That is from American Economic Journal: Applied Economics, you should be able to click through the captcha and get to the paper.
Covid-19 India prize (post with fixed links)
It goes to the COVIN Working Group for their paper “Adaptive control of COVID: Local, gradual, and trigger-based exit from lockdown in India.”
As India ends its lockdown, the team, led by Anup Malani, has developed a strategy to inform state policy using what is called an adaptive control strategy. This adaptive control strategy has three parts. First, introduction of activity should be done gradually. States are still learning how people respond to policy and how COVID responds to behavior. Small changes will allow states to avoid big mistakes. Second, states should set and track epidemiological targets, such as reducing the reproductive rate below 1, and adjust social distancing every week or two to meet those targets. Third, states should adopt different policies in different districts or city wards depending on the local conditions.
This project provides a path that allows states to contain epidemics in local areas and open up more of the economy. Going forward the team plans to help address shocks such as recent flows of laborers out of cities and estimate how effective different social distancing policies are at reducing mobility and contact rates.
This project has 14 authors (Anup Malani, Satej Soman, Sam Asher, Clement Imbert, Vaidehi Tandel, Anish Agarwal, Abdullah Alomar, Arnab Sarker, Devavrat Shah, Dennis Shen, Jonathan Gruber, Stuti Sachdeva, David Kaiser, and Luis Bettencourt) across five institutions (University of Chicago Law School and Mansueto Institute, MIT Economics Department and Institute for Data Systems and Society, IDFC Institute, John Hopkins University SAIS, and University of Warwick Economics Department).
Draft of the full paper is here. And for the visualizations see their website https://www.adaptivecontrol.org
Congrats to all the authors of the paper and their institutions. And here are links to the previous Emergent Ventures anti-Covid prize winners.
And I thank Shruti for her help with this.
Question about the Census and Census data
From an anonymous reader:
You consume more economic research and data than anyone I can think of and you always ask great questions, so I wanted to ask: Do you have any experience working with census data and any thoughts about the kind of data, timeliness of data, etc, that you think is lacking from government?
…Crises like the pandemic can force governments to make some changes…I think Census has a “no turning back” sense about knowing they need to try some innovative things (like the pulse survey), so I’d be happy to hear about any thoughts you have on economic data and surveys. Is there any low hanging fruit? Is there something that frustrates the hell out of many researchers? Are there moonshots in data or data-linkage that the census could attempt that you think could be valuable?
Please do leave your suggestions in the comments…
Correlates with Covid-19 death rates
We correlate county-level COVID-19 death rates with key variables using both linear regression and negative binomial mixed models, although we focus on linear regression models. We include four sets of variables: socio-economic variables, county-level health variables, modes of commuting, and climate and pollution patterns. Our analysis studies daily death rates from April 4, 2020 to May 27, 2020. We estimate correlation patterns both across states, as well as within states. For both models, we find higher shares of African American residents in the county are correlated with higher death rates. However, when we restrict ourselves to correlation patterns within a given state, the statistical significance of the correlation of death rates with the share of African Americans, while remaining positive, wanes. We find similar results for the share of elderly in the county. We find that higher amounts of commuting via public transportation, relative to telecommuting, is correlated with higher death rates. The correlation between driving into work, relative to telecommuting, and death rates is also positive across both models, but statistically significant only when we look across states and counties. We also find that a higher share of people not working, and thus not commuting either because they are elderly, children or unemployed, is correlated with higher death rates. Counties with higher home values, higher summer temperatures, and lower winter temperatures have higher death rates. Contrary to past work, we do not find a correlation between pollution and death rates. Also importantly, we do not find that death rates are correlated with obesity rates, ICU beds per capita, or poverty rates. Finally, our model that looks within states yields estimates of how a given state’s death rate compares to other states after controlling for the variables included in our model; this may be interpreted as a measure of how states are doing relative to others. We find that death rates in the Northeast are substantially higher compared to other states, even when we control for the four sets of variables above. Death rates are also statistically significantly higher in Michigan, Louisiana, Iowa, Indiana, and Colorado. California’s death rate is the lowest across all states.
