The 10 most-cited articles in American Economic Review, 2021

We present a new estimator for causal effects with panel data that builds on insights behind the widely used difference-in-differences and synthetic control methods. Relative to these methods we find, both theoretically and empirically, that this “synthetic difference-in-differences” estimator has desirable robustness properties, and that it performs well in settings where the conventional estimators are commonly used in practice. We study the asymptotic behavior of the estimator when the systematic part of the outcome model includes latent unit factors interacted with latent time factors, and we present conditions for consistency and asymptotic normality. (JEL C23, H25, H71, I18, L66)

Does the consumption of ideologically congruent news on social media exacerbate polarization? I estimate the effects of social media news exposure by conducting a large field experiment randomly offering participants subscriptions to conservative or liberal news outlets on Facebook. I collect data on the causal chain of media effects: subscriptions to outlets, exposure to news on Facebook, visits to online news sites, and sharing of posts, as well as changes in political opinions and attitudes. Four main findings emerge. First, random variation in exposure to news on social media substantially affects the slant of news sites individuals visit. Second, exposure to counter-attitudinal news decreases negative attitudes toward the opposing political party. Third, in contrast to the effect on attitudes, I find no evidence that the political leanings of news outlets affect political opinions. Fourth, Facebook's algorithm is less likely to supply individuals with posts from counter-attitudinal outlets, conditional on individuals subscribing to them. Together, the results suggest that…

This paper studies how changes in oil supply expectations affect the oil price and the macroeconomy. Using a novel identification design, exploiting institutional features of OPEC and high-frequency data, I identify an oil supply news shock. These shocks have statistically and economically significant effects. Negative news leads to an immediate increase in oil prices, a gradual fall in oil production, and an increase in inventories. This has consequences for the US economy: activity falls, prices and inflation expectations rise, and the dollar depreciates, providing evidence for a strong channel operating through supply expectations. (JEL E31, E32, F31, Q35, Q38, Q43)

We study a Chinese policy that awards substantial tax cuts to firms with R&D investment over a threshold or “notch.” Quasi-experimental variation and administrative tax data show a significant increase in reported R&D that is partly driven by firms relabeling expenses as R&D. Structural estimates show relabeling accounts for 24.2 percent of reported R&D and that doubling R&D would increase productivity by 9 percent. Policy simulations show that firm selection and relabeling determine the cost-effectiveness of stimulating R&D, that notch-based policies are more effective than tax credits when relabeling is prevalent, and that modest spillovers justify the program from a welfare perspective. (JEL D22, D24, H25, O14, O32, P31, P35)

Using data on 800,000 corporate publications and patent citations to these publications between 1980 and 2015, we study how corporate investment in research is linked to its use in the firm’s inventions, and to spillovers to rivals. We find that private returns to corporate research depend on the balance between two opposing forces: the benefits from the use of science in own downstream inventions, and the costs of spillovers to rivals. Consistent with this, firms produce more research when it is used internally, but less research when it is used by rivals. As firms become more sensitive to rivals using their science, they are likely to reduce the share of research in R&D. (JEL D22, D25, G31, I23, O31, O33, O34)

Can targeting information to network-central farmers induce more adoption of a new agricultural technology? By combining social network data and a field experiment in 200 villages in Malawi, we find that targeting central farmers is important to spur the diffusion process. We also provide evidence of one explanation for why centrality matters: a diffusion process governed by complex contagion. Our results are consistent with a model in which many farmers need to learn from multiple people before they adopt themselves. This means that without proper targeting of information, the diffusion process can stall and technology adoption remains perpetually low. (JEL O13, O18, O33, Q12, Q16)

The high-tech sector is concentrated in a small number of cities. The ten largest clusters in computer science, semiconductors, and biology account for 69 percent, 77 percent, and 59 percent of all US inventors, respectively. Using longitudinal data on 109,846 inventors, I find that geographical agglomeration results in significant productivity gains. When an inventor moves to a city with a large cluster of inventors in the same field, she experiences a sizable increase in the number and quality of patents produced. The presence of significant productivity externalities implies that the agglomeration of inventors generates large gains in the aggregate amount of innovation produced in the United States. (JEL D62, J24, L60, O31, 034, R32)

I estimate the effects of collaborative and adversarial intergroup contact. I randomly assigned Indian men from different castes to participate in cricket leagues or to serve as a control group. League players faced variation in collaborative contact, through random assignment to homogeneous-caste or mixed-caste teams, and adversarial contact, through random assignment of opponents. Collaborative contact increases cross-caste friendships and efficiency in trade, and reduces own-caste favoritism. In contrast, adversarial contact generally reduces cross-caste interaction and efficiency. League participation reduces intergroup differences, suggesting that the positive aspects of intergroup contact more than offset the negative aspects in this setting. (JEL C93, D83, D91, J15, O15, Z13, Z21)

We leverage a large-scale incentivized survey eliciting behaviors from (almost) an entire undergraduate university student population, a representative sample of the US population, and Amazon Mechanical Turk (MTurk) to address concerns about the external validity of experiments with student participants. Behavior in the student population offers bounds on behaviors in other populations, and correlations between behaviors are similar across samples. Furthermore, non-student samples exhibit higher levels of noise. Adding historical lab participation data, we find a small set of attributes over which lab participants differ from non-lab participants. An additional set of lab experiments shows no evidence of observer effects. (JEL C83, D90, D91)