Difference-in-differences with variation in treatment timing

Goodman-Bacon, Andrew

doi:10.1016/j.jeconom.2021.03.014

Cited by 2,529 publications

(929 citation statements)

References 39 publications

Supporting

Mentioning

916

Contrasting

Unclassified

Order By: Relevance

“…Thus, within this subsample, we have three groups of provinces: early switchers, easing into Phase 1 on 11 May, late switchers on 18 May and never switchers (till 25 May). Based on this classification, we can use the [22] decomposition theorem to estimate changes in Phase 1 treatment effects across different subgroups. Our estimates imply stable treatment effects.…”

Section: Further Analysis Of Lockdown Easingmentioning

confidence: 99%

“…Conclusions are unchanged by defining different pre-treatment periods within the joint lockdown and Phase 0 periods. A second concern that arises, as articulated in [22], is that the treatment effect may not be stable over time. In our context, this means that the expenditure effects of lockdown easing may be different across early-and late-switcher provinces, perhaps indicating that other unobservable time-varying factors are driving the province-level response.…”

Section: Further Analysis Of Lockdown Easingmentioning

confidence: 99%

See 1 more Smart Citation

Tracking the COVID-19 crisis with high-resolution transaction data

et al. 2021

View full text Add to dashboard Cite

Payments systems generate vast amounts of naturally occurring transaction data rarely used for constructing official statistics. We consider billions of transactions from card data from a large bank, Banco Bilbao Vizcaya Argentaria, as an alternative source of information for measuring consumption. We show, via validation against official consumption measures, that transaction data complements national accounts and consumption surveys. We then analyse the impact of COVID-19 in Spain, and document: (i) strong consumption responses to business closures, but smaller effects for capacity restrictions; (ii) a steeper decline in spending in rich neighbourhoods; (iii) higher mobility for residents of lower-income neighbourhoods, correlating with increased disease incidence.

show abstract

Section: Further Analysis Of Lockdown Easingmentioning

confidence: 99%

Section: Further Analysis Of Lockdown Easingmentioning

confidence: 99%

Tracking the COVID-19 crisis with high-resolution transaction data

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Our main analysis relies on a series of DID event-study analyses, examining the impact of UCC entry into a zip code on a range of outcomes, relative to yet-to-be-treated zip codes. Several authors have shown that when treatment is staggered, standard DID estimates can be biased in the presence of heterogeneous treatment effects (Goodman-Bacon, 2021;de Chaisemartin and d'Haultfoeuille, 2020a,b). Sun and Abraham (2020) show that the estimated coefficients from two-way fixed effect regressions are not robust when there are heterogeneous treatment effects, while Borusyak and Jaravel (2017) show that in this case it is also difficult to identify pre-trends.…”

Section: Methodsmentioning

confidence: 99%

Do Urgent Care Centers Reduce Medicare Spending?

Currie¹,

Karpova²,

Zeltzer³

2021

View full text Add to dashboard Cite

We examine the impact of the opening of a new urgent care center (UCC) on health care costs and the utilization of care among nearby Medicare beneficiaries. We focus on 2006-2016, a period of rapid UCC expansion. We find that total Medicare spending rises when residents of a zip code are first served by a UCC, relative to spending in yet-to-be-served zip codes, while mortality remains flat. We explore mechanisms by looking at categories of spending and by examining utilization. Increases in inpatient visits are the largest contributor to the overall increase in spending, rising by 6.65 percent within six years after UCC entry. The number of emergency room visits that result in a hospital admission also increases by 3.7 percent. In contrast, there is no change in the number of ER visits that do not result in admission to hospital, in visits to physicians outside a UCC, or in imaging and tests. Overall, these results provide little evidence that UCCs replace costly ER visits or that they crowd out visits to patients' regular doctors. Instead, the evidence is consistent with the possibility that UCCs-which are increasingly owned by or contract with hospital systems-induce greater spending on hospital care.

show abstract

“…The primary source of data comes from the Censoc project outlined in Goldstein et al (2021). 6 It uses death records from social security administration for individuals who die in old age and implement data-linkage techniques to link with full-count 1940 census records.…”

Section: Data Sourcementioning

confidence: 99%

The Effects of Education on Mortality: Evidence Using College Expansions

Fletcher¹,

Noghanibehambari²

2021

View full text Add to dashboard Cite

Difference-in-differences with variation in treatment timing

Cited by 2,529 publications

References 39 publications

Tracking the COVID-19 crisis with high-resolution transaction data

Tracking the COVID-19 crisis with high-resolution transaction data

Do Urgent Care Centers Reduce Medicare Spending?

The Effects of Education on Mortality: Evidence Using College Expansions

Contact Info

Product

Resources

About