Q&A: An Economic Lens on World Health

Associate Professor Marco Gonzalez-Navarro studies economic growth in developing countries with an emphasis on urban issues, including transportation, land use, and retail globalization. His study “Subways and Urban Air Pollution” is under revision for publication in the American Economic Journal: Applied Economics.

What’s the problem you wanted to identify or solve with this study?

We sought to answer this question: Do subways lead to improved air quality in urban areas? Many feel it’s obvious that rail transportation will lead to improved air quality because public transit will result in fewer cars on the road. However, previous research has not found conclusive evidence to support this. If public transportation gets some people to stop using their cars, traffic on the road may lessen, but that could also encourage non–transit users to drive more, potentially undoing the effect of those who left their cars at home.

On the other hand, subways can efficiently move many people in and out of downtown areas at peak hours, which are when vehicular congestion is at its worst. So it is plausible that subways could have a meaningful impact on air quality.

In our study, we found that by moving a small share of overall trips off the roads at those times and places, subways can make a meaningful dent in pollution.

What area did you focus on, and what methods did you use?

We took a global perspective, studying subway system openings all over the world wherever they occurred between 2000 and 2014; over this time period, the majority of subway openings were in China and Europe. We also used satellite data that measured a proxy for PM2.5—an EPA-regulated pollutant that’s responsible for millions of premature deaths worldwide.

Using these data, we found that air quality in these areas improved by about 4 percent in terms of reduced pollutants. The effect was larger near the city centers, and that persisted over the longest time horizon that we could measure with our data, which was about eight years. We estimated that a new subway system provides an external benefit of about $594 million per year by reducing deaths. Overall, our results suggest that the value of subway-induced improvements to air quality represents a substantial fraction of construction costs.

What’s the potential application of these findings?

Rail transportation systems are expensive to build and operate, typically requiring significant governmental subsidies to be viable. Fare revenue is usually insufficient to cover operating costs. Because this mode of transport requires subsidies, our study asks whether subways provide any benefits for citizens who don’t utilize them, which could justify such subsidies. One of these benefits is improved air quality, which has been shown to have important human-health impacts, including  reduced mortality. If a subway system leads to improved air quality for everyone living in a city, it could provide a rationale for that system to be financed by general taxation.

Associate Professor Michael L. Anderson focuses on topics such as transportation, education, population health, and health insurance. His research on the relationship between health insurance coverage and emergency room visits as well as other uses of medical services has been published in the  American Economic Journal: Economic Policy and the Review of Economics and Statistics.

What were you investigating with these studies?

Our goal with these studies was to measure the effect of health insurance on emergency room visits and hospitalizations. Under federal law, an ER must stabilize any patient who arrives there, regardless of their ability to pay. ERs are thus viewed as health care providers of last resort, and proponents of universal health insurance coverage have often argued that the costs of insurance coverage can be partially offset through a reduction in ER visits by the uninsured.

What did you find?

To estimate the effects of insurance coverage on the utilization of emergency room and hospital inpatient services, we looked at a sharp change in insurance coverage rates that resulted from young adults “aging out” of their parents’ insurance plans. Prior to the Affordable Care Act (ACA), young adults who were not in college aged out of their parents’ insurance at 19.

Using ER and hospital discharge records on millions of episodes of care across seven states—Arizona, California, Iowa, New Jersey, New York, Texas, and Wisconsin—we found that aging out resulted in an abrupt 5 to 8 percentage point reduction in the probability of having health insurance. We found accompanying drops in ER and hospital visits that implied that not having insurance led to a 40 percent reduction in ER visits and a 61 percent reduction in inpatient hospital admissions. That is, contrary to conventional assumptions, the loss of insurance caused ER visits to decrease. The drops in ER visits and inpatient admissions were concentrated in privately owned hospitals, with particularly large reductions at for-profit hospitals. ER visits to public hospitals, which typically have mandates to treat all patients regardless of insurance status or ability to pay, did not decrease.

What has been the broader impact of this research?

Under the ACA, employers and insurers are now required to cover enrollees’ children through age 26—at least for now. Nevertheless, our study results remain topical: the idea that insurance can cut costs by reducing ER visits persists as a common talking point. We should be honest about the costs and benefits of expanded health insurance coverage. The notion that people respond to health insurance coverage by reducing ER utilization simply isn’t supported by our results.

Associate Professor Aprajit Mahajan’s studies have covered a broad swath of subjects, including credit markets, health, and productivity in India and Mexico.

In this study, what’s the question you’re asking or the problem you want to solve?

Micronutrient deficiency and anemia are long-standing problems in India. Past attempts to address them have been made with tablets and food supplements, but the results have been mixed. We collaborated with the state government of Tamil Nadu to provide fortified rice to populations that have had high rates of anemia.

One prominent hypothesis for the failure of previous interventions, particularly at scale, is that they required substantial behavior modification: people had to remember to take a pill or to sprinkle a supplement over their food. There may also be financial or other barriers if they have to buy the pills and supplements.

How are you addressing that problem?

We wondered if we could design a solution that involved minimal behavioral change. Tamil Nadu has a well-functioning public food distribution program (PDS): 20 kilos of free rice are provided to all families monthly, or 35 kilos for poor households. Uptake rates are very high as well, so this seemed like a good place for our study. We’re asking, can we fortify rice with iron, B12, and other micronutrients and provide it through the existing PDS? Using this method of provision, households don’t have to change their behaviors—they’ll obtain and cook rice from the PDS just as they did in the past.

What are your next steps?

Our research team—from UC Berkeley, Stanford, Emory University, Universitat Pompeu Fabra in Spain, and J-PAL South Asia in India—has been designing the study since 2014, and the intervention will occur from November 2019 to November 2020. We’ll implement a randomized controlled trial, providing fortified rice to 110 shops distributing PDS food. We’ll follow the intervention for a year and evaluate the effects it has on the most vulnerable populations (women and children from six months to five years old).

If successful, this should provide a model for how to reduce anemia and micronutrient deficiencies through the PDS not just in Tamil Nadu but more generally in other rice-consuming states in India. We think it should work, but there are conditions under which it might not. For example, it may be that even though people consume the fortified rice, they can’t absorb the nutrients owing to conditions like chronic gut inflammation. Our thinking is that if we can’t move the needle on anemia in this way, we really need to rethink how it could be done.