That is from a new NBER working paper by Christopher R. Knittel and Bora Ozaltun.
Racial disparities in Covid-19 outcomes
This note seeks the socioeconomic roots of racial disparities in COVID-19 mortality, using county-level mortality, economic, and demographic data from 3,140 counties. For all minorities, the minority’s population share is strongly correlated with total COVID-19 deaths. For Hispanic/Latino and Asian minorities those correlations are fragile, and largely disappear when we control for education, occupation, and commuting patterns. For African Americans and First Nations populations, the correlations are very robust. Surprisingly, for these two groups the racial disparity does not seem to be due to differences in income, poverty rates, education, occupational mix, or even access to healthcare insurance. A significant portion of the disparity can, however, be sourced to the use of public transit.
That is from a new NBER working paper by John McLaren.
The course of Covid-19 in the United States
Via Amihai Glazer. And if you wish to verify, here is another (non-smoothed) presentation of the data.
In terms of the delta this picture is not as bad as what you sometimes hear, though data on cases are far worse, with a very long and indeed continuing plateau. And since deaths lag cases by a few weeks, you still might see reason to be alarmed. Nonetheless, the trend we can see is one of improvement, at least for a little over two months.
Do note it is better for everyone if you think the death rate is still rising!
Raj Chetty talk on “The economic impacts of COVID-19: Real-time evidence from private sector data”
I have not yet viewed it, but this talk is receiving rave reviews, here is the link. Or here:
Via the excellent Samir Varma.
Database of 1,100 superspreading events
It is about time someone put this together, here are some summary conclusions:
- Nearly all SSEs in the database — more than 97% — took place indoors
- The great majority of SSEs happened during flu season in that location
- The vast majority took place in settings where people were essentially confined together, indoors, for a prolonged period (for example, nursing homes, prisons, cruise ships, worker housing)
- Processing plants where temperatures are kept very low (especially meat processing plants) seem particularly vulnerable to SSEs
Here is the full material by Koen Swinkels, via Balaji.
Political and social correlates of Covid-19 mortality
Do political and social features of states help explain the evolving distribution of reported Covid-19 deaths? We identify national-level political and social characteristics that past research suggests may help explain variation in a society’s ability to respond to adverse shocks. We highlight four sets of arguments—focusing on (1) state capacity, (2) political institutions, (3) political priorities, and (4) social structures—and report on their evolving association with cumulative Covid-19 deaths. After accounting for a simple set of Lasso-chosen controls, we find that measures of government effectiveness, interpersonal and institutional trust, bureaucratic corruption and ethnic fragmentation are currently associated in theory-consistent directions. We do not, however, find associations between deaths and many other political and social variables that have received attention in public discussions, such as populist governments or women-led governments. Currently, the results suggest that state capacity is more important for explaining Covid-19 mortality than government responsiveness, with potential implications for how the disease progresses in high-income versus low-income countries. These patterns may change over time with the evolution of the pandemic, however. A dashboard with daily updates, extensions, and code is provided at https://wzb-ipi.github.io/corona/
That is from a new paper by Constantin Manuel Bosancianu, et.al., via Alex Scacco.
A new study of face masks
We use the synthetic control method to analyze the effect of face masks on the spread of Covid-19 in Germany. Our identification approach exploits regional variation in the point in time when face masks became compulsory. Depending on the region we analyse, we find that face masks reduced the cumulative number of registered Covid-19 cases between 2.3% and 13% over a period of 10 days after they became compulsory. Assessing the credibility of the various estimates, we conclude that face masks reduce the daily growth rate of reported infections by around 40%.
That is from a new paper by Timo Mitze, Reinhold Kosfeld, Johannes Rode, and Klaus Wälde.
