Panle Jia Barwick and Shanjun Li, Cornell University and NBER, and Deyu Rao and Nahim B. Zahur, Cornell University

Air Pollution, Health Spending and Willingness to Pay for Clean Air in China

Understanding the health impact of air pollution and consumer willingness to pay (WTP) for clean air is critical for understanding the benefit of environmental regulations. Based on the universe of credit and debit card transactions in China from 2013 to 2015, Barwick, Li, Rao, and Zahur provides analysis of the impact of PM_2.5 on healthcare costs for the entire population of a developing country. To address potential endogeneity in pollution exposure, the researchers construct an instrumental variable by modeling the spatial spillovers of PM_2.5 due to longrange transport. They incorporate the IV method into a distributed-lag model estimated with B-splines to flexibly capture the effect of past air pollution. Their analysis shows that PM_2.5 has significant impacts on health spending in both the short and medium term and that consumers exhibit avoidance behavior in spending. The annual reduction in national health spending from complying with the World Health Organization's annual standard of 10 µg/m³ would amount to over $40 billion, nearly 7% of China's total health spending in 2015. The researchers' estimates suggest a lower bound of annual household WTP of $9.25 for a 10 µg/m³ reduction in PM_2.5.

---

Robin Burgess, London School of Economics; Jonathan M. Colmer, University of Virginia; and Michael Greenstone, University of Chicago and NBER

The Economics of Marine Conservation

---

Mark J. Borgschulte, University of Illinois at Urbana-Champaign; David Molitor, University of Illinois at Urbana-Champaign and NBER; and Eric Zou, Illinois University

Smoked Out: The Effect of Wildfire Smoke on Labor Market Outcomes

Borgschulte, Molitor, and Zou examine the labor market effects of air pollution events caused by wildfire smoke. Wind transports smoke hundreds or even thousands of miles from its source, generating plausibly exogenous air pollution events in distant cities. They measure labor market responses to smoke using a novel linkage of satellite images of wildfire smoke plumes in the U.S., pollution monitor data, and labor market outcomes over the period 2006-2012. Each additional day of smoke exposure reduces per capita income by 0.03 percent that year, with income losses continuing into the year following exposure. Consistent with prolonged effects from transient smoke events, the researchers find that smoke exposure reduces labor force participation and increases Social Security claiming. Lost earnings from annual wildfire smoke exposure sum to 1.33 percent of annual labor income, on average. Borgschulte, Molitor, and Zou estimate that the welfare cost of these lost earnings is substantially larger than the mortality cost of wildfire smoke.

---

Nicholas Muller, Carnegie Mellon University and NBER

Individual Discount Rates During the Great Depression: Evidence from Firewood Prices in Portland, Oregon

Muller estimates individual discount rates on a monthly basis from 1922 to 1935. To do so, the analysis gathers price quotes for firewood data and estimates rates of time preference from the premium paid for dry, seasoned fuel, relative to green wood. Computing discount rates over 168 months facilitates a first-of-its-kind assessment of time series variation in individual discount rates. Muller examines the influence that wages, inflation, returns of stocks, yields on bonds, as well as cost-of-living indices have on individual discount rates. An assessment of how the macroeconomic shocks before, during, and after the Great Depression affected consumers' discount rates is conducted. Over the entire sample the discount rate is estimated to be 12 percent. Discount rates increased by a factor of four during the recession of 1923 to 1924, and fell by a factor of two following the stock market crash of October, 1929. Key factors in determining consumers' discount rates are: variation in wages, inflation, stock market returns, and bond yields. The results provide empirical support for the uncertainty hypothesis as an explanation for the precipitous fall in consumption expenditures following the crash.

---

Jeffrey G. Shrader, Jr, New York University

Expectations and Adaptation to Environmental Risks

Climate change is expected to have large, negative effects on the global economy. Adaptation by individuals and firms will determine, in part, how much damage ultimately occurs. Shrader introduces a method for estimating forward-looking adaptation based on changes in expectations about the weather, provides conditions under which public forecasts provide good measures for these expectations, and formalizes identification of ex ante adaptation using ex post observations. To apply the method, build a novel dataset of El Niño/Southern Oscillation (ENSO) forecasts and estimate adaptation by North Pacific albacore harvesters to ENSO-driven climate variation. The results show that, in this setting, nearly all of the effect of climate variation can be controlled through adaptation. Detailed, firm-level data allows for exploration of mechanisms, showing that vessels primarily adapt by timing entry into the fishery.

