WAGE VOLATILITY AND CHANGING PATTERNS OF LABOR SUPPLY

• Published date: 01 May 2019
• Date: 01 May 2019
• Author: Jay H. Hong,Hye Mi You,Byoung Hoon Seok
• DOI: http://doi.org/10.1111/iere.12363

INTERNATIONAL ECONOMIC REVIEW

Vol. 60, No. 2, May 2019 DOI: 10.1111/iere.12363

WAGE VOLATILITY AND CHANGING PATTERNS OF LABOR SUPPLY∗

BYJAY H. HONG,BYOUNG HOON SEOK,AND HYE MIYOU1

Seoul National University, Korea; Ewha Womans University, Korea; Hanyang University, Korea

Over the past few decades, the skilled–unskilled hours differential for U.S. men increased when the skill

premium rose sharply, in contrast with dominant income effects. Based on PSID data, we show that over

the 1967–2000 period, skilled men experienced a three times larger increase in wage volatility than unskilled

men. With the rise in wage volatility, our general equilibrium incomplete markets model generates a 2.7 hours

increase in the hours differential whereas it increased by 1.4 hours in the data. We find that hours adjustments

are important for self-insurance in the short run, whereas precautionary savings play a crucial role eventually.

1. INTRODUCTION

The literature has documented that over the past four decades, the skill premium defined

by the relative wages of U.S. men with a college degree (skilled) compared to those without

a college degree (unskilled) increased substantially (Katz and Murphy, 1992; Krusell et al.,

1994; Autor et al., 2008). The substitution effect from the increase in the skill premium predicts

skilled men to increase their hours worked relative to unskilled men, whereas the income effect

works in the opposite direction. In a recent survey, Saez et al. (2012) writes: “At the margin,

substitution possibilities . . . can be captured by a compensated labor supply elasticity. With

some notable exceptions, the profession has settled on a value for this elasticity close to zero for

prime-age males.” This implies that the income effect possibly governs the response of hours

worked of skilled men relative to unskilled men to a rise in the skill premium,2that is, skilled

men who achieve higher relative wages likely reduce their work hours relative to unskilled men.

However, what we observe in the data is exactly the opposite. According to the Panel Study of

Income Dynamics (PSID), the difference in weekly hours worked between skilled and unskilled

men actually increased by about one hour between 1970 and 2000, whereas the skill premium

rose by 17% over the same period.

This article explores the changing patterns of the second moment of wages, that is, increased

wage volatility, to resolve this discrepancy. As wages become more volatile, individuals under

incomplete markets work longer hours as well as accumulate more precautionary savings for

self-insurance. Flod´

en (2006) shows that an increase in future wage uncertainty raises current

labor supply due to a precautionary motive in a simple two-period model. Pijoan-Mas (2006)

also emphasizes the impact of wage volatility on labor supply decisions by showing that the

level of both precautionary savings and labor supply are higher in an economy with incomplete

∗Manuscript received June 2015; revised May 2018.

1We want to thank editor Dirk Krueger and the three anonymous referees for their extremely insightful com-

ments and suggestions. We are also grateful to Mark Aguiar, Mark Bils, Yongsung Chang, and seminar participants

at various institutions for their invaluable comments. Kyooho Kwon and Jinhee Woo provided excellent research

assistance. This work is supported by the Research Resettlement Fund for the new faculty of Seoul National Uni-

versity. Please address correspondence to: Hye Mi You, College of Economics and Finance, Hanyang University,

222 Wangsimni-ro, Seongdong-gu, Seoul 04763, South Korea. Phone: +82-2-2220-2580. Fax: +82-2-2296-9587. E-mail:

hyemi.you@gmail.com.

2A recent paper by Boppart and Krusell (2016) also shows that a utility function that features a dominant income

effect is consistent with the long-run reduction in hours worked and changes in the other aggregate variables in the

United States.

595

C

(2018) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

596 HONG,SEOK,AND YOU

markets than in a complete-markets economy. If skilled men were exposed to larger increases in

wage volatility than unskilled men over the past few decades, then this channel can potentially

explain the observed rise in skilled–unskilled hours differential concurring with a rising skill

premium.

This article estimates the time path of wage volatility separately for skilled and unskilled

men using data from the PSID. We document marked differences in the trends of the second

moment—as well as the first moment—of wages across skill groups over the 1967–2000 period.

Our estimation shows that (i) wage volatility (measured by the variance of residual wages) has

risen for both skill groups between 1967 and 2000 and (ii) skilled men have experienced much

larger increases in wage volatility compared to unskilled men: The variance of residual wages

for skill men doubled between 1967 and 2000, whereas that for unskilled men increased by 37%

over the same period. The former is consistent with the findings in the related literature and the

latter is our contribution to the literature.3

To quantify the effect of the changing wage structure in explaining the skilled–unskilled hours

differential, we develop a general equilibrium incomplete markets model with heterogeneous

agents of different skill levels. Agents face uninsurable idiosyncratic wage shocks, which consist

of a persistent and a transitory component drawn from a skill-specific distribution. In particular,

the initial steady state of the model is calibrated to the 1967 U.S. economy. We then feed

in the estimated wage processes from the PSID data to quantify how much the absolute and

relative increase in wage volatility can explain the widening gap between skilled and unskilled

hours.

