Some economics of place effects

It is not easy to study how affected you are by living in a certain place.

After all, people aren’t first observed, then randomly distributed across places, and then observed again. In more concrete terms, those who end up living in a place with high priced real estate and those who end up living in a place with low priced real estate may very well differ in several ways in terms of origin, from level of education and professional history, and it would be strange to say that the places where people ended up living “caused” these differences. Instead, it often seems more plausible that where people end up living is caused by previous mechanisms of social and economic sorting.

Nevertheless, here are two examples of the importance of knowing something about place effects. Imagine there is a program to help people from lower-income neighborhoods move to richer neighborhoods. Will the change of location affect their employment prospects or the outcomes for their children? Or imagine a retiree moving from one state to another. Will the change in place affect their health or the health care they receive? Some articles in the The Fall 2021 issue of Economic Outlook Journal tackle these questions of place effects head-on.

(Full disclosure: I have been editor-in-chief of the JEP since the first issue in 1987, so I may be predisposed to believe that the articles are interesting. Fuller disclosure: The JEP and all of its articles since the first issue have been available for free Online for a decade now, courtesy of the American Economic Association, so there is no financial benefit to me or anyone else in recommending the articles.)

It turns out that there are certain situations where one can look at those who have changed places, compare them with a “control group” who have not changed places, and draw plausible conclusions. Eric Chyn and Lawrence F. Katz provide an overview and contextual interpretation of this research in “Neighborhoods Matter: Assessing the Evidence for Place Effects” (Economic Outlook Journal, 35: 4, p. 197-222). Here’s how they describe perhaps the most important study in this area:

Beginning in 1994, the Moving to Opportunity housing mobility demonstration randomized access to housing vouchers and assistance to relocate to less troubled communities to approximately 4,600 families living in public housing projects located in low-income communities. deeply impoverished neighborhoods in five cities: Baltimore, Boston, Chicago, Los Angeles and New York. The program randomly divided families into three groups: 1) a group of poverty reduction vouchers (also known as the “experimental group”) who were offered advice on housing mobility and restricted housing vouchers. that could only be used to relocate to areas of low poverty (census regions with 1990 poverty rates below 10 percent); 2) a traditional voucher group who were offered regular Section 8 housing vouchers that had no additional location constraints (also referred to as the Section 8 group); and 3) a control group that received no assistance from the program.

When researchers started to think seriously about location effects, they looked for other situations in which people ended up leaving low-income areas. For example, studies have looked at situations in which people have had to relocate because public housing was demolished or when Hurricane Katrina destroyed existing housing, compared to findings from a similar group who did not. been forced to move. There are also studies in which a group of low-income households are offered information and advice to help them find rental options in higher middle-income areas, while a control group is not seen. offer such assistance and is therefore much less likely to move to these other areas. Note that somehow all of these approaches, in different ways, have an element of randomness that allows the researcher to make a plausible estimate of the causal effects of living in a different location.

The overall conclusion of this line of research is that when low-income households move to a high-income neighborhood, it has substantial effects for young children growing up in the new neighborhood, weaker or even no effects for those who do. are older adolescents at the time of the move, and little effect on employment or income outcomes for adults in these households. Interestingly, the positive effects for young children do not appear to be reflected primarily in school test scores, but rather manifest as the result of gains in non-cognitive skills as reflected in measures such as the number of absences or absences. school suspensions, and the chances of repeating a class. The authors write: “Studies of the Moving to Opportunity demonstration and of public housing demolitions in Chicago have found no evidence that relocation to less distressed areas has impacts on the economic outcomes of adults, but both contexts revealed significant long-term gains for young children… ”

In the same issue, Tatyana Deryugina and David Molitor discuss the “causal effects of place on health and longevity” (Economic Outlook Journal, 35: 4, p. 47-70). They point out that the magnitude of regional differences in health from one location to another is actually quite similar in the European Union and the United States. They write:

Three main results emerge from the comparison of the regional variation in life expectancy in the United States and in Europe. First, the average life expectancy is 2.8 years higher in Europe than in the United States. Second, the overall variation in life expectancy, as captured by the standard deviation or the interdecile interval of the distribution of life expectancy, is similar in both contexts. Third, most of the regional variation in life expectancy in Europe is explained by country of residence, while in the United States most of the variation is intra-state.

One of the main issues is whether it’s the location that matters or whether it’s the pre-sorting of people by income and occupation, which often correlates with location. For a given location, health could also be affected by issues such as local public health policies, or peer effects (such as if there is a culture of addiction or outdoor exercise), or by the contamination of the local environment. Moreover, these factors often overlap: for example, a low-income place may also receive less health care, be more affected by environmental pollution, and have peer effects that do not promote good health. It is not clear that “place” is always the most useful way to think about these kinds of factors, rather than focusing on the underlying determinants. The authors cite a classic example of research in this area:

As a striking example of the geographic differences in mortality across the United States, Fuchs (1974) compared the death rates in Nevada and Utah, which are neighboring states with similar climates and, at the time, levels of income and doctors per capita similar. Fuchs noted that, nonetheless, adult death rates were considerably higher in Nevada than in Utah, which he attributed to high rates of cigarette and alcohol consumption in Nevada as well as ” conjugal and geographic instability “. Even today, the average person born in Utah has a life expectancy that is 1.9 years higher than that of the average person born in Nevada.

But generally speaking, the approach is to look at the drivers between areas and then think carefully about what can be learned from such comparisons. For example, one data source is to look at the elderly, who are covered by Medicare, who move between states. Patterns of medical practice vary across the United States, so you may see Medicare patients with similar past patterns of healthcare use moving to a place where Medicare beneficiaries on average receive more care. , or where they receive less care, or those who do not. move at all. One can also look at physicians who move between regions and, looking at physicians who had similar health insurance cost models before they moved, see if their health insurance cost model tends to stay the same. even when they go to a different location. As the authors write:

Other indirect evidence that local conditions matter to health comes from articles that use movers to study how local conditions affect the delivery of health care and other non-health outcomes that could end up being account affect health. For example, Song et al. (2010) show that as Medicare beneficiaries move between regions, rates of medical diagnoses change. Finkelstein, Gentzkow, and Williams (2016) study Medicare beneficiaries who move between zones and show that place of residence affects the medical expenses of the movers. Molitor (2018) examines cardiologists who move and finds that, on average, their own practice patterns change by 60-80% of the difference in local standards between their new practice areas and their home area.

There are a lot of complexities when looking at the effects of health. So far, research based on people moving from one place to another (i.e. not on correlations looking at health effects in different places) suggests that health effects places are real, but does not yet give solid answers as to why they are real. . The authors write:

The observed geographic dispersion of life expectancy and the evidence of people moving between areas strongly suggest that where you live matters when you die. However, determining whether the health effects of the site are significant or insignificant has only been completed very recently. New evidence comparing movers to other movers to estimate the health effects of location suggests that, at least for some groups, location has an important health effect. However, more research is needed to take advantage of these findings and, in particular, to understand the effect of location at younger ages on long-term longevity. While there are many plausible mechanisms by which these location effects can materialize, the question of exactly what makes some places better for your health than others has so far not been answered directly. in no existing study.