The workforces of most high-income countries have been ageing at a rapid pace. The standard response to this has been to extend the careers of older workers. This column uses data from Italy to show how this may come at a cost to the careers of younger workers. Younger workers have faced increasing struggles to reach the upper segments of the wage distribution and higher-ranked jobs, in contrast to the experiences of older workers. The design of policies that increase the mean age of the workforce should consider these spillovers as they have important consequences over the lifecycle of workers.
The population and the workforce of most high-income countries have been ageing at a fast pace. In the US, for example, the share of workers aged 55 or older increased by 88%, from 12.9% in 1985 to 24.3% in 2020, marking the largest growth among all age groups (Toossi and Torpey 2017). In 2018, Japan became the country with the world’s oldest population, having the highest share of the population aged 65 or older (Stawasz et al. 2018). Similarly, in Italy, one of the main focuses of our empirical analysis, the mean worker age increased by 19% from 35.8 years old in 1985 to 42.7 years old in 2019.
The standard response to the greying of the population has been to extend the careers of older workers even further. Countries with extensive state-backed pension systems have often increased the minimum retirement age to ensure the long-term sustainability of the welfare state (Zweimuller et al. 2007, Bianchi et al. 2023). Similarly, employment protection laws usually entail stronger safeguards for older workers with high tenure to limit the risk that these workers could lose their jobs at a time in their life cycle in which re-employment opportunities are likely to be limited (Bellés-Obrero et al. 2022, Saez et al. 2024). Similarly, labour markets and firms may have responded to these demographic shifts by increasing the share of ‘age-friendly’ jobs for which older workers have a preference or a competitive advantage (Scott et al. 2022).
In general, all these responses have gone towards further increasing the supply of older workers in the labour markets of high-income countries. In Bianchi and Paradisi (2024), we explain why these changes may come at a cost for the careers of younger workers.
The rising pay inequality between older and younger workers
During the same period in which the workforce has been ageing, the wages of older workers have been increasing much more rapidly than those of younger workers in many high-income economies. For example, the pay gap between workers over 55 and those under 35 (from now on, the ‘age pay gap’) increased by 61% in the US between 1979 and 2018 and by 96% in Italy between 1985 and 2019 (Figure 1).
Figure 1 Age pay gap in log weekly wages, Italy
Notes: Figure 1 plots the gap between the log weekly wages of O55 workers and the log weekly wages of U35 workers from 1985 to 2019 for both mean and median wages. In each year, the data pool information about all workers who were at least 16 years old, had worked at least six months, had earned strictly positive wages, had full-time contracts, and had not retired by December 31.
Database: UNIEMENS, Istituto Nazionale della Previdenza Sociale (Italy).
We study how an increased stock of older workers can contribute to widening the age pay gap. It is a standard economic belief that younger and older workers are imperfect substitutes in the production function. Under this assumption, an increase in the relative supply of older workers should decrease the wages of older workers relative to those of younger workers, ultimately shrinking the age pay gap. However, since the age pay gap has instead widened, it has become evident that other factors must be at play.
Negative career spillovers in labour markets with limited slots
We propose a conceptual framework of ‘negative career spillovers’ to explain how the increased presence of older workers can harm the labour market outcomes of younger workers. Specifically, our model highlights the role of two features of the labour markets in high-income economies.
First, the model recognises that workers often receive rents for longer tenure at their firm – for example, in the form of stickier wages or fixed job allocations. Various factors can generate rents associated with longer tenure, such as backloaded wages (Ke et al. 2018), firm-specific human capital (Gathmann and Schönberg 2010), knowledge spillovers (Sandvik et al. 2020, Cornelissen et al. 2023), and layoff costs (Boeri and Jimeno 2005).
Second, in the model, some firms face constraints in adding higher-level positions. In most high-income economies, the labour markets have been experiencing declines in labour productivity, in GDP growth, and in firm dynamics. These factors can all contribute to firms facing increasing difficulties in expanding their ranks, especially at the top.
