Income Redistribution, Consumer Credit, and Keeping Up with the Riches

• Published date: 01 December 2020
• Date: 01 December 2020
• Author: MATHIAS KLEIN,CHRISTOPHER KRAUSE
• DOI: http://doi.org/10.1111/jmcb.12690

DOI: 10.1111/jmcb.12690

MATHIAS KLEIN

CHRISTOPHER KRAUSE

Income Redistribution, Consumer Credit, and

Keeping Up with the Riches

In this study, we set up a dynamic stochastic general equilibrium (DSGE)

model with upward looking consumption comparison and show that con-

sumption externalities are an important driver of consumer credit dynam-

ics. Our model economy is populated by two different household types. In-

vestors, who hold the economy’s capital stock, own the rms and supply

credit, and workers, who supply labor and demand credit to nance con-

sumption. Furthermore, workers condition their consumption choice on the

investors’level of consumption. We estimate the model and nd a signicant

keeping up mechanism by matching business cycle statistics. In reproduc-

ing credit moments, our proposed model signicantly outperforms a model

version in which we abstract from consumption externalities.

JEL codes: E21, E32, E44

Keywords: income redistribution, consumer credit, relativeconsumption

motive, business cycles

T     of consumption

externalities between different income groups for replicating consumer credit dynam-

ics over the business cycle. Forthis purpose, we propose a dynamic stochastic general

We thank the editor Sanjay K. Chugh and two anonymous referees for helpful comments and sug-

gestions. This paper has beneted greatly from comments made by participants at many conferences and

seminars, in particular from those of Ludger Linnemann, Roland Winkler, WolfgangLeininger, Wolfram

Richter, Philip Jung, Johannes Brumm, KenJudd, Stefan Wild, Michal Marenˇ

cák, Gregor von Schweinitz,

Thomas Krause, Geraldine Dany,and Manuel Buchholz. The opinions expressed in this article are the sole

responsibility of the authors and should not be interpreted as reecting the views of Sveriges Riksbank.

Open Access funding enabled and organized by Projekt DEAL.

Correction added on 8 March 2021, after rst online publication: An afliation to Karlsruher Institut für

Technologie (KIT) has been added for Christopher Krause.

M K is a research economist at Sveriges Riksbank (E-mail: Mathias.Klein@riksbank.se).

C K is a post-doctoral researcherat Karlsruher Institut für Technologie (KIT) (E-mail:

christopher.krause@kit.edu).

Received October 1, 2018; and accepted in revised form August 5, 2019.

Journal of Money, Credit and Banking, Vol. 52, No. 8 (December 2020)

© 2020 The Authors. Journal of Money, Credit and Banking published by Wiley Periodicals

LLC on behalf of Ohio State University

This is an open access article under the terms of the Creative Commons Attribution-NonCom-

mercial-NoDerivs License, which permits use and distribution in any medium, provided the

original work is properly cited, the use is non-commercial and no modications or adaptations

are made.

1938 :MONEY,CREDIT AND BANKING

equilibrium (DSGE) model with upward looking consumption comparison that suc-

cessfully reproduces credit movements during the Great Moderation. We estimate

deep model parameters and thereby contribute to the literature as we show that con-

sumption externalities are a signicant determinant of short-run credit uctuations.

Recent empirical studies show that consumption externalities signicantly affect

individuals’ consumption decisions. Bertrand and Morse (2016) nd empirical sup-

port for so-called “trickle-down consumption,” meaning that rising income and con-

sumption at the top of the income distribution induces households in the lower parts

of the distribution to consume a larger share of their income. Focusing on the pe-

riod between the early 1980s and 2008, the authors present evidence for a negative

relationship between income inequality and the savings rate of middle-income house-

holds. Carr and Jayadev (2015) show that rising indebtedness of U.S. households is

directly related to high levels of income inequality. The authors conclude that rela-

tive income concerns explain a signicant part of the strong increase in household

debt for the period 1999–2009. Using data from the German Socio-Economic Panel,

Drechsel-Grau and Schmid (2014) demonstrate that upward looking comparison is a

signicant determinant of individuals’ consumption decisions.

