ArticlePDF Available

The role of risk attitudes and expectations in household borrowing: evidence from Estonia

Authors:

Abstract and Figures

This study investigates the role of risk attitudes and financial expectations in households’ borrowing behaviour. The central research question is whether risk aversion and optimistic expectations provide additional information beyond the main economic and sociodemographic characteristics in predicting applications for credit and the size of debt. The paper uses microdata from the Estonian Household Finance and Consumption Survey (HFCS) and estimates probit and Heckman models. My analysis shows that risk-tolerant households apply for loans more often than risk-averse households do and that their loans are larger. The variables describing the household's expectations for its future financial situation are on their own related to the decision to apply for a loan, but they do not contain any relevant additional information beyond the main economic and sociodemographic characteristics of the household.
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rbec20
Baltic Journal of Economics
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rbec20
The role of risk attitudes and expectations in
household borrowing: evidence from Estonia
Eva Branten
To cite this article: Eva Branten (2022) The role of risk attitudes and expectations in
household borrowing: evidence from Estonia, Baltic Journal of Economics, 22:2, 126-145, DOI:
10.1080/1406099X.2022.2112485
To link to this article: https://doi.org/10.1080/1406099X.2022.2112485
© 2022 The Author(s). Published by Informa
UK Limited, trading as Taylor & Francis
Group
Published online: 23 Aug 2022.
Submit your article to this journal
Article views: 182
View related articles
View Crossmark data
The role of risk attitudes and expectations in household
borrowing: evidence from Estonia
Eva Branten
Department of Economics and Finance, Tallinna Tehnikaulikool, Tallinn, Estonia
ABSTRACT
This study investigates the role of risk attitudes and nancial
expectations in householdsborrowing behaviour. The central
research question is whether risk aversion and optimistic
expectations provide additional information beyond the main
economic and sociodemographic characteristics in predicting
applications for credit and the size of debt. The paper uses
microdata from the Estonian Household Finance and
Consumption Survey (HFCS) and estimates probit and Heckman
models. My analysis shows that risk-tolerant households apply for
loans more often than risk-averse households do and that their
loans are larger. The variables describing the households
expectations for its future nancial situation are on their own
related to the decision to apply for a loan, but they do not
contain any relevant additional information beyond the main
economic and sociodemographic characteristics of the household.
ARTICLE HISTORY
Received 1 March 2022
Accepted 9 August 2022
KEYWORDS
household debt; mortgage
loans; non-mortgage loans;
borrowing decisions; income
and price expectations; risk
attitudes; Household Finance
and Consumption Survey
JEL CODES
G51; D14
1. Introduction
1
This paper studies the relations between risk attitudes and nancial expectations and
dierent aspects of borrowing by households in Estonia. The aim of the study is to
assess whether risk aversion and optimistic expectations of households about their real
income and house prices provide additional information beyond that given by the
main economic and sociodemographic characteristics in predicting borrowing behaviour.
Data from the 2013 and 2017 waves of the Estonian Household Finance and Consumption
Survey are employed for the analysis.
It is important to assess these relationships since over-optimism and willingness to take
on too much risk may lead to excessive borrowing, which could entail a risk to the house-
holds themselves and to overall nancial stability. The knowledge gained through the
study could be of use in predicting borrowing behaviour and credit growth, and this
could be of interest for nancial stability analysis.
The theoretical foundations of this study are based on the Life-cycle/Permanent
Income Hypothesis (Modigliani and Brumberg (1954), Friedman (1957)), which suggests
© 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/
licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
CONTACT Eva Branten eva.branten@ttu.ee Department of Economics and Finance, Tallinna Tehnikaulikool,
Estonia
BALTIC JOURNAL OF ECONOMICS
2022, VOL. 22, NO. 2, 126145
https://doi.org/10.1080/1406099X.2022.2112485
that consumers aim to hold their consumption levels steady over time at the level of their
permanent income by borrowing and saving. Consumption smoothing provides a link
between nancial expectations and borrowing, while risk averseness aects the shape
of the utility function. Brown et al. (2005) describe two versions of the life-cycle model,
showing there to be a positive relationship between optimistic expectations for future
income and the size of a loan, and Brown et al. (2013) use a life-cycle model to show theor-
etically how risk aversion impacts the size of debt. The models by Brown et al. (2005) and
Brown et al. (2013) provide the theoretical motivation for this study.
