Cultural Determinants of Household Saving Behavior

• Published date: 01 August 2020
• Author: NICOLA FUCHS‐SCHÜNDELN,PAOLO MASELLA,HANNAH PAULE‐PALUDKIEWICZ
• Date: 01 August 2020
• DOI: http://doi.org/10.1111/jmcb.12659

DOI: 10.1111/jmcb.12659

NICOLA FUCHS-SCH ¨

UNDELN

PAOLO MASELLA

HANNAH PAULE-PALUDKIEWICZ

Cultural Determinants of Household

Saving Behavior

Relying on the epidemiological approach, we show that culture is a signif-

icant driver of household saving behavior. Second-generation immigrants

from countries that put strong emphasis on thrift or wealth accumulation

tend to save more in Germany. We confirm these results in data from the

United Kingdom. By linking parents to their children, we show that these

two cultural components affect the saving behavior of both first-generation

immigrants and their children.

JEL codes: D14, Z10

Keywords: saving, culture, immigrants.

CULTURE SIGNIFICANTLY INFLUENCES HOUSEHOLD BEHAVIOR.A

vast literature finds that cultural components matter for decisions concerning fe-

male labor supply (e.g., Fern´

andez 2007, Alesina and Giuliano 2010, Blau 2015),

male employment (Moriconi and Peri Forthcoming), fertility (Fern´

andez and Fogli

2009), and family living arrangements (e.g., Giuliano 2007, Furtado, Marc´

en, and

Sevilla 2013). Evidence on the link between culture and household saving behavior

is, however, still scarce and inconclusive, possibly due to data limitations. Carroll,

We thank conference and seminar participants at the European UniversityInstitute, Goethe University

Frankfurt, University College London, Center for European Economic Research (ZEW), Annual Meeting

of the Society for Economic Dynamics (2017), Workshop on Microeconomics (Leuphana University,

2017), Spring Meeting of Young Economists (2016), the International German Socio-Economic Panel

User Conference (2016), the Annual Meeting of the German Economic Association (2016), and the IZA

Summer School in Labor Economics (2017) for helpful comments. We also thank Friedrich Scheller for

sharing Stata code.

NICOLA FUCHS-SCH¨

UNDELN is at Goethe University Frankfurt and CEPR (E-mail: fuchs@wiwi.uni-

frankfurt.de). PAOLO MASELLA is at University of Bologna (E-mail: p.masella@unibo.it). HANNAH PAULE-

PALUDKIEWICZ is at Goethe University Frankfurt(E-mail: hannah.paule@hof.uni-frankfurt.de).

Received December 4, 2017; and accepted in revised form April 3, 2019.

Journal of Money, Credit and Banking, Vol. 52, No. 5 (August 2020)

C

2019 The Ohio State University

1036 :MONEY,CREDIT AND BANKING

Rhee, and Rhee (1994, 1999) do not find evidence that differences in saving rates

between cultural groups can be attributed to cultural factors. Yet, recent papers

find cultural effects on portfolio allocation decisions (Guiso, Sapienza, and Zin-

gales 2004, 2008, Osili and Paulson 2008, Haliassos, Jansson, and Karabulut 2017).

Moreover, it has been shown that the individual experience with the macroeconomic

and financial environment significantly influences households’ financial decision

making through the formation of preferences and beliefs such as risk preferences

and trust (Malmendier and Nagel 2011, 2015, Osili and Paulson 2014, Kn¨

upfer,

Rantapuska, and Sarvim¨

aki 2017). It thus seems surprising that household savings—

the decision that precedes the portfolio choice—should not be affected by cultural

elements.

We revisitthe question whether culture affects household saving behavior by rely-

ing on household data that provide measures of the propensity to save and information

about the cultural origin of household members and their ancestors, and by resorting

to several worldwide surveys containing rich arrays of attitudinal variables that can

be used to more precisely isolate what cultural components explain variation across

households in saving rates. We follow Guiso, Sapienza, and Zingales (2006) and

define culture as “those customary beliefs and values that ethnic, religious, and social

groups transmit fairly unchanged from generation to generation.”

