Economic research shows that Southern cities benefit the most from increasing concentrations of skilled workers. Mississippi needs to catch up.

This is the second of a two-part series analyzing the economics behind Mississippi’s brain drain. Part one looked to game theory to explain why many young Mississippians choose to leave the state. 


The analysis of good and bad equilibria in the first part of this article led to something of a chicken-or-the-egg problem. Mississippi’s economy does not provide enough high-skill jobs, so skilled workers leave the state. Businesses that need high-skill labor then avoid Mississippi, because they worry about the lack of employable workers.

Identifying what started the vicious cycle in the first place – and what could have been done about it — is really only possible long after the fact. Agricultural mechanization, for example, has made hundreds of thousands of jobs in Mississippi obsolete, but it would have been very difficult to make that prediction in the 1930s.

A game of chance?

One reason specific predictions are difficult is because industry locations — the growth or decline of which can determine whether an area grows or stagnates — are often determined essentially by chance.

A well-known example of this phenomenon is the carpet industry in Dalton, Georgia, which got its start from a wedding gift given by a teenager in 1895.[1] A girl had woven a bedspread for a new couple. Her talent was noted, and she and some others in her town started weaving items in the same style. Their business grew, and eventually a specific technique they had perfected was realized to be useful in mechanized carpet manufacturing. Carpet companies located their factories in Dalton to be able to hire the weavers there for their expertise in the technique, and this led other carpet companies to realize that Dalton was a perfect location for them as well. Fifty years after the girl made her wedding gift, Dalton, Georgia, was the carpet capital of the world.

That such an insignificant thing as a wedding present could lead to an industry employing nearly 20,000 people a century later is very intimately tied in with the notions of good and bad equilibria described in the first part of this series.

In his book Geography and Trade, Nobel Prize-winning economist Paul Krugman wrote that, just as our simple models predict a vicious or virtuous cycle of decline or growth seemingly without some underlying cause, it is often the case that insignificant events “start a cumulative process in which the presence of a large number of firms and workers acts as an incentive for still more firms and workers to congregate at a particular location.”

If historical accidents can initiate the process of sustained growth or steady decline, how can we expect to predict what will get Mississippi out of its bad equilibrium?

Even though we cannot say for certain what will (or won’t) bring about change in Mississippi, we do know a little about the conditions in the modern world that make such positive change more likely. We also have evidence about specific mechanisms that can lead to growth.

Cities are the hubs of innovation

The study of cities is one subfield of economics that can provide some answers to Mississippi’s problems. It’s a line of research that Harvard economist Ed Glaeser has helped re-popularize recently, but its basic ideas were well understood more than a century ago, when the legendary British economist Alfred Marshall wrote about “the [great] advantages which people following the same skilled trade get from near neighbourhood to one another.”

The relevant aspect of this research for Mississippi’s problem is that the growth of cities often resembles the good equilibria discussed in the first part of this article. It is a dynamic in which a virtuous cycle of economic and population growth develops through the exchange of ideas among skilled workers. As Marshall described it in 1890:

inventions and improvements in machinery, in processes and the general organization of the business have their merits promptly discussed: if one man starts a new idea, it is taken up by others and combined with suggestions of their own; and thus it becomes the source of further new ideas.

More recent research in this field clarifies the role education plays in bringing about this type of good equilibrium. In their paper Cities, Skills and Regional Change, Glaeser and his co-authors Kristina Tobio (also of Harvard) and Giacomo Ponzetto (of Pampeu Fabra University in Barcelona, Spain) looked at the effects on future growth of various characteristics of different metropolitan areas across the country. In their paper, the researchers construct a model in which population and income changes can be attributed to various factors like firm productivity, land values, local amenities, and the number of entrepreneurs. They then link their theoretical model with real-world data on population and education levels in metropolitan areas in different regions of the U.S. from 1970 to 2000.

Southern cities benefit most from education gains

The model’s results reinforce the link between education the growth of population and income. What’s most interesting – and most relevant for Mississippi – is that the growth effects of education they identify are particularly strong in the South.

When they consider all metropolitan areas in the U.S., an increase of 5 percentage points in the share of the adult population with bachelor’s degrees[2] causes population growth over 30 years to be about 8 percentage points higher than it would have been otherwise.

For the South, however, their results show a similar increase in the college-educated population to be associated with a 19 percentage-point increase in population, by far the largest predicted growth in any region of the country, and more than double the national average.

These results suggest that our state’s low share of workers with college degrees (the second-lowest in the nation) not only predicts that Mississippi will grow more slowly than the rest of the country, but also that it will grow even more slowly relative to other Southern states.

