Labor‐Technology Substitution: Implications for Asset Pricing

• Author: MIAO BEN ZHANG
• DOI: http://doi.org/10.1111/jofi.12766
• Date: 01 August 2019
• Published date: 01 August 2019

THE JOURNAL OF FINANCE •VOL. LXXIV, NO. 4 •AUGUST 2019

Labor-Technology Substitution: Implications for

Asset Pricing

MIAO BEN ZHANG∗

ABSTRACT

This paper studies the asset pricing implications of a firm’s opportunities to replace

routine-task labor with automation. I develop a model in which firms optimally under-

take such replacement when their productivity is low. Hence, firms with routine-task

labor maintain a replacement option that hedges their value against unfavorable

macroeconomic shocks and lowers their expected returns. Using establishment-level

occupational data, I construct a measure of firms’ share of routine-task labor. Com-

pared to their industry peers, firms with a higher share of routine-task labor (i) invest

more in machines and reduce more routine-task labor during economic downturns,

and (ii) have lower expected stock returns.

LABOR ECONOMISTS ARGUE THAT IN recent decades, automation has increasingly

replaced workers who perform procedural and rule-based tasks, that is, rou-

tine tasks. In addition, Jaimovich and Siu (2014) find that the disappearance of

routine-task jobs has occurred mainly during recessions, with such job disap-

pearance accounting for almost all job loss in the three most recent recessions.1

∗Miao Ben Zhang is at Marshall School of Business, University of Southern California. This

paper is based on the second chapter of my doctoral dissertation at UT Austin. I would like

to thank the members of my dissertation committee for their constant support and invaluable

guidance throughout: Aydogan Alti, Andres Donangelo (co-chair), Sheridan Titman (co-chair), and

Mindy Zhang. I thank David Autor; Jonathan Berk; Tom Chang; Jonathan Cohn; Wayne Ferson;

Cesare Fracassi; John Griffin; Jerry Hoberg; Chris Jones; Matthias Kehrig; George Korniotis;

Lars-Alexander Kuehn (discussant); Tim Landvoigt; Jun Li (discussant); Zack Liu; Vikram Nanda

(discussant); Robert Parrino; Vincenzo Quadrini; Jay Shanken; Stathis Tompaidis; Selale Tuzel;

Rossen Valkanov; Parth Venkat; Yuzhao Zhang (discussant); and seminar participants at Boston

College, Carnegie Mellon University, Emory University, INSEAD, HKU, UCLA, University of

Notre Dame, UNC Chapel Hill, University of Miami, University of Toronto, UNSW, USC, UT

Austin, UT Dallas, 2014 USC Marshall Ph.D. Conference in Finance, 2014 AFBC PhD Forum,

2016 WFA Meeting, 2016 SFS Finance Cavalcade, and 2016 CICF for helpful suggestions and

comments. I thank David Autor and Ryan Israelsen for sharing their data. Special thanks to Nir

Jaimovich, Minyu Peng, Stefan Nagel (Editor), an Associate Editor, and two anonymous referees.

This research was conducted with restricted access to Bureau of Labor Statistics (BLS) data. The

views expressed here do not necessarily reflect the views of the BLS. I thank Erin Good, Donald

Haughton, Jessica Helfand, Mark Loewenstein, and Michael Soloy at the BLS for their assistance

with the data. The author does not have potential conflicts of interest to disclose, as defined in the

Journal of Finance’s disclosure policy.

1Jaimovich and Siu (2014) show that routine-task jobs constitute 89%, 91%, and 94% of all

job loss in the 1990, 2001, and 2008 to 2009 recessions, respectively. Examples of routine-task

DOI: 10.1111/jofi.12766

1793

1794 The Journal of Finance R

Connecting these findings to a firm’s production, it seems that adopting ma-

chines to replace routine-task labor, that is, labor-technology substitution, is

an economically significant decision that varies with the business cycle. Such

state-contingent decisions can reflect important investment opportunities that

firms encounter.

In this paper, I study whether the opportunities for labor-technology substi-

tution are a source of macroeconomic risk that is priced in the cross-section

of stock returns. Compared to growth opportunities (Berk, Green, and Naik

(1999)), opportunities for labor-technology substitution have two distinctive

features in my model. First, labor-technology substitution features cost sav-

ing rather than scale expansion. Second, labor-technology substitution may

interrupt firm production. For example, adopting technologies is known to be

accompanied by plant restructuring (Cooper and Haltiwanger (2006)), worker

retraining (Atkin et al. (2017)), and organizational restructuring (Bresnahan,

Brynjolfsson, and Hitt (2002)), all of which are likely to interrupt firm pro-

duction. Given this interruption, firms optimally choose to switch technologies

when their productivity is low. Hence, if the economy experiences a negative

productivity shock, firms that have not yet switched technologies (due to their

superior productivity in the past) are able to do so. The increase in firm value re-

sulting from this switching acts as a hedge against negative shocks and lowers

firms’ risk premia. In other words, firms with a higher share of routine-task la-

bor maintain more abundant technology-switching options to hedge their value

against unfavorable aggregate shocks.

