The mystery of regional inequality

Optimism soared after the Berlin Wall fell in 1989. Germany’s reunification promised not only greater political and social harmony, but economic revival—particularly for East Germany’s moribund economy. But while East Germans did benefit from integration, stark economic differences from West Germany have persisted. Wages, in particular, are much higher in the West and have remained so for decades.

Germany is not alone in this spatial dichotomy between haves and have-nots. The south of Italy and the Andalusian region of Spain, for example, experience similar second-tier status relative to their compatriots. A 2016 report from the Organisation for Economic Co-operation and Development on more than 40 countries notes that “regional disparities … in GDP per head, disposable income, safety and air pollution are widening in many.”

A December 2019 working paper (iwp 29) by Tommaso Porzio of Columbia Business School, an Opportunity & Inclusive Growth Institute visiting scholar, with Sebastian Heise of the Federal Reserve Bank of New York, investigates the causes of German pay disparity using a massive database and a hybrid labor market model. The paper concludes that labor market frictions within regional job markets—not just between them—impede good job matches, thereby limiting productivity and commensurate wage growth. The focus should be on better job matching in local labor markets rather than moving workers to more prosperous regions.

“Policies aimed at promoting more migration,” write Heise and Porzio, “are dominated in utility terms by policies that facilitate the reallocation of labor within a given labor market.”

Something as basic as a subsidy to firms to post more vacancies, they say, “would improve the matching market in workers’ home region and thus raise average welfare.” Improving job markets within a region is a better strategy than encouraging people to migrate to higher-pay regions.

Mining the gap

The economists reach these conclusions after analyzing data on German labor markets and documenting several main facts, then building a model to account for these empirical realities and, finally, using the model to explore potential mechanisms behind—and remedies for—the less prosperous East.

Heise and Porzio rely on data from several sources, including government statistics on 1.3 million German business “establishments” (distinct units of companies that operate in each county). Another source is a database of job histories of 1.9 million workers from 1993 to 2014, matching each worker with the establishment they worked at in every period. The scholars also gather information regarding family composition, living conditions, and education, and still other data on geographic cost-of-living differences, to adjust wages accordingly.

By analyzing these data, they document three facts:

(1) They verify a large, persistent wage gap between the two regions, about 26 percent higher in the West, and show that it’s not driven primarily by demographic factors, industry composition, or establishment sizes. There’s nothing “observable,” in economics jargon, about workers or industry composition that explains the gap. Instead, more than 60 percent of it is due to the reality that West German firms pay more because they’re more productive. “In other words,” they write, “workers’ location matters.” A map of average daily wages by county demonstrates that workers in West Germany enjoy far higher wages than those in the East. (See map.)

(2) All workers enjoy large wage gains by moving from East to West. This supports the idea that the gap isn’t driven by something unmeasured or less obvious. But the gains are asymmetric: East and West workers both receive higher wages if they move away from their birth region. This suggests that all workers need extra compensation for migration.

(3) The data show substantial worker mobility for both East and West Germans but with strong bias for moves closer to home. High mobility suggests that moving costs are fairly low, say the economists. The home bias may be due to personal preference, skill advantages, or social networks that assist in job hunting.

Modeling the market

Can economic theory explain these facts? In particular, can it account for the persistence of the large regional wage gap despite the lack of physical or language barriers and the fact that relatively few East Germans permanently move to the West, despite far higher pay scales.

The economists build a model that combines two approaches to understanding labor markets: “job ladder” theory that examines how workers become distributed within and among firms during careers and “spatial job allocation” theory that looks at how workers move across geographic labor markets.

The hybrid model reveals how workers sort themselves among jobs in different regions and firms. A key assumption is that these labor markets are imperfect. That is, real-world barriers like moving costs, poor job market information, or discrimination stand in the way of ideal matches between employees and employers.

Shaped with parameters of the German economy, the model allows the economists to explore and quantify these labor market “frictions” that might explain the regional wage gap and to study the impact of various policies to address barriers to better job markets.

They identify several factors that play a role, albeit limited, in sustaining the wage gap and preventing full integration of German labor markets. One is strong worker preferences for staying near their home region, or “home bias.” Another is “migration costs,” the cost of packing up and moving from one region to the other. The economists refer to these as “spatial frictions” since they limit allocation across space.

While these costs are “sizable,” the economists conclude that “the estimated barriers are significantly smaller than previous estimates.” Other research, they observe, hasn’t accounted for the fact that much of the wage gain from migration was not a compensation to workers for the frictional costs of spatial moves, but higher pay resulting from moves up the job ladder—a factor their model explicitly takes into account.

The model also highlights the importance of “reallocation frictions” due to job posting and search costs. These “prevent the reallocation of workers to more productive firms even within a region,” explain Heise and Porzio. “The model shows that a wage gap between regions can persist even in the absence of any spatial frictions.”

Policy alternatives

What does all of this mean for potential solutions to the East-West wage gap?

One option is paying for worker’s moving expenses, thereby eliminating migration cost friction. Heise and Porzio estimate that doing so would have only a small impact on average wages and national GDP, and “almost no effect on the spatial wage gap.” Aggregate welfare would show no improvement. “The small increase in aggregate GDP and wages,” the economists write, “is completely offset by the utility cost of having more workers located away from home.” Simply put, relocated workers might be slightly better paid and more productive, but they’ll be far less happy.

The economists then compare policies that foster regional integration with programs that make it easier for workers to match with a good job where they are, for example, by subsidizing firms’ search for suitable local candidates. The latter deliver just slightly lower average wage gains and GDP gains than the former, but generate significantly larger welfare gains to the average worker, who “obtains higher wages without the need to relocate.” In other words, it’s better to improve local job markets than to help workers move.

Home, sweet home

In a closing section, the economists look into the tight connection workers feel to their home region, strong enough to sacrifice better-paying jobs elsewhere. Workers are likely to move back home after becoming parents, they find, perhaps seeking family support. They also discover that East-born workers moving to the West are likely to move to counties with many Eastern workers. And workers from both regions are likely to move toward their home state within their home region, given the opportunity.

All of this suggests that it isn’t just the historic separation of East and West that drives the results on home attachment. Rather, home is where the heart remains, regardless of the specific context.

The upshot is that supporting migration to better-paying regions is less advantageous—to both workers and the national economy—than hiring subsidies that improve the functioning of the local labor market. Staying close to home is a powerful instinct, it seems, regardless of pay.