Understanding Declining Fluidity in the U.S. Labor Market

Abstract

We document a clear downward trend in labor market fluidity that is common across a variety of measures of worker and job turnover. This trend dates to at least the early 1980s if not somewhat earlier. Next we pull together evidence on a variety of hypotheses that might explain this downward trend. It is only partly related to population demographics and is not due to the secular shift in industrial composition. Moreover, the decline in labor market fluidity seems unlikely to have been caused by an improvement in worker-firm matching, the formalization of hiring practices, or an increase in land use regulation or other regulations. Plausible avenues for further exploration include changes in the worker-firm relationship, particularly with regard to compensation adjustment; changes in firm characteristics such as firm size and age; and a decline in social trust, which may have increased the cost of job search or made both parties in the hiring process more risk averse.

Labor Reallocation, Employment, and Earnings: Vector Autoregression Evidence

“…We caution against this because our estimates of a zero relationship between macroeconomic aggregates may not generalize to other countries or time periods. 6 See, among many others, Davis and Haltiwanger (2014), Faberman and Justiniano (2015), Hyatt and Spletzer (2013), and Molloy et al (2016) 7 Some models such as Bils et al (2014) and Schoefer (2015) include wage rigidity as a means of solving the Shimer (2005) puzzle that in the standard Mortensen and Pissarides (1994) framework it is otherwise difficult to have a strong relationship between output and employment. Although these models have the implication that a higher earnings level may lead to lower employment and so less job creation, we caution against such a prediction for two reasons.…”

“…Excess reallocation, called labor market churn, is procyclical and increases along with other measures of the health of the labor market. 1 A number of studies including Davis and Haltiwanger (2014), Hyatt and Spletzer (2013), and Molloy et al (2016) present evidence that measures of aggregate labor reallocation have declined dramatically since the start of the millennium, and may also done so in previous decades. But whether and how job reallocation and churn affect other labor market outcomes and the economy as a whole remains unknown, despite increasing interest from researchers and policymakers.…”

“…Our empirical findings, and especially our tests for joint multivariate significance, can assist in the interpretation of previous empirical findings. Studies such as Davis and Haltiwanger (2014), Hyatt and Spletzer (2013), and Molloy et al (2016) consider the causes and consequences of different labor reallocation measures in separate empirical specifications. However, if changes in one labor reallocation measure leads to changes in another, then failure to include or control for multiple measures can lead to omitted variable bias in the analysis of labor reallocation.…”

Search and Multiple Jobholding

“…27 Davis and Haltiwanger [2014] indicate that an increasing share of employment is concentrated in occupations that require licensing, and that this could contribute to increasing the cost of occupational mobility. Molloy et al [2016] find little support for this explanation.…”

“…That is, s e is exogenous in the model, but it is likely that with endogenous on-the-job search efforts, the increase in ω 2 over time would contribute to reducing s e . Regarding the implications, Molloy et al [2016] point out that the decline in U.S. labor market dynamism may have negative consequences, such as a less efficient allocation of resources, but also positive effects coming from reduced uncertainty over one's labor market trajectory. Against this backdrop, our analysis makes a point that has so far been overlooked: conditional on meeting an employed worker, there is a nonnegligible probability that this worker will opt for the multiple jobholding option.…”