Han, Yu
(2019).
The Roles of Demographic Changes on Labor Market Dynamics and Consumption Inequality.
PhD thesis, Universität zu Köln.
|
PDF
Thesis_CMR_YuHan_A5_for_Library.pdf Download (1MB) | Preview |
Abstract
This thesis consists of three essays on the impacts of demographic changes on labor markets and consumption goods markets. More specifically, the first two essays examine the role of demographic changes on labor market dynamics with empirical evidence and a theoretical framework, respectively. The third essay investigates the differential impacts of demographic changes on the consumption of retirees and workers. The impacts on labor market dynamics come from the substantial changes in the demographic composition of labor markets, which happen among the vast majority of developed economies. In the U.S., for instance, the share of young workers (aged 15–24 years) in the labor force, let us name it the youth share afterward, has increased from 19% since 1960s to 25% by the late 1970s due to the postwar baby boom. Recently, this value decreased to less than 15% as the inflow of young workers dwindled. In Japan, this value declined from 23% in the late 1960s to less than 10% currently. If workers at different age levels are disproportionately affected by the business cycle, then changes in the youth share will have a direct composition effect on aggregate labor market dynamics, even if age-specific labor market dynamics remain unchanged during demographic changes. It is often thought that young workers would be more easily affected by the business cycle, as they lose jobs more often and their (un)employment rate is generally more cyclically sensitive. In the first essay, I first show that this guess is supported by data. In particular, the labor market volatility of young workers, in terms of unemployment volatility, is significantly higher than those of other age groups, and the corresponding values of prime-age and old workers are relatively close. This age-specific difference points out the importance of demographic changes in accounting for aggregate business cycle volatility. Besides, I observe that there is a hump-shaped pattern in the youth share in the U.S. from 1960 to 2007, at the same time, aggregate unemployment volatility comoves with this youth share over time. More importantly, I find an even closer comovement between the unemployment volatility of young workers and this youth share. This suggests a new channel: Unemployment volatility within the group of young workers increases with their share in the labor force. I then test empirically this new channel. Using an unbalanced panel of 20 OECD countries from 1960 to 2007, I find that the variation in the youth share has a quantitatively large and statistically significant, positive effect on the unemployment volatility of young workers. The causality of this relationship is identified by the exogeneity of the youth share, as it is predetermined at least 15 years prior. I refer to this novel fact as a spillover effect. To quantify the relative importance of this spillover effect with respect to the composition effect as mentioned before, which is proposed as the only demographic explanation for business cycle volatility by Jaimovich and Siu (2009), I decompose the overall effect of demographic changes on aggregate unemployment volatility. In contrast to Jaimovich and Siu (2009), I find that the spillover effect accounts for most of the effect of demographic changes. In accounting for “The Great Moderation”—the substantial decline in cyclical volatility experienced in the U.S. since the mid-1980s, I find that demographic changes can explain one quarter of the decline in unemployment volatility. Of this, the spillover effect accounts for two-thirds of the overall effect, while the composition effect accounts for only a third. In the second essay, I explore a potential explanation of this novel fact. I argue that an increase in the youth share depresses the price of goods produced by young workers. This generates a congestion in their labor market dynamics and raise their cyclical unemployment volatility. To be specific, I build on Jaimovich, Pruitt, and Siu (2013) by incorporating a job matching model with endogenous job separation into the real business cycle model, to connect labor demand with the dynamics of labor market transition rates. I start by distincting goods produced by young and old workers. Then, these two distinct goods are combined with capital for the production of final goods. Therefore, this distinction generates a labor demand structure that differentiates between young and old workers. With this labor demand structure, a greater share of young workers is associated with a lower relative price of goods produced by them, lowering firms' profits. This induces spillover effects in labor market dynamics, both in the job separation and finding rates. For the former, the volatility of the job separation rate increases with the youth share, because young workers become more vulnerable to productivity shocks. For the latter, the volatility of job finding rate also increases with the youth share, since firms’ vacancy posting also becomes more sensitive to productivity shocks, which works the same way as Hagedorn and Manovskii (2008). Therefore, both aspects imply that the unemployment volatility of young workers increases with the youth share. As corroborating evidence for this mechanism, I find that the decline in the relative price of goods produced by young workers due to a greater share of young workers is supported by empirical data. Further evidence comes from the empirical dynamics of transition rates: The second moments of the transition rates of young workers are higher if the share of young workers is higher, and vice versa. This empirical pattern is exactly the same as my model predicts. The dependence of the labor market dynamics of young workers on demographic structure have important policy implications. First, it points out the necessity for policy-makers to pay attention to the current state of demographic structure and its projected path, if they need information about the labor market response to potential shocks, especially for young workers. Second, the positive correlation between business cycle volatility and demographic changes also suggests that the Great Moderation was more likely driven by structural factors other than good policies. The third essay focuses on consumption goods markets, by assessing the differential impacts of demographic changes on the consumption of retirees and workers. The research topic is of particular interest, not only because of the increasing income and wealth inequality in industrialized countries in the past few decades (Piketty, 2015), but also because these countries will experience considerable growths in their older population while significant declines in their working population. In the U.S., the dependency ratio—the number of population over 65 years to working age population, is expected to grow from 0.22 in 2015 to 0.40 in 2060. Some other countries have already stepped into the society of aging. In Japan, the dependency ratio is already up to 0.47 in 2017, almost three times as much as that in 1990. The current ratios for Italy, Germany, and France are all over 0.3. Using a New Keynesian model featuring the life-cycle behavior of Gertler (1999), I predict an increase in consumption inequality between retirees and workers in the U.S., measured by the ratio of consumption per retiree to consumption per worker, using its projected demographic evolution. In my calibrated model, this ratio is predicted to decline by 40% for the U.S., from 0.68 to 0.41, between 1990 and 2060, due to population aging. As the zero lower bound will bind more frequently with an older population, in addition, I also investigate the distributional effects of the zero lower bound during cyclical downturns. I find that the presence of the zero lower bound can mitigate the asymmetric effects of shocks on workers and retirees in dynamics, because there is a lower decline in the real return on assets, which particularly benefits retirees.
| Item Type: | Thesis (PhD thesis) | ||||||||
| Creators: |
|
||||||||
| URN: | urn:nbn:de:hbz:38-94366 | ||||||||
| Date: | 11 March 2019 | ||||||||
| Language: | English | ||||||||
| Faculty: | Faculty of Management, Economy and Social Sciences | ||||||||
| Divisions: | Weitere Institute, Arbeits- und Forschungsgruppen > Center for Macroeconomic Research (CMR) | ||||||||
| Subjects: | Economics | ||||||||
| Uncontrolled Keywords: |
|
||||||||
| Date of oral exam: | 11 March 2019 | ||||||||
| Referee: |
|
||||||||
| Refereed: | Yes | ||||||||
| URI: | http://kups.ub.uni-koeln.de/id/eprint/9436 |
Downloads
Downloads per month over past year
Export
Actions (login required)
 |
View Item |