This is a Synthetic World: Three Essays on Causal Inference, Applied Econometrics, and Machine Learning for Policy Analysis
A World Where Policy Cannot Be Avoided
Pandemics are not individual-level events. Unlike epidemics which are more localized,
Exposure is:
- Non-optional
- Non-local
- (Generally) Non-isolable
Outcomes depend heavily on:
- Government action
- Collective behavior
COVID-19 forced policy decisions at massive scale
The Central Policy Tension During COVID-19
Non-pharmaceutical Interventions
Current Literature
Much existing econometric literature on the effect of COVID policy emphasizes health impacts (e.g., case growth, hospitalizations) and whether NPIs mitigated the spread of the virus.
For example, researchers studied how opening of NFL stadiums impacted the spread of the virus or how earlier lockdowns affected case rates in hte United States.
Far less is known about secondary outcomes:
- Economic activity, labor markets, environmental outcomes
A key question remains then:
Can policies protect public health without causing large economic or other social harms?
Why Identification Is Hard
- Pandemics create a latent, system-wide “background treatment”:
- All units are exposed in some way
- Policies respond to this background risk
- Spillovers and network effects are pervasive
- Implications for evaluation:
- SUTVA is often violated
- Untreated control units are hard to identify
- Standard DiD or SCM approaches can fail
- Observed outcomes may conflate policy effects with the underlying pandemic shock
Implications for Policy Analysis
If standard econometric techniques cannot provide satisfactory answers to these problems, this limits our ability to measure the broader impact of COVID policy
If we have limited knowledge of this, then we stand less prepared for future pandemics.
Role of Machine Learning
- Enhances counterfactual estimation by:
- Flexibly modeling complex dynamics (such as, low rank matrix estimation to denoise outcome data)
- Selecting relevant control groups via formalized algorithms such as k-means or the \(\ell_1\) norm for variable selection
- Mitigating spillover effects using proximal causal inference methods
Technical Motivation
However, many modern causal inference methods lack ready-to-use software
Existing SCM tools in R, Stata:
- Can be very limited in scope
- Cannot easily implement or extend novel identification strategies
- Are not free.
To address this gap, I developed mlsynth (machine learning synthetic control), an open-source Python package
mlsynth allows analysts to use modern machine learning tools from the synthetic control literature to address spillover effects, mitigate the arbitrarity of donor pool selection, and perform annalysis when no donors are present.
Paper 1: Hawaii’s COVID-19 Quarantine
Question:
What were the economic costs of Hawaii’s March 2020 mandatory quarantine?
Method:
Synthetic Historical Control (SHC)
Outcomes:
- Tourism demand
- Labor market conditions
- Statewide economic activity
Why Hawaii?
- Most extreme U.S. COVID containment policy, more so than any other state in the country
- Introduced a mandatory 14-day quarantine on all arrivals (March 26, 2020) until September 2020
- Effectively sealed borders and shut down tourism
- No comparable U.S. state intervention
Core Policy Question
What are the economic costs of decisive early action under uncertainty?
Early containment may prevent long-run damage
A key question is the degree to which the intervention impacted the economy, in contrast to places like Florida who did not such restrictions.
Difficulty in studying this case
Hawaii is:
Furthermore, because not every state collects tourism data, no credible untreated comparison units
Standard DiD and SCM designs will fail
Methodological Solution
Synthetic Historical Control (SHC)
- Uses the treated unit’s own past
- Constructs counterfactuals from historical segments
- Requires no geographic control units
- Designed for:
- One-off shocks
- System-wide interventions
SHC Intuition
- Match the last pre-treatment window
- Using weighted combinations of:
- Earlier historical segments where intervention was not active.
- Learn how Hawaii typically evolves, absent either pandemic or lockdown
- Extrapolate forward absent treatment
Economic Outcomes
Tourism Demand
Visitor Days
Hotel Occupancy
All measured as: - Annualized year-over-year growth rates
Data
- Monthly time series
- 30 years of pre-treatment data (1990 to 2020).
- Sources:
- Hawaii Department of Business Economic Developmemt and Tourism
- Treatment begins: March 2020
Inference
- Placebo tests in time
- Conformal prediction intervals
- Distribution-free
- 90% out-of-sample coverage
Contribution
- First causal estimate of Hawaii’s quarantine effects
- Demonstrates SHC in an extreme policy setting
- Provides evidence on economic costs of near-zero-COVID policies
- Broad relevance for crisis policymaking
Paper 2: Impact of National Lockdown on Air Quality in India
The COVID-19 Lockdown
- On March 25, 2020, India began a strict lockdown to combat the spread of COVID-19.
- Among the strictest globally
- Sudden halt to:
- Transportation
- Industrial activity
- Construction
- While the first order indicator is spread of the virus, there was also a large carbon emissions shock
Motivation
- Air pollution is one of India’s most severe public health challenges
- Fine particulate matter (PM2.5) is the primary pollutant of concern
- Chronic exposure affects over 1.4 billion people
- Persistent exceedance of WHO and national standards
Sources of PM2.5 in India
- Vehicular emissions
- Industrial production
- Construction activity
- Biomass burning (household and agricultural)
- Seasonal and meteorological factors
Severity and Spatial Variation
- National average PM2.5 far exceeds WHO guidelines
- Large regional heterogeneity:
- Indo-Gangetic Plain: extreme winter pollution
- Major metros consistently among the world’s most polluted (especially Delhi).