We adopt the so-called epidemiological approach in order to identify the effect of

culture on savings (Fern´

andez 2008). The epidemiological approach links the behav-

ior of second-generation migrants, who all face common economic and institutional

factors in the host country, to attitudes and behavior of individualsin the home coun-

try of their ancestors. If culture is persistent, second-generation immigrants should

still possess values and beliefs of their parents’ home country culture. Differences

in the saving behavior of second-generation immigrants of different countries of an-

cestry that correlate with cultural attitudes in their country of ancestry are hence an

indicator of cultural effects. We analyze the saving behavior of second-generation

immigrants in Germany,1additionallyproviding supporting evidence from the United

Kingdom.2

The epidemiological literature links attitudes or behavior of second-generation

immigrants to either attitudes (e.g., Luttmer and Singhal 2011) or behavior (e.g.,

Fern´

andez and Fogli 2009) of individuals in their country of ancestry. While we

present results linking behavior of second-generation immigrants to both attitudes

and behavior of individuals in their country of ancestry, we believe that linking it to

attitudes is more appropriate when studying saving decisions because of data issues

concerning the availability and comparability of national saving rates. We thus rely

in our main analysis on attitudinal variables from the countries of ancestry. With

reference to the economic-psychological literature (W¨

arneryd 1999), we distinguish

1. Paule (2012) analyzes the saving behavior of first-generation German immigrants. In contrast, we

study second-generation immigrants, whose saving behavior is less likely to be affected by immigration

related issues.

2. We are not aware of any U.S. data set that provides detailed information on household saving

choices and makes it possible to identify second-generation immigrants and their country of ancestry.

NICOLA FUCHS-SCH ¨

UNDELN, PAOLOMASELLA, AND HANNAH PAULE-PALUDKIEWICZ :1037

between five different cultural aspects that may affect saving behavior. These are (i)

thrift, (ii) the prevalent desire for effective accumulation and improvement, (iii) self-

control, (iv) attitudes toward the future (long-term versus short-term orientation),

and (v) reaction to uncertainty. To measure these different cultural aspects at the

home country level, we construct cultural proxy variables based on some of the

most widely recognized cultural cross-country data sets, namely, the World Values

Survey/European Values Study (WVS/EVS), the Hofstede data set, and the Schwartz

Value Survey (SVS).

While we focus on Germany in our main analysis, we confirm our results using data

from the United Kingdom. For both Germany and the United Kingdom, we identify

significant effects of culture on savings. The two cultural motivesthat we persistently

find to matter are, first, the desire for accumulation, which we measure by the

importance assigned to wealth, material possession, and money; and, second, attitudes

toward thrift. Both variables, measured at the level of the country of ancestry, have a

significant effect on the saving behavior of second-generation migrants. These effects

are robust in a series of checks including different sample adjustments, alternative

estimation techniques, and the inclusion of additional control variables. Our estimates

suggest that, holding everything else equal, immigrants from a country at the 75th

percentile in the distribution of the attitudes toward thrift have a 1.3 percentage

points higher saving rate than immigrants from the country at the 25th percentile,

and a similar comparison relating to the distribution of the desire for accumulation

suggests a 0.9 percentage points higher saving rate.

One major advantage of our study compared to other papers from the epidemi-

ological literature is that based on our panel data we can link second-generation

immigrants to their parents and study the saving behavior of both. By exploiting the

household survey structure and directly linking parents to their children, we confirm

significant effects of attitudes concerning wealth accumulation and thrift on both

the saving behavior of parents and their children, and also document some assimi-

lation. Moreover, we find that long-term orientation measured in the home country

significantly affects the saving behavior of parents, but only the saving behavior of

second-generation migrants who speak the parental language well. Finally,we find no

robust relationship between saving behavior and self-control or attitudes toward un-

certainty, respectively, from the home country for neither second-generation migrants

nor their parents. This indicates that either the measures we are using to measure self-

control and attitude toward uncertainty are inappropriate to capture these cultural

aspects, or there exist no cultural effects of these channels on saving behavior, while

the attitudes might still matter on the individual level.

This paper contributes to the growing literature that assesses the importance of

culture for various economic outcomes, and in particular to studies that analyze

the effect of culture on household saving behavior. Using Canadian household data,

Carroll, Rhee, and Rhee (1994) provide a first exploration of the relationship between

culture and the propensity to save of households. While they find that the saving

behavior of immigrants differs significantly from the saving behavior of natives,they

do not find significant differences in the saving behavior of immigrants coming from