Regarding the interrelation between consumption externalities and privatedebt dy-

namics, there is yet no conclusive evidence. Bertrand and Morse (2016) provide in-

direct evidence that nonrich households rely on easier credit access to nance their

desired keeping up with richer coresidents. Moreover, they nd a positive relation-

ship between the number of personal bankruptcy lings and top income levels. Geor-

garakos, Haliassos, and Pasini (2014) show that a higher average income increases

the tendency to borrow of households with incomes belowaverage. Contrary, Coibion

et al. (2014) nd that low-income households in high-inequality regions accumulate

less debt than similar households in low-inequality regions. However, their ndings

are mainly driven by mortgages, whereas for our variableof interest, consumer credit,

the authors only nd mixed results. Against this background, we investigate this re-

lationship within a structural model and show that relative consumption concerns are

an essential driver of aggregate credit dynamics.

Understanding how unsecured consumer credit uctuates over the business cycle

is of central importance because of several reasons. First, consumer credit is an im-

portant source of personal nance. For our period of interest, the Great Moderation,1

credit averages 23% of aggregate personal consumption in the United States, indi-

cating that more than one-fth of households’ private expenditures were nanced by

relying on consumer credit.2Second, short-run credit movements in the United States

1. Following Bertrand and Morse (2016) and Iacoviello and Pavan (2013), among others, we date

the Great Moderation as the time span between the early 1980s (here 1982 q1) and the outburst of the

nancial crisis (2008 q2). We choose the Great Moderation as the underlying time span, because this

period is characterized by a signicant widening of income disparities and several innovationsin nancial

markets, which ultimately made credit access for households easier. Notably, all our qualitative ndings

are robust when extending the sample by the Great Recession.

2. Our consumer credit measure includes revolving and nonrevolvingcredit. Revolving credit primar-

ily consists of outstanding credit card balances and accounts for roughly one-third of aggregate consumer

MATHIASKLEIN AND CHRISTOPHER KRAUSE :1939

TAB LE 1

C-R M I T U.S. (19821-20082)

ρ(xt,Dt)σx/σD

Output 0.1523 0.4568

Consumption 0.1658 0.2783

Investment 0.0852 1.7524

Hours worked 0.3603 0.5080

Real wage −0.3207 0.3994

N: ρ(xt,Dt) is the cross-correlation of variable xwith credit, σx/σDis the std. deviation of variablexrelative to the std. deviation of

credit. Consumer credit has been deated using the price index of personal consumption expenditures. All variables are logged and HP-

ltered (smoothing parameter of 1600) to obtain cyclical components. For data denitions and sources, see Online Appendix A.

are characterized by a highly volatile behavior.As Table 1 reports, credit is more than

twice (three times) as volatile as output (consumption). Third, and most importantly,

business cycle correlations with other main aggregate variables contradict standard

theory in which credit represents an instrument to smooth consumption in bad times.

Table 1 shows positive comovements between credit and output and consumption,

respectively. In contrast to these empirical observations, one would expect counter-

cyclical correlations when credit is primarily used to smooth consumption.

Our proposed model economy is populated by two types of households. Investors,

who hold the economy’s entire capital stock, own rms, and supply credit, and work-

ers, who supply labor and demand credit to nance their desired level of consump-

tion. Moreover, we include a mechanism through which workers value their own

level of consumption relative to the investors’ level of consumption. We refer to

this mechanism as keeping up with the Riches.3This extension allows us to capture

the “trickle-down-consumption” channel of Bertrand and Morse (2016), where the

income-poor try to catch up with the income-rich. In the baseline model, uctuations

are driven by four stochastic innovations, namely, a neutral technology, investment-

specic technology, price markup, and wage markup shock. In standard DSGE mod-

els, both technology shocks are main drivers of uctuations in real variables (Smets

and Wouters 2003, Justiniano, Primiceri, and Tambalotti 2010). Although in general

markup shocks play a minor role in driving output dynamics, we stress their impor-

tance in driving credit movements. In particular, in our model setup both innovations

shift resources between investors and workers, which in combination with consump-

tion externalities amplies the credit changes compared to a framework that abstracts

from these externalities.

credit. Nonrevolving credit includes auto loans as well as consumer installment loans. For a detailed anal-

ysis of consumer credit moments across categories and sample periods, we refer the interested reader to

Fieldhouse, Livshits, and MacGee (2016).

3. This term is inspired by the literature on keeping up with the Joneses. While studies that incorporate

this mechanism model relativeconsumption concerns in relation to the average consumer (e.g., Galí 1994),

in our setup poorer households (workers) aim to keep up with richer ones (investors).