localizer pixel
Jan 5, 2021

Why workers in cities may be more productive

“Workers are more productive in large cities” (Bacolod et al., 2020). This has been well established for years, but the reasons behind the phenomenon remain a bit of a chicken-or-egg question: Does being together make workers more productive, or do inherently more productive workers tend to congregate? And, what exactly makes them more productive in the first place?

Economists Marigee Bacolod (Naval Postgraduate School), Marcos Rangel (Duke), and Bernardo Blum and William Strange (University of Toronto) offer a novel contribution to this question with a study using Lumosity data. In it, they find that cities are magnets for people adept at certain types of learning. Their paper “Learners in Cities” distinguishes between “crystallized intelligence”—accumulated knowledge and facts—and “fluid intelligence,” or the ability to solve new problems. By looking at the scores of people who played a variety of Lumosity games and comparing them to the density of the population where they live,(1) they found that this second type of intelligence was higher in people who decided to live in cities.

Whereas the arithmetic game Raindrops asks people to draw on their crystallized memory of addition, subtraction, division, and multiplication, games like Memory Match and Brain Shift ask players to strategize how to achieve a goal. Raindrops tests mental math abilities we might use on a daily basis but which we learned long ago, while Memory Match challenges working memory, or the ability to manipulate new information temporarily stored in the brain. Brain Shift likewise tests learning rather than retrieval by asking its players to switch between different game demands, and to strategize how best to do that. Over time, the researchers saw not only that the initial scores of people who live in cities were higher, but the rate at which they improved was greater than people in less densely populated areas. So, it seemed that the urbanites were better learners, because they more easily adapted to and mastered these new learning scenarios.

Taken together, scores on the games that gauge working memory and task-switching (here, Brain Shift and Memory Match) can tell us something about an individual’s executive function. Executive function refers to “the mental processes that enable us to plan, focus attention, remember instructions, and juggle multiple tasks successfully.” Task-switching and working memory are crucial components of executive function. Therefore, combining scores in task-switching and working memory games may provide an estimate of executive functioning ability. Since researchers saw that high zipcode density correlated with higher scores in both crystallized intelligence and fluid intelligence, they concluded that executive function corresponded with urban areas, too. Interestingly, and as the researchers note, adjacent research found executive function to be almost entirely genetically driven, (2) suggesting that it isn’t the cities that affect executive function, but that those who inherit high levels of executive function may sort themselves into cities.

As the authors note, cities don’t just attract learners—they also “have a dynamic effect on productivity, i.e., they ‘teach’ workers to be more productive over time.” This occurs both through the competition for work, richer work offerings, and the cognitive challenges inherent in city living itself. It may be for these reasons that the density of the city matters, too: the study found that higher population density meant higher learning ability. So, you might experience a higher learning benefit from, say, moving to Chicago than to Houston, because Chicago is over twice as dense.

If density and learning—and possibly productivity—are intertwined, then urban and city planning affect a local economy, the authors point out. That is, throttling the expansion of a city can cause innovation and learning to suffer: if there isn’t enough space for people to move into, or if cities become prohibitively expensive, young learners are less likely to move to those places. As the authors put it, “Land use regulation can impede access to agglomeration, and the lost benefits of agglomeration can be a huge welfare cost.”

What does it mean for city-based learners that the pandemic has transitioned so many to at-home work? Do the factors that contribute to learning and that attract learner types remain relevant when work is not done in person? According to economist Enrico Moretti (UC Berkeley), people in the early 1990s predicted that the internet would democratize information so that it wouldn't matter where people lived. Thus, the thinking went, cities would no longer agglomerate knowledge-based industries that would attract the educated elite—the internet would be a generally democratizing force.(3) That didn’t happen, and as Moretti’s 2012 book The New Geography of Jobs argues, educated elites clustered in cities more and more, so that cities that had successfully capitalized on the new knowledge economy continually outstripped their less successful, smaller counterparts whose populations both shrank and became less educated over time (lifespans declined, too).

Recently, though, a reversal in line with the 90s hypothesis has occurred: some big cities have seen an exodus as the pandemic wears on and people work from home (in San Francisco 1 in 13 people had broken a lease by mid-year).(4) While the expense of cities is likely the main cause of this shift to less expensive areas (Sacramento has grown as San Francisco has shrunk), we can also surmise that, once people land the city-based jobs that they seek out, they may not depend on the cities to sustain that relationship to learning. Parsing the reasons why people are leaving the largest cities is a topic for another paper, but the timing of Bacolod, Blum, Rangel, and Strange’s study provides food for thought as we (and cities) face an uncertain future.

By Aimee Fountain

Notes:

  1. Bacolod and colleagues used anonymized location data for 3-digit zip codes, following Safe Harbor guidelines for data de-identification.

  2. See Friedman, Naomi. P., Akira Miyake, Susan E. Young, John. C. DeFries, Robin P. Corley, and John K. Hewitt. 2008. “Individual differences in executive functions are almost entirely genetic.” Journal of experimental psychology: General 137 (2):201.

  3. See richmondfed.org/publications/research/econ_focus/2019/q1/interview

  4. See sfchronicle.com/business/article/SF-tenants-break-leases-in-startling-numbers-15347851.php

Share this on Facebook
Tweet this
Email this