The model generates a 2.7-hour increase in the skilled–unskilled hours differential during the

transition, which is qualitatively consistent with the data. The greater rise in the volatility of

wage rates for skilled men causes them to increase their hours worked relatively more to build

up a larger stock of precautionary savings. This effect is strong enough to swamp the wealth

effect from the rise in the skill premium. The model quantitatively over-predicts the increase

in the relative labor supply of skilled men by 1.3 hours, which we believe is due to household

labor supply decisions missing in this study. We argue that hours adjustment is important for

self-insurance in the short run, whereas precautionary savings play a dominant role in the long

run. Once precautionary savings are built up, the wealth effect causes skilled men to reduce

their relative labor supply.

As a validity check, we assess the model’s implications for consumption and wealth. Based on

nondurables consumption data from the Consumer Expenditure Survey (CEX), we document

that the relative consumption of skilled to unskilled households continued to increase since the

early 1980s. This path is well in line with the relative consumption of skilled men generated by

the model. The model can also replicate the increasing trend in net worth of skilled households

relative to unskilled households in both the PSID and the CEX data. In addition, our model’s

implications for changes in the cross-sectional variation of hours worked and the correlation

between wages and hours are broadly consistent with what we observe in the data.

This article is closely related to Heathcote et al. (2010), who explore the implications of

observed changes in the wage structure (including a rising skill premium, a declining gender

gap, and the increasing overall residual wage dispersion) for cross-sectional inequality in hours,

earnings, and consumption. Our work differs from theirs in that we focus on the differences in

the evolution of wage volatility across different skill groups. We explore the quantitative effects

3Examining the causes for rises in wage volatility is outside the scope of this study, yet we view the increasing adoption

of performance–pay contracts and deunionization in the U.S. labor market as important factors behind the increases

in wage volatility. Lemieux et al. (2009) show that about 20% of the rise in variance of male log wages is associated

with performance–pay contracts and that performance–pay contracts have been widely adopted for managers and sales

personnel, as well as in finance, insurance, and real estate, all of which are regarded as skill intensive compared to

others. Card et al. (2004) claim that deunionization is an important contributor to the rise in male wage inequality.

Acikgoz and Kaymak (2014) explore skill-biased technical changes as a potential explanation behind the decline in the

U.S. unionization rate.

WAGE VOLATILITY AND LABOR SUPPLY 597

of the differences in increased wage volatility on the evolution of the relative hours worked of

skilled to unskilled men.

We can also relate our article to Erosa et al. (2016) and Castro and Coen-Pirani (2008), who

examine differences in male labor supply by skill groups. However, Erosa et al. (2016) focus on

their life-cycle features to study aggregate labor supply responses to changes in the economic

environment. Their main purpose is to quantify the aggregate responses of labor supply to wage

changes and evaluate the importance of the extensive versus intensive margins in this response.

Instead, our study aims to explain changing patterns of relative hours worked using differences

in the evolution of wage volatility across skill type. Our work also differs from Castro and

Coen-Pirani (2008), in that they explore demand-side factors to explain changes in business

cycle fluctuations of hours by skill groups,4whereas we consider supply-side factors to address

the trends in hours by skill levels.

Several recent studies, including Santos (2014), Elsby and Shapiro (2012), Michelacci and

Pijoan-Mas (2008), and Michelacci and Pijoan-Mas (2012), have tried to explain phenomena

related to our motivating facts using a different mechanism. These studies all explore the

effects of current experience/hours on future labor market outcomes as potential explanations

for the phenomena, although the details of the mechanism vary. Santos (2014) considers a

model in which current work hours affect future productivity in order to explain the increase

in the correlation between hours and wages for the last quarter of the 20th century in the

United States. He shows that in the CPS data, this dynamic effect has become stronger for

higher wage quintiles, whereas it has weakened for lower wage quintiles, leading to the recent

rise in the hours–wages correlation. Elsby and Shapiro (2012) focus on the extensive margin

of labor supply and explore changes in the returns to experience as driving forces behind

the changing patterns of employment/nonemployment rates by education. They find that the

return to experience for high school dropouts has fallen significantly since the 1970s, which

contributed to a downward trend in their employment rate. Michelacci and Pijoan-Mas (2008)

and Michelacci and Pijoan-Mas (2012) exploit a search-matching framework to explain the link

between wage inequality and hours worked. They assume that current hours of work affect

the future probability of getting outside job offers. In this framework, greater wage inequality

makes outside offers more attractive, inducing agents to work longer hours. Michelacci and

Pijoan-Mas (2008) use this mechanism to explain the divergence in hours worked between the

United States and Continental Europe, whereas Michelacci and Pijoan-Mas (2012) show that

this model is consistent with the positive correlation between the increase in wage inequality

and the rise in hours worked across occupation and industry groups in the U.S. census data. We

propose an alternative mechanism based on a self-insurance motive. Our work shows that the

evolution of the second moment of wages plays a quantitatively important role in explaining

the recent changes in U.S. male hours worked by skill groups.

The remainder of this article is organized as follows: Section 2 describes the stylized facts on

changes in the U.S. wage structure and the trends in hours worked by skill group. In Section 3,

we describe a general equilibrium model with heterogeneous agents and incomplete markets.

We then describe the calibration procedure in Section 4. Section 5 presents main quantitative

results, and Section 6 discusses several sensitivity analyses. Section 7 then concludes the article.

2. DATA

This section documents how the U.S. wage structure and hours worked have changed by skill

type between 1967 and 2011. Our data of wages and hours worked are drawn from the PSID

1968 through 2012. We also use the Current Population Survey (CPS) March Supplements over

this period to document the relative wages and hours worked of skilled and unskilled men for

comparison. To avoid issues associated with changing selection into labor force participation

4Castro and Coen-Pirani (2008) attempt to explain increased cyclicality of skilled hours since the mid-1980s and

propose changes in capital–skill complementarity as potential explanations.