Embedded Files
Abstract: Healthcare systems worldwide face increasing nursing shortages, but the consequences remain poorly understood. This paper examines how nursing shortages in hospitals affect healthcare provision and patient health, leveraging the 2011 Swiss franc stabilization. Due to wage rigidity in the German healthcare sector, cross-border commuting became more attractive for German registered nurses, causing German hospitals to lose staff based on their proximity to the border. Using rich patient- and hospital-level administrative data in a matched difference-in-differences approach, I find that German border hospitals lost, on average, 12.5% of their nursing staff. In response, hospitals reduced care intensity, leading to a decline in surgeries. While hospitals attempted to prioritize care reductions (triage), even patients with urgent medical needs experienced cutbacks, resulting in a sharp rise in mortality rates and a stagnation in life expectancy. These findings highlight the fragility of healthcare systems to labor market regulations and labor scarcity.
You can download my job market paper here. In addition, you can find a short summary of the paper below:
Fig. 1: Number of German cross-border commuters newly starting to work in the Swiss healthcare sector each quarter (raw means and adjusted for linear trends)
Fig. 1: Number of German cross-border commuters newly starting to work in the Swiss healthcare sector each quarter (raw means and adjusted for linear trends)
The stabilization of a strong Swiss franc in 2011 incentivized more German healthcare workers to start commuting to Switzerland. This increase was driven by registered nurses formerly working in German hospitals close to the border as these hospitals couldn't adjust wages due to rigidities, and registered nurses face low legal barriers and high demand in Switzerland. Consequently, nurse staffing rates declined by 12.5% on average in German hospitals near the border after 2011 and relative to similar but unaffected hospitals further inland. Other occupations in or outside the healthcare sector were not affected.
Fig. 2: Effect of the Swiss Franc Stabilization in 2011 on the number of patients treated per nurse in hospitals
Fig. 2: Effect of the Swiss Franc Stabilization in 2011 on the number of patients treated per nurse in hospitals
As German hospitals are subject to care mandates preventing them from decreasing output at the extensive margin (i.e., treating fewer patients), hospitals decrease output at the intensive margin (i.e., less treatment per patient). For patients in affected border hospitals, this translates into a 10% higher patient-to-nurse ratio and a decrease in the probability of receiving surgery by 12%, even when accounting for patients' medical needs. This decrease in care intensity is not explainable by a change in demand, e.g. through changes in patient composition.
Fig. 3: Effect of the Swiss Franc Stabilization in 2011 on mortality rates in hospitals
Fig. 3: Effect of the Swiss Franc Stabilization in 2011 on mortality rates in hospitals
Consequently, in-hospital mortality rates increased by more than 4% (or around 0.1%-points). Hospitals aim to allocate scarce resources toward patients with higher medical needs ("triage"). However, patients with urgent diagnoses and high age also face lower treatment intensity leading them to face the highest increase in mortality. For instance, for patients with myocardial infarction ("heart attack"), the probability of in-hospital death increases by almost 18% (or 1.5%-points).
Fig. 4: Effect of the Swiss Franc Stabilization in 2011 on aggregate life expectancy
Fig. 4: Effect of the Swiss Franc Stabilization in 2011 on aggregate life expectancy
To sum up, nursing scarcity in hospitals results in a measurable reduction in the amount of care per patient and an increase in mortality, especially among vulnerable subgroups. This even translates into a decrease in regional life expectancy - a commonly used demography-adjusted indicator to capture changes in death rates - by 0.32 statistical life years. The effect size is both statistically and economically significant given it refers to 34% of the standard deviation and 27% of the interquartile range. Extensive sensitivity and robustness checks rule out potential alternative explanations that could confound the results, such as changes in healthcare demand, healthcare supply outside hospitals, and economic conditions.
Page updated
Google Sites
Report abuse