Empirical Essays on Firm-Level Digitalization and on Short-Time Work

Abstract

"The behavior of firms in the economy is an important determinant of market outcomes. For labor markets specifically, how work is organized within firms is pivotal. In three empirical essays using data from Germany, I study firm behavior and the (re)structuring of work in response to a negative economic and health shock - the Covid-19 pandemic - with a focus on investments in digital technologies and on short-time work. Thereby, this dissertation aims to contribute to an advanced understanding of the role of firms in labor markets. The first study demonstrates that the pandemic led to an increased diffusion of digital technologies in Germany. It is joint work with Christina Gathmann, Laura Pohlan and Duncan Roth. We use information from a self-designed survey module matched to administrative data to show that two-thirds of firms invested in digital technologies, three quarters of which cite the pandemic as an explicit reason for the investments. Investments took place mostly in hardware and software to enable a decentralized organization of work and were complemented by firm-provided training. Based on event-study specifications combined with entropy balancing, we document that digital investments lead to an insurance effect, shielding firms and their workers against the economic impacts of the shock. Investing firms have better employment developments and rely less on short-time work. Medium-skilled as well as young employees are the main beneficiaries of this insurance effect. The remaining two studies shift the focus towards the labor market insurance program of short-time work (STW), specifically the German scheme called Kurzarbeit. STW is a type of job retention scheme that works through subsidizing hours not worked during an economic crisis. The firm can flexibly reduce its wage costs in response to a shock by only paying for hours that were actually worked. For the affected employees meanwhile, the state covers a substantial percentage of the lost earnings. At the same time, the worker-firm relationship is not severed during STW, and friction costs resulting from job loss and rehiring are avoided. Hence, for transitory shocks, STW can be an efficient policy measure. During the pandemic, usage of Kurzarbeit reached unprecedented levels, with a peak of six million employees being in STW in the spring of 2020. The second study, co-authored with Malte Schierholz and Bernd Fitzenberger, probes the reporting of STW in firm surveys. For establishments, Kurzarbeit provides considerable flexibility. While the establishment has to notify the Federal Employment Agency about the number of workers that might be affected by STW prior to using it, it can then flexibly choose the actual utilization of STW. Real-time data on STW figures coming from firm surveys, however, exhibit an over-reporting bias, the sources of which we explain in the study. Using a high-frequency establishment survey linked to newly available administrative data on Kurzarbeit, we trace parts of the bias to establishments with STW being more likely to participate in a business survey and to peculiarities in the reporting of STW in multi-establishment firms. Further, using a yearly survey, we find that the remaining over-reporting decreases when establishments are asked retrospectively about a month with STW, and disappears with three months of retrospection. Three months is also the time by which the notification for reimbursements has to be handed in by the establishment, allowing us to link the over-reporting to how Kurzarbeit operates. In the third study, I investigate the effects of Kurzarbeit. I document that, besides financial reasons, firms value and use STW because it allows for labor hoarding. During the pandemic, I show that lower-productivity firms are more likely to select into STW, a pattern not observed during the financial crisis. This selection pattern is explained by the differing types of crises and their heterogeneous impact on establishments along the productivity distribution. I then turn to the labor market effects of the scheme in the pandemic, specifically employment effects (that is, the number of jobs saved from destruction by STW) and windfall effects (that is, jobs that are supported by STW but which would not be lost in the scheme's absence). Adjusting for the aforementioned selection in a matched event-study design at the establishment-level, I find significant positive employment ramifications of Kurzarbeit, with utilizing firms having 3-4% higher employment levels while in STW. After exiting STW, these effects quickly dissipate, a result driven by outflows among STW firms being initially lower, but higher after the end of STW. Post-STW outflows are not just delayed layoffs; a majority of those outflows are voluntary. Extrapolating the obtained employment effects and comparing them to the total number of STW-supported jobs reveals the existence of sizable windfall effects (i.e., deadweight losses), a finding that is supported by further indicators. Taken together, the three essays collected in this dissertation study how German firms restructured work processes in response to the shock of the Covid-19 pandemic. The shift to remote work was facilitated by an investment push in digital technologies, the repercussions of which outlast the crisis. Derived from firms' extensive usage of STW, the findings on the effects of Kurzarbeit can help inform policymakers in tweaking the scheme. In particular, the results on windfall effects suggest that Kurzarbeit could be improved by ensuring that the support scheme is well-targeted to those firms that actually need it." (Author's abstract, IAB-Doku) ((en))

Cite article

Kagerl, C. (2024): Empirical Essays on Firm-Level Digitalization and on Short-Time Work. Erlangen, 170 p. DOI:10.25593/open-fau-1397