My study relates to a large strand of literature studying the determinants of household
debt. Data from the Eurosystem Household Finance and Consumption Survey have been
used by Bover et al. (2014) to investigate dierent measures of debt, and by Du Caju et al.
(2016) to study how dierent household-level characteristics aect the likelihood of a
household being over-indebted. The nancial vulnerability of euro area households has
been analysed using the Household Finance and Consumption Survey data by
Ampudia et al. (2016), while Terraneo (2018) employs HFCS data to analyse the
nancial fragility of households in Southern Europe.
In the literature discussing the relations between householdsexpectations and their
economic behaviour, relatively few studies have focused on household borrowing. My
study is similar in the object of its research and its use of microdata to the studies of
Brown et al. (2005) and Brown et al. (2008), but there are some additional aspects of bor-
rowing that are explored. Brown et al. (2005) and Brown et al. (2008)nd that optimistic
nancial expectations are positively related to the size of non-mortgage debt and mort-
gage debt of households, respectively.
The risk preferences of households have been highlighted in several studies as a factor
related to household debt. The importance of risk attitudes in explaining household
indebtedness in Southern European countries has been pointed out using data from
the Household Finance and Consumption Survey by Massó and Abalde (2020). My
study also shares similarities with the study by Brown et al. (2013) in that it analyses mort-
gage loans and non-mortgage loans separately. Brown et al. (2013) show from US data
that aversion to risk is negatively related to the level of debt held by the household.
Compared to previous studies, my study contributes to a greater understanding of
household nancial behaviour by looking at more aspects of the borrowing of households
in relation to their expectations and risk attitudes, such as the decision of whether to
apply for a loan or not, the size of debt they decide to take on, and the resulting debt
service burden. My study also contributes by applying models that use the panel com-
ponent in the dataset and entail a logic that the values of the explanatory variables are
from the previous survey wave, compared to the dependent variables. The rationale
behind this is the assumption that the situation for the loan liabilities of households in
2017 survey was aected by the decisions taken previously, which are assumed to be
aected by the past values of dierent explanatory variables.
There are relatively few studies on how attitudes to nancial risk have formed in
Eastern European countries and the Baltic states (e.g. Dohmen et al. (2016), Pońsko
(2018)) and how these aect the nancial behaviour of households. Estonias case is inter-
esting since the credit and nancial markets in Estonia are relatively young and house-
holds have had relatively little experience of nancial risk-taking. Households borrowed
actively before the global nancial crisis and have also done so since a few years after
BALTIC JOURNAL OF ECONOMICS 127
the global nancial crisis. The investment opportunities for households and the interest in
investing have also broadened since then. This has made questions about nancial risk-
taking increasingly topical.
My analysis shows that risk-tolerant households apply for loans more often than risk-
averse households do, and their loans are larger even when other characteristics of
households are controlled for. For mortgage loans, risk aversion is associated with a
smaller probability of having a loan, whereas for non-mortgage loans, risk aversion is
negatively related to the size of the debt. Householdsexpectations about their future
nancial situation do not contain any relevant additional information for predicting
the applying for a loan and the size of debt beyond that given by the main economic
and sociodemographic characteristics. The study also shows that risk aversion is related
to several characteristics of the household, such as income, reference persons edu-
cation, age, employment status, gender and the perceived ability to get nancial assist-
ance from friends or relatives.
The paper proceeds as follows. Section 2 describes the theoretical background of the
study and reviews the related literature. Section 3 gives an overview of the data used in
this study. Section 4 describes the methods applied. Section 5 presents the results: section
5.1 on the factors of optimistic expectations and risk aversion, section 5.2 on the factors of
borrowing decisions and section 5.3 on the factors deciding the size of debt and the debt
burden of households. Section 6 concludes.