To see how this might happen, let’s look at how Mississippi’s two largest metropolitan areas compare to similarly sized metro areas in Alabama.  The share of adults over the age of 25 with bachelor’s degrees in the Gulfport-Biloxi metropolitan area is 20.7 percent. The same figure for Tuscaloosa, Alabama, which has roughly the same population, is 27.4 percent.[3] In Jackson, 30.3 percent have a college degree, while in Huntsville, Alabama, a comparably sized city, 36.9 percent do. The research of Glaeser, et al., suggests that in the coming decades we can expect Tuscaloosa and Huntsville to grow about 25 percent faster than Gulfport and Jackson.

For a closer look at how Mississippi’s cities stack up educationally, explore this map (Author’s Note: Statistics cited in text come from 1-year ACS estimates for total metropolitan areas, while map shows 5-year ACS estimates within city boundaries):

Many of the most dynamic, fastest growing cities in the South also have much higher concentrations of more educated workers. More than 40 percent of the populations of Austin, Durham, and Raleigh have college degrees. In Atlanta, Dallas, and Nashville, college graduates make up more than 30 percent of the population, all higher than our most-educated metropolitan area, and significantly higher than Mississippi as a whole, in which only 20.7 percent of the population has a college degree.

Poised for entrepreneurial growth

Perhaps equally important to Mississippi is that the research cited above also concludes that the effect of higher education levels on economic productivity in the South is significantly greater, and of a somewhat different nature, than in other regions. (In their model, productivity in a modern economy is essentially output per worker.)[4]

Again, it is not surprising that more skilled workers lead to much higher productivity, and the data show that a 5 percentage-point increase in the share of workers with college degrees increases overall economic productivity by about 25 percentage points over 30 years for the nation as a whole. In the South, however, this effect is again greater, with a 5 percentage-point increase in college graduates being associated with a 33 percentage-point increase in productivity.

The productivity increase in the South is also of a different nature than in the rest of the country. For the entire U.S., about 85 percent of the increased productivity associated with a more skilled workforce comes from an increase in firm-level productivity; that is, most of the increase comes from existing firms’ doing their jobs more efficiently, with only about 15 percent of the increase coming from a larger number of entrepreneurs starting new businesses.

In the South, however, more than 40 percent of the increase in productivity is associated with an increase in entrepreneurship. The overall economic efficiency gains come almost in equal part from a more diversified set of businesses in an area as from existing firms’ productivity gains.

The difference in the source of the productivity gains in the South is particularly important in light of our analysis above of the potentially large effects of changes to an area’s economy, like the effect of agricultural mechanization on the Delta or the movement of the carpet manufacturing industry to Dalton, Georgia. In the model used in the Glaeser, et al.’s research, increased entrepreneurship in an area acts to limit the effect of negative economic events: just as a well-diversified portfolio of stocks is less risky than having all your money in one company, an area with many different businesses is more protected from adverse technological shocks or other “accidents of history” that can set off the downward spiral into bad equilibrium.

It starts with education

The results of the research above provide evidence for a significantly positive effect of increased education on Mississippi’s ability to attract and retain skilled workers. Population and economic growth go hand in hand, and data from the last part of the 20th century show that, in the South, the effects of education on these variables are two-fold: they are both larger and more robust, in the sense that the resulting economic growth is of a form that protects an area’s economy against once again falling into a bad equilibrium trap.

Identifying what works and what doesn’t to improve educational outcomes is difficult, controversial, and, fortunately for me, outside the scope of this article (and my expertise). Mississippi may be one of the least educated states in the nation, but that low starting point may actually be good thing looking forward, as there is ample room for increasing the share of skilled workers in the state’s economy. When it comes to education in Mississippi, there is a lot of low-hanging fruit to be picked.

The benefits of education are greater than the sum of its parts. No one would question the benefits of education at the individual level: a person can almost always better their life outcomes through education of one form or another. But as we saw above, an educated life helps more than just the person who gets that education; it is part of a larger process that can determine whether a state thrives and succeeds or stagnates and fails. And it’s time for Mississippi to get to work.


[1] See Krugman, Geography and Trade, 1991.

[2] The researchers use this measure as a proxy variable for overall education levels; as I mentioned in the first part of the article, the arguments are applicable to other types of education, such as job training or vocational apprenticeships, as well. The results should not be interpreted to conclude that a college education is all that matters.

[3] In one sense, it should be expected that Tuscaloosa, home of the University of Alabama, would have a higher share of college graduates than Gulfport-Biloxi, which has no major university. On the other hand, universities are an important source of urban development, thus raising the question whether Mississippi was wise to place its largest universities in small or mid-sized towns.

[4] When applying their model to earlier periods in US history, the authors’ measure of productivity can be affected by location-specific advantages (such as proximity to waterways) and the level of transportation technology; however, as technology progresses to higher and higher levels, as in a modern economy, this measure is essentially the same as an output-per-worker interpretation.

In this article