When the relative supply of older workers increases, these frictions imply that firms’ hierarchies are not always sufficiently flexible to create paths toward top jobs for all workers who are qualified to receive a promotion. Given that older workers’ wages are stickier, younger workers often bear the brunt of this market-level shock in the form of much slower career growth.
The data are consistent with the existence of negative career spillovers
We empirically test the key predictions of this model using matched employer-employee administrative data from Italy between 1985 and 2019. Moreover, we show that the main findings hold if we use more limited administrative data from Germany and labour force survey data for fourteen other high-income countries.
Younger workers have indeed faced increasing struggles to reach the upper segments of the wage distribution and higher-ranked jobs, in contrast to the trend experienced by older workers. In Italy, the likelihood of workers under 35 belonging to the top quartile of weekly wages declined by 34% from 1985 to 2019, while the probability for those over 55 increased by 32% (Figure 2).
Figure 2 Positions in the wage distribution, Italy
Panel A: U35 workers
Panel B: O55 workers
Notes: Panel A (B) shows the ratio between the share of U35 (O55) workers in each quartile in year t and the share of U35 (O55) workers in the same quartile in 1985. In each year, the data pool information about all workers who were at least 16 years old, had worked at least six months, had earned strictly positive wages, had full-time contracts, and had not retired by December 31.
Database: UNIEMENS, Istituto Nazionale della Previdenza Sociale (Italy).
As predicted by the model, the predominant driver of the widening age pay gap has been the opposite trajectories followed by younger and older workers along the wage distribution rather than changes in the level of wages paid in different parts of the distribution or for different jobs. We therefore compute the pay rank gap, which represents the portion of the age pay gap’s expansion attributable to variation in the positions of younger and older workers in the wage distribution while keeping the level of real wages in the economy fixed at baseline. The data indicate that the deterioration in the pay rank of younger workers and the improvement in that of older workers accounted for 78% of the total growth in the age pay gap.
Consistent with the increased concentration of older workers in higher-paying jobs, new entrants have entered the labour market progressively lower in the wage distribution (64% of the total loss in pay rank), and their position in the wage distribution has started growing more slowly for several years after entry.
Our analysis then documents the types of firms in which the age pay gap has widened the most. The data indicate that an increase in the number of older workers in top jobs is more detrimental to younger workers in firms with more limited opportunities to add higher-ranked positions: older, larger firms with lower employment growth have found it particularly challenging to map out satisfactory career trajectories for their younger workers, who have experienced higher-than-average positional losses in the pay distribution.
In addition to losing ground within firms, younger workers have been progressively concentrated among worse firms. In fact, older workers’ progressive entrenchment in higher-paying firms has reduced the probability of younger workers securing employment in these firms. Between 1985 and 2019, the share of younger workers decreased by 2 percentage points (a 26% loss relative to the 1985 level) among firms in the top decile of mean pay and increased by 3 percentage points among firms in the bottom decile (Figure 3).
Figure 3 Distribution of workers between firms, Italy
Panel A: Share of O55 workers
Panel B: Share of U35 Workers
Notes: Firm groups are one hundred groups that have the same number of workers and are ordered in ascending mean firm pay. Panel A (Panel B) shows the change in O55 (U35) workers for each firm group from 1985 to 2019. To limit noise, the displayed values are averages computed using each group and its two adjacent firm groups (one in the case of the first and last group). In each year, the data pool information about all workers who were at least 16 years old, had worked at least six months, had earned strictly positive wages, had full-time contracts, and had not retired by December 31.
Database: UNIEMENS, Istituto Nazionale della Previdenza Sociale (Italy).
These spillovers are important for policymaking
Our theoretical and empirical findings indicate that having more older workers in the labour market can harm the career outcomes of younger workers, at least in a context in which higher-paying slots are rationed. The design of policies that increase the mean age of the workforce should consider these spillovers because they have important consequences over the lifecycle of workers. For example, lower early-career wages may deter some workers from purchasing durables, investing in real estate, or starting families – decisions that cannot always be deferred until later career stages.
Source : VOXeu