---

Tatyana Deryugina and David Molitor

Long-Run Health Dynamics in the Wake of Disaster: Evidence from Hurricane Katrina

Natural disasters may have sizable long-run consequences for victims' health, but such effects are difficult to assess. Using administrative Medicare data that capture 97 percent of US individuals aged 65 and older as well as many long-term disabled, Deryugina and Molitor estimate the short- and long-run effects of Hurricane Katrina on mortality of those living in New Orleans prior to the storm. Mortality increases due to Hurricane Katrina are overwhelmingly concentrated in the week of the storm, and longer-run increases are largely absent. Surprisingly, eight years after the storm, Hurricane victims have lower cumulative mortality, despite the fact that the Hurricane raised very short-run mortality and caused massive and long-lasting dislocation. The researchers show that the most likely explanation for long-run improvements is the relocation of New Orleans residents to areas with better health outcomes.

---

Leslie A. Martin and Samuel J. Thornton, University of Melbourne

To Drive or Not to Drive? A Field Experiment in Road Pricing

Martin and Thornton describe the results of a large road use pricing experiment that installed GPS responders in 1400 vehicles and implemented usage, time-of-day, and cordon charges via a system of virtual accounts. Using six-second location data collected over an eight to ten month period, they find a mean price elasticity of -0.13 to per kilometer charges, which is consistent with the literature on short-term demand response to fuel price increases. The researchers find that uniform road use charges lead primarily to reductions in high-speed driving and off-peak road use. Charges targeted at peak times or central areas are more successful in reducing driving under congested road conditions. They also study the margins on which drivers respond. Work commutes and school pickups and drop-offs are unchanged. Most reductions come from households making fewer trips to shops and malls. Finally, Martin and Thornton show that because low-income drivers drive less and respond more, they benefit the most from replacing existing sources of road revenues with fees that better reflect each driver's contribution to road use externalities.

---

Paramita Sinha, Research Triangle Institute; Martha L. Caulkins, University of Maryland; and Maureen L. Cropper, University of Maryland and NBER

Do Discrete Choice Approaches to Valuing Urban Amenities Yield Different Results than Hedonic Models? (NBER Working Paper No. 24290)

Amenities that vary across cities are typically valued using either a hedonic model, in which amenities are capitalized into wages and housing prices, or a discrete model of household location choice. In this paper, Sinha, Caulkins, and Cropper use the 2000 Public Use Microdata Sample (PUMS) to value climate amenities using both methods. They compare estimates of marginal willingness to pay (MWTP), first assuming homogeneous tastes for climate amenities and then allowing preferences for climate amenities to vary by location. They find that mean MWTP for warmer winters is about four times larger using the discrete choice approach than with the hedonic approach -- mean MWTP for cooler summers is twice as large. The two approaches also differ in their estimates of taste sorting. The discrete choice model implies that households with the highest MWTP for warmer winters locate in cities with the mildest winters, while the hedonic model does not. Differences in estimates are due to three factors: (1) the discrete choice model incorporates the psychological costs of moving from one's birthplace, which the hedonic models do not, (2) the discrete choice model allows for city-specific labor and housing markets, rather than assuming a national market, (3) the discrete choice model uses information on market shares (i.e., population) in estimating parameters, which the hedonic model does not.

---

Erich Muehlegger, University of California at Davis and NBER, and Richard Sweeney, Boston College

Pass-through of Input Cost Shocks under Imperfect Competition: Evidence from the U.S. Fracking Boom (NBER Working Paper No. 24025)

The advent of hydraulic fracturing lead to a dramatic increase in US oil production. Due to regulatory, shipping and processing constraints, this sudden surge in domestic drilling caused an unprecedented divergence in crude acquisition costs across U.S. refineries. Muehlegger and Sweeney take advantage of this exogenous shock to input costs to study the nature of competition and the incidence of cost changes in this important industry. They begin by estimating the extent to which US refining's divergence from global crude markets was passed on to consumers. Using rich microdata, the researchers are able to decompose the effects of firm-specific, market-specific and industry-wide cost shocks on refined product prices. They show that this distinction has important economic and econometric significance, and discuss the implications for prospective policy which would put a price on carbon emissions. The implications of these results for perennial questions about competition in the refining industry are also discussed.

---

Bryan Bollinger, Duke University; Jesse Burkhardt, Colorado State University; and Kenneth Gillingham, Yale University and NBER

Peer Effects in Water Conservation: Evidence from Consumer Migration

Social interactions are widely understood to influence consumer decisions in many choice settings. Bollinger, Burkhardt, and Gillingham identify causal peer effects in water conservation utilizing variation from consumer migration. After using a machine learning approach to classify extremely high-resolution remote sensing data, they show that the water conservation effects can be attributed to peer effects in the diffusion of dry landscaping. the researchers provide evidence that without a price signal, peer effects are muted, highlighting an important relationship between information transmission and prices. These results inform water use policy in many areas of the world threatened by recurring drought conditions.

---