2. Literature review
One of the main concepts that can explain the relations between the expectations house-
holds have for their future nancial situation and their borrowing is the Life-cycle/Perma-
nent Income Hypothesis (Modigliani and Brumberg (1954), Friedman (1957)). The Life-
cycle/Permanent Income Hypothesis suggests that consumers aim to hold their consump-
tion levels steady over time at the level of their permanent income by borrowing and
saving. Consumption smoothing creates a positive relationship between optimistic
nancial expectations and borrowing. Important aspects that should be considered
when applying the concept of the Life-cycle/Permanent Income Hypothesis are precau-
tionary saving (Leland (1968)) and borrowing constraints (Zeldes (1989)). Precautionary
saving implies that uncertainty about future income reduces current consumption and
increases saving. Borrowing constraints can reduce consumption directly if they are cur-
rently binding, or indirectly through the possibility of borrowing constraints becoming
binding in the future. The life-cycle model also provides an important theoretical back-
ground for studying how risk aversion aects borrowing, since risk averseness is
reected in the utility function of the household.
Brown et al. (2005) demonstrate based on a life-cycle model that there is a positive
relationship between optimistic expectations for future income and the size of a loan,
and Brown et al. (2013) use a life-cycle model to show theoretically how risk aversion
impacts the size of the debt. Brown et al. (2005) model loan sizes as the outcome of sim-
ultaneous decisions by borrowers and lenders, while expectations enter the model
through the probabilities of occurrence of a high income state and a low state. The expec-
tations are the same for borrowers and lenders. In the model of Brown et al. (2013), the
relationship between risk aversion and the optimally chosen level of debt is found by
128 E. BRANTEN
solving the individuals lifetime expected utility maximisation problem, given the budget
constraint.
Quite a lot of studies have investigated the determinants of household debt and the
debt burden. Some authors have used macro-level aggregate data, while others have
used individual-level or household-level microdata. Microdata from the Eurosystem
Household Finance and Consumption Survey have been used by Bover et al. (2014),
one of whose several research questions analyses how dierent household-level charac-
teristics aect dierent measures of debt, and by Du Caju et al. (2016) to analyse how
these aect the likelihood of the household being over-indebted. HFCS data have been
employed for analysing the nancial fragility of households by Ampudia et al. (2016)
and Terraneo (2018). These studies include several euro area countries.
2
Although the
focus in these studies is somewhat dierent than in this paper, they provide important
input for the choice of the control variables in my models.
In Bover et al. (2014), for example, among the dependent variables are the dummies of
whether the household has mortgage debt or non-mortgage debt and the sizes of mort-
gage and non-mortgage debt. They nd that important factors related to the incidence
and size of debt of a household are age, income and education, although age, income
and education proles of borrowers dier to a great extent in dierent euro area
countries. Du Caju et al. (2016) investigate how the labour market status and other house-
hold demographic and socioeconomic characteristics aect the likelihood of over-indebt-
edness (measured by dierent measures). Their results suggest that the households
reference person being unemployed signicantly raises the likelihood of over-indebted-
ness. Ampudia et al. (2016) and Terraneo (2018) also analyse dierent measures of
over-indebtedness and stress that the debt burden and risk of default of households
are markedly dierent in countries under observation in their studies.
Risk preference has been observed as a factor aecting household debt in several
studies, including Brown et al. (2013) and Massó and Abalde (2020), which are the
closest to my study. Brown et al. (2013)nd from US data that risk aversion is negatively
related to the level of debt of the household. Like the study by Brown et al. (2013), my
study analyses the eects of risk attitudes on both the total debt of the household and
the mortgage debt and non-mortgage debt separately. The importance of risk attitudes
in explaining household indebtedness in Southern European countries has been investi-
gated from the HFCS data by Massó and Abalde (2020).