To study the empirical relation between routine-task labor and risk premia, I

construct a measure of a firm’s share of routine-task labor (RShare) using new

microdata from the Occupational Employment Statistics (OES) program of the

Bureau of Labor Statistics (BLS). The OES microdata provide employment

and wages for over 800 detailed occupations in 1.2 million establishments

in the United States over three-year cycles, covering 62% of total national

employment from 1990 to 2014. Following the labor economics literature (Autor

and Dorn (2013)), I classify occupations into routine-task labor and nonroutine-

task labor. I then define a firm’s RShare as the ratio of the total wages paid to

its routine-task labor relative to its total wage expense. I compare firms with

different RShare within industry sector to ensure that RShare does not simply

reflect the heterogeneous business models across industry sectors.

My measure of firms’ share of routine-task labor is correlated with a number

of firm characteristics in a manner that is consistent with my model. In the

data, high-RShare firms have a lower proportion of machines in their total capi-

tal than their industry peers with a low RShare. This relation is consistent with

my model’s assumption that routine-task labor and machines are substitutes.

High-RShare firms also have higher operating costs and higher operating lever-

age, which is consistent with the model’s assumption that routine-task labor

jobs over the past 30 years include clerks, production line assemblers, travel agents, bank tellers,

and tax preparers. See Acemoglu and Autor (2011) for a review of the literature on decreasing

routine-task jobs.

Labor-Technology Substitution 1795

is more costly to use than machines. Finally, high-RShare firms have higher

cash flows. This relation is consistent with the model’s implication that firms

with higher cash flows face a higher opportunity cost for switching technolo-

gies and thus are more likely to retain their routine-task labor than to switch

to machines.

The main empirical findings in this paper are twofold. First, I find strong

negative relations between firms’ RShare and their exposure to systematic

risk and expected stock returns. I use time-invariant and time-varying market

betas (Lewellen and Nagel (2006)) in the Capital Asset Pricing Model (CAPM)

as my main proxies for firms’ exposure to systematic risk. As alternative prox-

ies, I examine the beta of aggregate cash flow news and the beta of GDP growth.

I use future stock returns to proxy for firms’ expected returns. I find that sort-

ing portfolios of firms by RShare within industry generates a monotonically

decreasing pattern in both the betas and future excess returns. In all specifica-

tions, the betas of the high-RShare quintile portfolio are more than 20% lower

than those of the low-RShare quintile portfolio, suggesting that high-RShare

firms are less risky. In addition, comparing the high- and low-RShare quintile

portfolios yields a negative return spread of −3.9% per year.

My second main empirical finding is that, in response to unfavorable

aggregate economic shocks, high-RShare firms increase the extent of their

labor-technology substitution more than low-RShare firms. Specifically, when

I compare firms in the cross-section, I find that when GDP growth is low, high-

RShare firms, compared to their low-RShare industry peers, (1) invest more in

machines (especially in computers), (2) reduce more routine-task labor, and (3)

reduce even more routine-task labor if they invest more in machines. Tothe best

of my knowledge, this is the first empirical evidence to show that routine-task

labor is substituted by machines within firms during economic downturns.2

Moreover, these results suggest that high-RShare firms have more abundant

technology-switching options that can be exercised during economic downturns.

To strengthen the link between my model’s mechanism and the cross-

sectional risk premia, I test two additional predictions of the model. First,

I show that high-RShare firms cut operating costs more and lose less market

value than low-RShare firms over recessions. In contrast, I do not find such

differences between firms over expansions. Hence, high-RShare firms do seem

to be able to better hedge their value against unfavorable economic shocks

through exercising switching options, resulting in lower expected returns for

these firms. Second, my model predicts that high-RShare firms have higher

operating leverage, which offsets part of the hedging effect of the switching

option. Hence, if operating leverage is controlled for, high-RShare firms should

have even lower betas than low-RShare firms. I confirm this prediction in

tests that control for the linear and nonlinear associations between operating

2Most studies on routine-biased technological change use individual-level data such as the

Decennial Census data or the Current Population Survey data. These data have limitations in

linking individuals to firms. Hence, it is difficult for these studies to explore firms’ employment of

routine-task labor and investment in machines jointly to establish the substitution argument.