Buy your apples from Abraham Aardvark
It is not just researchers and co-authors who benefit from having names close to the beginning of the alphabet:
The names of traders should not matter if information is symmetric across traders. By examining export data from Chinese customs, we find persistent lexicographic biases in firm-level export records. Firms whose names are lexicographically earlier in the Chinese-character rank export more to countries that have greater language proximities to Chinese, while firms whose names are lexicographically earlier in the English-romanization rank export more to countries that have greater language proximities to English. The lexicographic biases signify linguistic visibility as a source of comparative advantage in international trade.
Here is the full paper by Hua Cheng, Cui Hu, and Ben G. Li. Via the excellent Kevin Lewis.
Policing the Police
Here is the new paper by Tanaya Devi and Roland Fryer, full title being “Policing the Police: The Impact of “Pattern-or-Practice” Investigations on Crime”:
This paper provides the first empirical examination of the impact of federal and state “Pattern-or-Practice” investigations on crime and policing. For investigations that were not preceded by “viral” incidents of deadly force, investigations, on average, led to a statistically significant reduction in homicides and total crime. In stark contrast, all investigations that were preceded by “viral” incidents of deadly force have led to a large and statistically significant increase in homicides and total crime. We estimate that these investigations caused almost 900 excess homicides and almost 34,000 excess felonies. The leading hypothesis for why these investigations increase homicides and total crime is an abrupt change in the quantity of policing activity. In Chicago, the number of police-civilian interactions decreased by almost 90% in the month after the investigation was announced. In Riverside CA, interactions decreased 54%. In St. Louis, self-initiated police activities declined by 46%. Other theories we test such as changes in community trust or the aggressiveness of consent decrees associated with investigations — all contradict the data in important ways.
Via Ilya.
Have we already defunded the police?
Here is more complete data on police expenditures, interesting throughout, via Charles Fain Lehman. The sociology of this issue I find fascinating. Usually in Progressive lore, if you defund an agency, you lower its quality and make it all the more dysfunctional. But in this case, defunding the bureaucracy, namely the police, is supposed to solve the problem. Is there anywhere a well-worked out model of why this particular bureaucracy might be different from the others? (Maybe it is, I would gladly link to such an argument!) Or, dare I say it, is this just mood affiliation and once again…politics isn’t about policy. I’ll give 4-1 odds on the latter.
The return to studying economics is fairly high
Selection bias may confound the identification of field-specific returns to higher education. This study investigates the wage return to studying economics by leveraging a policy that prevented students with low introductory grades from declaring the major. Regression discontinuity estimates show that policy-complying economics majors — who appear representative on observables — earned $22,000 (58%) higher annual early-career wages than they would have with their second-choice majors, despite otherwise-unchanged educational investment and attainment. Cross-industry wage variation explains half of the return, with economics majors channeled towards high-wage economics-related industries. Differences between institution-specific or nationally-representative average wages by major well-approximate the estimated causal return.
That is from a new paper by Zachary Bleemer and Aashish Mehta, via the excellent Kevin Lewis.
Is this why budget deficits might prove sustainable?
By Adrien Auclert, Hannes Malmberg, Frederic Martenet, and Matthew Rognlie:
We use a shift-share approach to quantify the general equilibrium effects of population aging on wealth accumulation, real interest rates, and capital flows. Combining population projections with household survey data from the US and 24 other countries,we project the evolution of wealth-to-GDP ratios by changing the age distribution,holding life-cycle asset and income profiles constant. We find that this compositional effect of aging is large and heterogeneous across countries, ranging from 85 percent-age points in Japan to 310 percentage points in India over the rest of the twenty-first century. In a general equilibrium overlapping generations model, our shift-share provides a very good approximation to the evolution of the wealth-to-GDP ratio due to demographic change when interest rates remain constant. In an integrated world economy, aging generates large global imbalances in the twenty-first century, pushing net foreign asset positions to levels several times larger than those observed until today.
Via Steven Bogden. This is very likely an important piece.