In my initial post on this blog, “Why Honeycomb Commons,” I mentioned that “I… advocate for social systems which affirm people rather than control them and economies and technologies which are built with a human and local scale in mind.” I want to provide some evidence for this advocacy today, which will also serve as a nice follow-up to the most recent post on brownfields. As mentioned in that post, many Midwestern cities have seen a massive population decay in their central-city areas in the last half-century. This decay has taken a toll on both their economies and their tax bases, as many employment dollars get paid to those commuting in from the suburbs, where they are taxed in different towns or counties. To remedy these issues, many economic development departments adopt the “chasing smokestacks” strategy, attempting to woo large firms to locate in prime locations they lay out for them. As I will make clear over the course of this blog, I believe “chasing smokestacks” is one of the worst strategies a city can take. Instead of seeking companies, I’d like to assert, cities desiring economic development should prioritize people: both their current residents and those they wish to attract.
The first lens through which I’ll analyze this question is that of migration. What follows is an attempt to explain why people might come to a city and where they might locate within it.
There are two perspectives from which to approach this issue: firstly, the question of whether “retail follows rooftops,” or more broadly, whether “people follow jobs” in migration patterns, and secondly, the question of how far folks are willing to commute from home to work. While academic literature and public opinion on these topics seldom yield consensus, a picture of location patterns can nevertheless be painted: It seems that while on an inter-state or inter-regional level people do migrate to follow jobs, at smaller spatial levels, residential location decisions are made based more on amenities than proximity to work, with people willing to commute on average approximately half an hour to and from work.
Do People Follow Jobs?
While the idea that retail follows rooftops seems to be somewhat accepted in real estate circles, there has been much more discussion over the broader question of whether people follow jobs or jobs follow people. A 2013 article by Jim Russell in the Pacific Standard illustrates the disagreement on this issue within the ranks of urban planners and economic geographers. Russell describes the perspective of academics on both sides of the issue: some argue that problems like poor school districts and lack of amenities cause urban flight, while others propose that these people were simply drawn to more job-rich areas. Russell himself concludes that people do follow jobs, but unfortunately provides very little in the way of an argument to support that claim. This seems to be a common journalistic thread, no one has a clear answer: a 2011 article in the Oregon Business magazine attempts to answer the question, but in the end offers little other than a few statistics and pieces of anecdotal evidence, concluding that jobs probably follow people except for recent, more mobile, college grads.
Academic literature on the topic seems to edge towards “jobs follow people,” though it is also mixed. An econometric analysis by Hastorun and Cangero (pdf) finds that the employment opportunities in a state do indeed have an effect on migration, showing that a 1% change in relative employment opportunity and relative real compensation result in .3% and .41% changes in migration, respectively. However, these researchers find that while states with high levels of amenities (parks, recreation, open space, low density, low crime) were in-migrant states despite economic struggles, economic opportunities were only sometimes enough to outweigh the negative migration impacts from lack of amenities. This seems to put them on the jobs follow people side of the argument. At a different scale, an analysis by Geoffrey Meen on the relationship between industrial and housing construction finds that, instead of crowding each other out, the two are self-reinforcing: “Jobs move to workers as relocating firms seek out highly skilled workers who, in turn, search for high quality housing locations.” Based on her own literature review, Michelle White confirms this, theorizing that “The gains to firms from suburbanizing are directly linked to the distribution of population: firms have an incentive to follow—but not lead—workers to the suburbs,” while households are more mobile.
The best source to combine a collection of academic literature is a meta-analysis performed by Hoogstra, Florax, and Dijk in 2005. Their analysis surveyed 37 different projects with a total of 308 study results to determine overall trends in academic studies of the people vs. jobs question. Their method was to use econometrics to look at the findings of the studies collectively. The results of their study were unfortunately, once again mixed. While their findings provided support for the “jobs follow people” argument, they were not always conclusive and contained a lot of variation. Evidence in favor of “people follow jobs,” while limited primarily to one large study, was markedly more common when comparisons were made at the level of U.S. states or the Bureau of Economic Analysis’s U.S. regions. Based on the collection of literature and opinion, then, it is reasonable to conclude that while economic factors may be highly prevalent in inter-state and inter-regional migration, regional and metropolitan scale location decisions are more often based on quality of life assets and amenities.
How Far Will People Commute?
If we conclude that jobs will tend to follow people at the level of metro area, we must then ask how close the jobs need to be to the people. The answer here, again, is that it depends. While the U.S. average commute time is 25.4 minutes, it varies somewhat from place to place. In Indianapolis, for example, the Marion county average time is only 22.6 minutes, but census tracts in Noblesville and in some rural areas outside of suburban Indianapolis sometimes average as much as 35 minutes per commute (See Figure 1 for more data on average commutes). With Indianapolis’s interstate assets, this means that the Marion County area can capture a large number of employees from surrounding areas, but only to those locations with interstate proximity. Figure 2 puts Marion County’s commute figures in geographic terms, visualizing the differences between commute ranges for various quartiles of the data in two locations.
So how can we make sense of this information? While it is true that some will commute from very long distances, people are motivated by very real factors to keep driving times down. Long commutes have been shown in repeated studies to lead to increased stress, blood pressure, exhaustion, and other negative health effects. In fact, researchers in 2004 found that people with longer commutes report systematically lower levels of subjective well-being and calculated that for every extra hour of commuting time, workers would need to be compensated with a 40% increase in salary to make it worthwhile. These issues place clear constraints on companies hoping to attract employees from wide-ranging geographies. A company located in the downtown area, even if immediately connected to the interstate, simply cannot reach all of the citizens in the suburbs within a reasonable commute time. If most of the higher-educated workers have moved to the comparatively amenity-rich suburbs, then why would a firm seeking these employees ever locate in the central city, choosing to face higher land costs and possible confrontation with brownfield properties?
This is why economic development must ultimately be about people. We cannot hit the jackpot by bringing a high-level firm to a struggling area; we must change the neighborhood itself. In a future post, I will outline how we can nurture the economy in a particular area by highlighting those residents who are already present, but in the context of this post, it is clear that for particular sections of a city, jobs do follow people. If we ever want to have high quality places of employment in struggling regions of a city (especially those not located near an interstate), then we must make those neighborhoods into places where people actually want to live. Furthermore, it does little good for a neighborhood to bring in jobs that none of the current residents can access. If we are to fully utilize the assets and resources of inner-city regions, we must be willing to forgo quick fixes and participate in the gradual work that over time shapes plots of land into communities.
(This post was adapted from material originally researched in late 2013. If you think I missed something, or have something to add, please let me know in the comments below.)