- Pollution is persistent, not episodic
Health Impacts
- Long-term exposure increases:
- Cardiovascular disease
- Respiratory illness
- Stroke and lung cancer
- Air pollution causes ~1.5 million premature deaths annually
- Substantial reductions in life expectancy, especially in Delhi
Economic Impacts
- Significant healthcare expenditures
- Reduced labor productivity
- Losses in tourism and agriculture
- Estimated costs ≈ 1–1.5% of GDP annually
- Policy response: National Clean Air Programme (NCAP)
Lockdown as a Natural Experiment
- Policy applied uniformly across the country
- Sharp timing and clear intervention onset
- A rare opportunity to study large-scale emissions reductions
- Particularly relevant for national-level air quality
Observed Changes During Lockdown
- PM2.5 declined sharply across monitored locations
- Reductions often in the range of 30–50%
- Major metros experienced large AQI improvements
- Effects strongest in early lockdown phases
What Existing Studies Show
- Consistent evidence of pollution declines during lockdowns
- Findings documented across:
- Effects observed for PM2.5, NO₂, and PM10
Dominant Empirical Approaches
- Pre–post comparisons
- Year-over-year comparisons
- Graphical trend analysis
- Satellite imagery
- Simple regression models
Limits of Existing Evidence
- Largely descriptive
- No explicit counterfactual construction
- Confounding from seasonality and weather
- Limited policy relevance
Causal Approaches in the Literature
- Interrupted Time Series (ITS)
- Synthetic Control and augmented SCM
- Causal machine learning methods
- Mixed findings across contexts
Methodological Gap
- Few credible causal designs for national-scale interventions
- Makes the SHC method a natural candidate
Data
- Source: SHRUG – Socioeconomic High-resolution Rural-Urban Geographic Data Platform for India
- Spatial coverage: 635 districts across India
- Temporal coverage: Jan 1998 – Dec 2020 (monthly observations)
- Outcome: PM2.5 concentrations, population-weighted at district level
National-level analysis: - Population-weighted average across all districts → single monthly time series
Treated districts (city-level analysis): - Delhi, Mumbai, Bangalore, Kolkata (major tech hubs/metropolitan areas). - Aggregated subdistricts to district-level averages
Contribution of This Chapter
- First causal evaluation of India’s nationwide lockdown on PM2.5.
- If successful, illustrates how government action like this can significantly mitigate emissions in the short term.
Paper 3: Locking Away Prosperity? Evaluating the Labor Impacts of Vaccine Mandates
Motivation
COVID-19 NPIs initially focused on lockdowns and masks
Early research emphasized timing and public health benefits
Subsequent work documented economic and labor market costs
Less is known about later-stage pandemic policies
One key policy passed by many jurisdictions was the idea of vaccine passports/mandates.
From Lockdowns to Mandates
- Vaccines became widely available by early 2021
- Policymakers sought alternatives to broad lockdowns
- Vaccine mandates emerged as a targeted intervention
- Goal: protect public health while sustaining economic activity
Why Focus on Restaurants?
- High-contact, indoor settings
- Central to urban leisure economies
- Sensitive to both consumer confidence and labor supply
- A leading concern during mandate debates
Policy Question
How did city-level vaccine mandates affect restaurant employment?
New York City (August 2021)
San Francisco (August 2021)
New Orleans (August 2021)
Los Angeles (November 2021)
Each implemented strict vaccine mandates for indoor dining and other public venues in 2021. Other cities (i.e., Chicago) did too, but they did so when a new COVID variant (Omicron) became viral, a critical confounder for identification.
Why These Cities Matter
- Large, economically important metros
- Early adopters of vaccine mandates
- Clear policy onsets and enforcement timelines
- Urban outliers in the national policy landscape
Expected Employment Channels
- Supply side:
- Worker compliance or exit
- Staffing frictions
- Demand side:
- Consumer confidence
- Willingness to dine indoors
Net effects are theoretically ambiguous. Whatever the theoretical relationship, the question that remains unanswered is the degree to which these mandates affected employment.
Contribution of This Chapter
- First causal analysis of vaccine mandates on restaurant employment
- Focuses on labor market outcomes
- Evaluates mandates as an economic policy tool
Empirical Strategy
- Compare treated cities to synthetic counterfactuals, using staggered adoption SCM methods
- Use MSAs without similar mandates as donors
- Focus on year-over-year employment growth
- Separate analysis by city
Data
- Monthly MSA-level employment data
- Source: BLS / FRED
- Time span: 1991–2022
- Outcome:
- Full-service restaurant employment
- Leisure and hospitality employment
- Both as Year on Year Growth rates
What this Dissertation Adds
- Pandemic policy is, in principle, a joint health–economic decision problem
- Governments do not choose policies based on epidemiological outcomes alone
- Economic consequences shape political feasibility, compliance, and long-run welfare
- While health effects may be transient, economic impacts can persist even after the virus goes away
- Industry contraction
- Labor market scarring
- Changes in regional economic structure
- Outcomes studied in this dissertation capture these longer-run margins:
- Tourism: regional exposure to mobility and border restrictions
- Restaurant employment: labor market recovery in high-contact sectors
- Air pollution: economic activity and mobility responses beyond reported statistics
- Credible counterfactuals for these outcomes allow policymakers to:
- Assess short-run costs versus long-run consequences
- Distinguish temporary disruptions from lasting damage
- Evaluate whether decisive action under uncertainty pays off
- Substantively, this work helps answer: > Can non-pharmaceutical interventions protect public health without imposing persistent economic harm?