Author:
The recent surge in inflation fueled by the Covid-19 pandemic and the war in Ukraine brought the question of who suffers most from price increases back into the political arena. This column presents evidence from Germany, indicating that labour market adjustments provide a surprisingly large buffer against increasing inequality in living standards following energy price shocks.
It is a well-documented fact that people’s spending habits vary widely. This means the burden of price changes is not borne equally by all consumers (Jaravel, 2021). In recent years, for example, the media repeatedly echoed concerns about the disproportionate effect that increases in energy prices have on poorer households (who tend to spend a larger share of their income on energy). While at first glance this suggests that energy price increases are accompanied by an increase in the inequality of living standards, this view neglects what happens in the labour market.
People who are hit harder by the price shock might look to boost their earnings. In theory, they are incentivised to renegotiate their wages with their current employer or even change jobs. If either a pay rise or a job switch allows a person to recover part of their personal cost increases through a corresponding increase in earnings, the distributional implications of a given price shock are muted. This column presents the results of a recent study I conducted using German micro-data to understand the role of such labour supply dynamics in shaping the distributional consequences of price shocks.
Energy spending across Germany
As a starting point, I consider geographic variation in how much different households spend on their energy bills (as a share of total expenditures). Using data from the Einkommens- und Verbrauchsstichprobe, a survey in which German households are asked to document their expenditures for three months systematically, I show that German administrative counties differ vastly in the relevance of energy expenditures for the overall consumption budget of households. Figure 1 presents the average household’s energy expenditure share by county, combining the expenses on gasoline, gas, heating oil, electricity and other energy sources. While in some counties, such as Potsdam-Mittelmark, the average household spends 13% of their budget on energy, this amount is more than halved in higher-density counties, such as Frankfurt or Berlin. Besides population density, these differences are also correlated closely with local commuting infrastructure and the quality of housing. The region where someone lives has a notable effect on how much they need to set aside for energy costs.
This also suggests that households across Germany have differing vulnerabilities to energy price shocks. Figure 2 plots the price increase for various energy types from 1998 to 2018. Consumer prices for gasoline and electricity roughly doubled over this period, while the price of heating oil increased by a factor of three. For comparison, Germany’s overall inflation over the same period is around 32%, underlining the significant contribution of energy to price increases.
Figure 1: Energy expenditure shares by German administrative county (2018)
Source: author’s calculations based on Einkommens- und Verbrauchsstichprobe (2018), Basic File 3.
Note: The figure presents county-level estimates of energy expenditure shares, adjusted for differences in the socio-demographic composition of counties.
Figure 2: Consumer price series for energy goods
Sources: Eurostat for electricity and gas, en2x for gasoline and oil, and DESTATIS for other energy types, including coal, wood, other solid fuels, and central heating.
Note: The figure plots a series of average yearly prices, normalised to 100 in 1998.
Energy prices and earnings
The stark increases in energy prices and the variation in consumption across counties allow me to study the labour supply responses of workers living in high- and low-energy expenditure areas. To do so, I use data from the Social Security Registry in Germany. The research institute at the Federal Employment Agency (IAB-FDZ) provides (for a random subsample of all social security employees) anonymised information about their full employment history, including earnings, job tasks and their employer. I can further link each person in the sample to their respective county of residence and county of work, allowing me to map county-specific energy cost increases to their labour market outcomes.
The results indicate that after energy prices increase, earnings tend to increase more strongly in counties in which households spend a larger share of their budget on energy than in low-expenditure counties. On a year-to-year basis, this allows people to recover around 40% of county-specific increases in energy costs. Over a period of five years, income adjustments cover more than 85% of additional costs. Earnings dynamics, therefore, provide a non-negligible buffer against the unequal consequences of the price change.
Pay rises versus job swaps
Part (but not all) of this income recovery is due to job mobility. People who are hit harder by energy price increases are more likely to change their employers and do so in a way that corresponds to gains in earnings. For job switchers, the increase in earnings fully offsets the higher energy costs. At the same time, firms respond to the higher mobility of their employees. In an attempt to retain employees, they raise wages. This implies that employees who do not switch jobs can recover part of the increased costs through income adjustments.
But the positive relationship between cost increases and income is not uniform across the population. It is particularly pronounced among younger workers, for whom the costs of switching between employers are lower. Older workers tend to suffer more acutely from energy cost shocks. The same is true for immigrant workers and workers in Eastern Germany, whereas native-born workers living in the West profit from the labour market adjustments.
What do the results mean for overall inequality?
My study implies that labour market dynamics limit the price-driven increase in inequality across German counties. But there are some important caveats. First, higher job mobility is no silver bullet: some parts of the population clearly fail to translate cost increases into increased earnings. Second, to isolate cleanly the impact of consumers’ labour supply responses on inequality, the underlying study abstracts from or accounts for other relevant factors, such as the effect of energy prices on firms or other sources of income besides earnings.
Previous research shows that energy price shocks do not affect all firms in the same way. This means that the sorting of employees across firms is an important consideration when assessing the distributional impact of price shocks (Kehrig and Ziebarth, 2017). Moving beyond energy prices and thinking more broadly about inflation also implies that price increases affect other aspects of consumers’ income, such as dividends on assets (Del Canto et al., 2023). Inflation also generally shifts the value of nominal wealth, redistributing away from older (wealthier) people to the young (Doepke and Schneider, 2006).
Understanding the distributional implication of price shocks is a complex task, and more research is needed to better inform political responses to inflation surges. This column does, however, highlight that employees’ decisions about changing jobs or renegotiating wages can partially shield them against cost increases. This suggests that future research on the issue should include realistic labour market dynamics in their models (in the paper, I propose ways to do so). From a political point of view, it also suggests that some policies designed to ease the burden of higher prices might discourage people from switching jobs or asking for higher wages, partially reducing the policy’s effectiveness.
Price shocks such as the recent energy crisis garner a lot of focus within the media and in political discourse. The disproportionate impact on lower-income households in terms of expenditure share means this is largely justified. But labour market dynamics should not be overlooked. To design effective policies to support those most vulnerable to inflation, understanding mechanisms such as earnings dynamics in the face of price shocks is vital.
Author: Yannick Reichlin
References
Del Canto, Felipe N., John R. Grigsby, Eric Qian, and Conor Walsh. 2023. Are inflationary shocks regressive? A feasible set approach. (No. w31124). National Bureau of Economic Research.
Doepke, Matthias, and Martin Schneider. 2006. “Inflation and the Redistribution of Nominal Wealth.” Journal of Political Economy, 114 (6): 1069–1097.
Jaravel, Xavier. 2021. “Inflation Inequality: Measurement, causes, and policy implications.” Annual Review of Economics, 13: 599–629.
Kehrig, Matthias, and Nicolas L. Ziebarth. „The effects of the real oil price on regional wage dispersion.” American Economic Journal: Macroeconomics, 9(2): 115–148.