Resumen de Essays on local labor markets
This thesis is composed of three essays in which I study how heterogeneity in productivity interacts with the size of local labor markets and a set of outcomes of interest. The motivation for each essay is empirical: economic theory plays nevertheless a fundamental role throughout, both ex ante in defining the relevant set of hypothesis to test in the data and ex post in providing a framework for gauging the economic significance of the results.
The notion that the dispersion of the fundamental shocks hitting the economy might change over time is an idea that has gained traction since the notion of the Great Moderation was introduced. While the Great Moderation refers to a protracted period of low macroeconomic volatility, there is evidence that microeconomic volatility has also declined substantially over the past decades, without showing a reversal in the most recent years as it has been the case for macroeconomic volatility. While this topic has recently attracted considerable attention, there has been little effort to put it into the bigger picture of the secular decline in firm entry, dynamism and overall changing nature of competition that is underway.
In the first chapter, co-authored with Heiko Stueber, we take a first step towards filling this gap and we assess the welfare consequences from the point of view of consumers. We provide evidence that the decline in microeconomic volatility has been accompanied by a fall in uncertainty for companies and entrepreneurs. Exploiting administrative data for the population of German establishments we show that reduced idiosyncratic uncertainty is empirically consistent with fewer varieties available for consumption. Less idiosyncratic uncertainty reduces the likelihood that businesses grow large at the competitors' expenses, thus decreasing the value of running a firm and discouraging entry. We show that variety is more responsive to uncertainty in large cities, so that a fall in idiosyncratic uncertainty has hampered not only consumer surplus but also the extent of agglomeration. We build a tractable spatial equilibrium model of firm entry to quantify the negative implications of a decline in uncertainty for consumers and the spatial distribution of economic activity. We find that the 2.75 percent point decline in uncertainty observed during 1990-2014 in Germany has led to an average 9\% decline in consumer surplus across cities, and a modest compression in the city size distribution.
The restructuring or opening of large plants and its consequences for the city where they are located is a hot topic for local policy-makers. A recent and famous example is the competition started among US cities when Amazon announced its intention to open a second headquarter. In the second chapter, also co-authored with Heiko Stueber, we investigate a negative consequence of sharing the labor market with a few very large firms: the systemic risk that arises when a few firms are too big relative to the average size of their competitors. Intuitively, if a large firm experiences difficulties in either production or demand and, as a consequence, it is forced to dismiss a large number of workers, its aggregate impact will not average out: slack in labor demand will drive down the wage and the economy will enter a recession that has a micro-origin. Since higher uncertainty reduces investment, large firms can represent a liability for the economy unless there exist proper policies in place.
The existing evidence in favor of granularity-driven business cycles takes mostly a narrative approach. In this chapter, we exploit spatial heterogeneity in local business cycles and variation in the concentration of economic activity in the top local firms across cities and years to show that a deviation from steady state in concentration predicts the start of local recessions. Deeper recessions tend to be preceded by a substantial build-up in local concentration, and for recessions during which local concentration stays stubbornly high after the peak it takes longer to reach the trough that marks the start of the recovery. We supplement our findings with narrative evidence on the impact of large plants on local employment fluctuations in Germany over the past 25 years. We conclude with an overview of the policies adopted in four major European countries to limit the labor-market impact of shocks that affect individual firms. We discuss how in many of these countries more should be done to further bring down the systemic risk introduced by large firms.
The belief that individual specialization may matter for worker productivity dates back to Adam Smith, but it has received surprisingly scant empirical attention to date. In the third chapter, I construct a novel measure of specialization using data on the frequency-adjusted count of tasks involved in an occupation. I use this measure to test for the existence of a specialization wage premium and study whether large cities tend to attract more specialized workers, consistent with the Smithian argument that the division of labor is limited by the extent of the market. I provide evidence supporting the existence of a specialization wage premium, and - by showing how average specialization rises with market size - of a specialization urban wage premium.
Next, I connect cross-city differences in specialization-driven productivity with the divergence in location choices between college and non-college US workers observed during 1980-2000. I find that specialization-driven productivity alone has been an important force behind the rising concentration of college workers in 1980 skill-abundant cities. At the same time, however, I also find that specialization growth over this period has been the highest for non-college workers, thus suggesting it may have counteracted the rise in the skill premium.
The re-discovery of the importance of specialization as an engine of individual productivity has several interesting implications for future research, from the rising importance of between-firm wage dispersion to the complementarity with automation.
