Heterogeneous preferences, investment, and asset pricing

• Published date: 01 December 2021
• Author: Bo Liu,Lei Lu,Congming Mu,Jinqiang Yang
• Date: 01 December 2021
• DOI: http://doi.org/10.1111/fima.12350

DOI: 10.1111/fima.12350

ORIGINAL ARTICLE

Heterogeneous preferences, investment, and asset

pricing

Bo Liu1Lei Lu2Congming Mu3Jinqiang Yang4,5

1School of Management and Economics, Universityof Electronic Science and Technology of China, Chengdu, China

2Asper School of Business, University of Manitoba, Winnipeg, Manitoba,Canada

3College of Finance and Statistics, Hunan University, Changsha, China

4Shanghai Institute of International Finance and Economics, Shanghai, China

5School of Finance, Shanghai University of Finance and Economics, Shanghai, China

Correspondence

CongmingMu, College of Finance and Statis-

tics,Hunan University, Changsha, China.

Email:mucongming@sina.com

Fundinginformation

NationalNatural Science Foundation of China

(71573033),Fundamental Research Funds

forthe Central Universities, National Natu-

ralScience Foundation of China (72072108,

71972122,71772112, 71522008, 71472117),

FokYing-TongEducation Foundation of China

(#151086),Innovative Research Teamof

ShanghaiUniversity of Finance and Economics

(#2016110241)

Abstract

We present a production-based model in which agents

have heterogeneous risk aversion and heterogeneous dis-

count rate. Compared to the exchange economy, the aggre-

gate consumption-capital ratio and aggregate consumption

volatility is reduced. The risk premium and the volatility of

stock return increase when moving from the exchangeecon-

omyto the production economy. We also find that the volatil-

ity of Tobin’sqexhibits an inverted-U-shape and Tobin’s qis

procyclical.

1INTRODUCTION

The economy with heterogeneous agents has attractedacademic attention in the past few decades. Contrasted to the

representative agent economy,agents can optimally share risk in financial markets through lending/borrowing in the

economy with heterogeneous agents (e.g., Longstaff & Wang, 2012). When agents differ in their risk aversion, they

have different motives of precautionary savings and the less risk-averse agent borrows from the more risk-averse

agent. When agents have heterogeneous discount rate(e.g., Jouini & Napp, 2007; Warner & Pleeter, 2001), they have

a different saving decision and the less impatient agent lends to the more impatient agent (e.g., Brunnermeier & San-

nikov,2014;Paul,2020).As the wealth distribution among agents varies, the aggregate risk aversion and aggregate

discount ratealso vary over time, leading agents’ aggregate savings behavior to be stochastic and thus firms’ aggregate

© 2021 Financial Management Association International

Financial Management. 2021;50:1169–1193. wileyonlinelibrary.com/journal/fima 1169

1170 LIU ET AL.

investment and future output will be affected. In this paper,we study the effects of the heterogeneity in preferences

on firm’s investment and asset prices in the presence of production.

Tohighlight the importance of heterogeneous preferences, we incorporate agents’ heterogeneity in risk aversion

and discount rate into a production-based general equilibrium model (e.g., Cox et al., 1985; Eberly & Wang, 2011).

We consider three scenarios: (1) the economy with heterogeneity in risk aversion only,(2) the economy with hetero-

geneity in discount rate only, and (3) the economy with heterogeneity in both risk aversion and discount rate. We

focus on the economy with heterogeneity in risk aversion only and briefly discuss the other two cases. Toshow the

features of production economy that are different from those in exchange economy, we compare the economy with

finite adjustment cost to that with infinite adjustment cost. With infinite adjustment cost, the aggregate consumption

is equal to the aggregate output and the growth of capital stock is set to be zero. Thus, we interpret this economy as

an exchangeeconomy.

Wehave the following several findings when agents have heterogeneous risk aversion.First, the volatility of the less

risk-averse agent’s wealth share increases in the adjustment cost and thus it is lower in the production economy than

that in the exchangeeconomy. As the adjustment cost increases, it will be more costly for agents to adjust investment

for production and then it is more difficult for them to smooth consumption. The shocks to capital stock will have a

larger impact on agents’ consumption and thus they have more incentive to share risk by borrowing/lending, leading

to a higher volatility of wealth distribution.

Second, the presence of production reduces the aggregate consumption-capital ratio and aggregate consump-

tion volatility. Given a firm’s capital stock, agents find that it could be easier to smooth their future consumption

through physical investment in the firm, leading to a lower (current) aggregate consumption-capital ratio. Moreover,

the extra investment opportunities makeagents’ future consumption smoother (through adjusting firm investment)

and then the aggregate consumption becomes less volatile. As adjustment cost increases, it is more difficult for agents

to adjust their investment to smooth future consumption. Thus, the agents will reduce their physical investment in

the firm, and the aggregate (current) consumption-capital ratio increases and the volatilityof aggregate consumption

is enhanced. As a result, both the aggregate consumption-capital ratio and aggregate consumption volatility increase

in investment adjustment cost. In particular,the volatility of aggregate consumption in the production economy dis-

plays a U-shape with respect to the wealth share of the less risk-averse agent, while it is constant in the exchange

economy.

Third, given the less risk-averse agent’s wealth share, Tobin’sqincreases in adjustment cost. Higher adjustment

cost implies that it is more difficult for a firm to adjust its capital, and thus it is more valuable for its asset in place.

Given adjustment cost, Tobin’sqdecreases in the less risk-averse agent’s wealth share in the economy with infinite

adjustment cost, while it increases in the production economy. In particular,the volatility of Tobin’s qin the produc-

tion is positive and exhibits an inverted-U-shapemeaning that Tobin’s qis procyclical, while in the economy with infi-

nite adjustment cost, it is negative and displays a U-shape implying that Tobin’sqis counter-cyclical. Moreover, the

volatility of investment-capital ratio decreases in adjustment cost. This result is intuitive: As investmentadjustment

cost increases, it is more difficult for the firm to change its investment plan and then firm’s investment becomes less

dispersed.

Finally,given the less risk-averse agent’s wealth share, the interest rate increases when moving from the exchange

economy to the production economy. With the production, agents find it is easier to smooth their future consump-

tion through physical investment and thus reduce their saving incentives, leading to a higher interest rate. More-

over, given the less risk-averse agent’s wealth share, both risk premium and volatilityof stock return also increase

when moving from the exchange economy to the production economy. In our setting, the risk comes only from the

capital stock and thus the uncertainty in current capital stock generates long-term uncertainty on future invest-

ment. Therefore, agents require more compensation to hold stocks. In addition, the long-term uncertainty on cap-

ital stock generates a time-varying capital price and thus increases the volatility of stock return. Meanwhile, both

risk premium and volatility of stock return decrease with the adjustment cost in the production economy.As adjust-

ment cost increases, the expected growth rate of capital stock decreases, which reduces the long-run effect of the