Relatively few earlier studies have focused on the relations between the expectations
of households about their future nancial situation and their borrowing decisions. Biało-
wolski (2019) studies how economic sentiment inuences the saving and borrowing
behaviour of households in Poland. One of his results is that consumer condence
aects the acquisition of debt for durables and mortgages positively, but has a negative
impact on the acquisition of debt for unexpected expenditures or consumption purposes.
Kłopocka (2017)nds from data for Poland that indexes of consumer condence have
predictive power on their own and are informative for future household saving and bor-
rowing rates alongside the information contained in economic fundamentals. Hyytinen
and Putkuri (2018) analyse the forecast errors that households make and how these
relate to the borrowing behaviour of households and their over-indebtedness. They
nd that the households with the largest optimistic forecast errors have larger debt-to-
income ratios and are more likely to perceive diculties in coping with their liabilities.
BALTIC JOURNAL OF ECONOMICS 129
Brown et al. (2005) and Brown et al. (2008)nd from British microdata that optimistic
expectations of households for their nancial situation are positively related to the size
of their debt.
3. Data
This study employs microdata from two waves of the Estonian Household Finance and
Consumption Survey carried out in 2013 and 2017. The Household Finance and Consump-
tion Survey is a joint project by the European Central Bank and national central banks and
statistical institutes of the euro area countries, Croatia, Hungary and Poland, and the
surveys are conducted in a standardised form in all these states. The Household
Finance and Consumption Survey collects information on the assets, liabilities, incomes,
and consumption of households, and also on the expectations and opinions of house-
holds and on demographic variables for household members (Eesti Pank, 2019). The
results of the Estonian Household Finance and Consumption Survey are provided in Mer-
iküll and Rõõm (2016,2019).
The models describing the factors of optimistic expectations and risk aversion in this
paper include all the households that participated in the 2017 survey for which the
values for relevant dependent and independent variables are available. The models
describing the factors of debt-related variables are based on the households that partici-
pated in both the 2013 and 2017 surveys and where the age of the Canberra denition
reference person is between 20 and 64 years in the 2013 survey.
3
All the households
that participated in the 2013 survey were contacted and invited to participate in the
2017 survey. The 2017 survey also included new split households, where a member of
a household that was interviewed in 2013 had moved into a separate household, and
new households drawn randomly from the Population Register (Eesti Pank, 2019).
Number of observations in the following tables denotes the number of households. Over-
view of the variables used in this paper is provided in Table A1 in Appendix.
The means and standard deviations of the variables are presented in Table 1. The stat-
istics are calculated for observations used in each model of this paper separately. Since
presenting each models statistics separately would be lengthy, Table 1 shows ranges
of means and standard deviations of the variables included in the models in 1) section
5.1, 2) section 5.2, and 3) section 5.3.
The HFCS data are multiply-imputed data
4
and the estimations of the models are
carried out in the multiple imputation (MI) regime in Stata. Imputed variables in the
HFCS dataset include components of net assets, income and consumption; details of
loans such as the interest rate, the year the loan was taken, the maturity of the loan,
and its initial value and current value; indicators for credit constraints; and some com-
ponents of pension plans (Eesti Pank, 2019).
4. Method
Probit and Heckman models are estimated using the microdata dataset. Probit models are
estimated to investigate the factors that are related to optimistic expectations and risk
averseness of households, namely whether 1) the household expects that its income
would grow more than prices over the next year, 2) the household expects the price of
130 E. BRANTEN
the residence the household is living in to increase by more than 5% over the next 12
months, and 3) the household is not willing to take any nancial risk. Probit models are
also estimated to study borrowing measures, namely the factors of i) whether the house-
hold had applied for credit within the previous three years, ii) whether the household
received as much credit as it applied for
5
, and iii) whether the household did not apply
for credit because of a perceived credit constraint. Probit (or logit) models are commonly
used to study binary dependent variables. The logit model, which is a relatively similar
approach to the probit model employed in this study, has been used in, for example, Pat-
tarin and Cosma (2012) to ascertain the relations between attitudes towards credit and
Table 1. Summary statistics of the variables.
Models in section 5.1 Models in section 5.2 Models in section 5.3
Number of
observations
(range) 2 6102 637 4162 278 3801 318
Variable
Mean
(range)
Standard
deviation
(range)
Mean
(range)
Standard
deviation
(range)
Mean
(range)
Standard
deviation
(range)
APPL 0.290.31 0.4520.461
REC 0.89 0.318
PC 0.07 0.257
HD 0.500.68 0.4690.500 0.62 0.487
HMD 0.28 0.446
HNMD 0.53 0.499
LS 22 587 40 649
MDS 44 531 51 120
NMDS 3 251 4 942
DSTI 0.10 0.103
EIP 0.07 0.256 0.080.10 0.2680.303 0.080.11 0.2680.319
EHP 0.07 0.250 0.070.09 0.2500.290 0.08 0.2700.277
RA 0.73 0.445 0.560.68 0.4680.496 0.480.68 0.4680.500
AINC 21 55630
020
23 90527 366
INC1 0.20 0.3970.399 0.090.14 0.2910.351
INC2 0.20 0.401 0.090.11 0.2820.314
INC3 0.20 0.4010.402 0.200.21 0.3970.404
INC4 0.20 0.399 0.25 0.435
INC5 0.20 0.4000.401 0.290.37 0.4530.484
AGE20-34 0.260.37 0.4360.484 0.260.38 0.4360.485
AGE35-49 0.360.43 0.4810.495 0.360.48 0.4810.500
AGE50-64 0.200.38 0.4030.486 0.140.38 0.3460.486
AGE<35 0.22 0.4140.415
AGE35-54 0.33 0.469
AGE>54 0.45 0.498
EDU1 0.15 0.359 0.110.13 0.3170.341 0.070.12 0.2550.324
EDU2 0.48 0.500 0.490.52 0.5000.501 0.490.52 0.5000.501
EDU3 0.37 0.4820.483 0.37 0.4830.484 0.360.44 0.4800.497
MALE 0.42 0.494
EMPL 0.560.57 0.496 0.790.87 0.3320.409 0.79 0.409
MAR 0.33 0.470
HT1 0.270.31 0.4460.463
HT2 0.250.30 0.4330.460
HT3 0.390.48 0.4870.500
HS 2.202.21 1.299 2.67 1.408
FA 0.30 0.457
OWN 0.760.85 0.3580.428
Note. Figures are calculated using survey weights of households (variable HW0010 in the HFCS User Database). Sources:
Estonian Household Finance and Consumption Survey; authors calculations and compilation.
BALTIC JOURNAL OF ECONOMICS 131
the use of consumer credit. The logit model has also been applied in Bover et al. (2014)to
model the probability of holding debt, and in Du Caju et al. (2016) to model the probabil-
ities of dierent measures of indebtedness. The ordered logit approach has been used in
Colasante and Riccetti (2021) to model the factors of risk attitudes.
The Heckman model is estimated to ascertain the factors of the size of the outstand-
ing liabilities of households.
6
The selection equation estimates the probability of the
household having debt and the outcome equation estimates the size of the debt, con-
ditional on borrowing. The rationale behind using the Heckman model is the assump-
tion that the group of households that have a loan liability is dierent from the group
of households that do not have a loan liability in certain characteristics, meaning that
the sample of households that have debt is not randomly drawn from the population.
The Heckman model is then used to correct for the non-random sampling. The actual
estimation of the Heckman model supports this assumption, since the inverse of the
Millsratio that is included in the outcome equation turns out to be statistically
signicant.
The variables of the exclusion restriction that are included in the Heckman selection
equation but not in the outcome equation are a dummy that takes the value 1 if the refer-
ence person of the household is employed, and a variable indicating the size of the house-
hold. Exclusion restrictions are selected following the intuition that employment status
and household size play a signicant role in the probability of a household applying
for and having a loan above all, while they are of lesser importance for explaining the
size of the debt. Before the exclusion restrictions were selected, correlations between
the dependent variable of the selection equation and dierent control variables and cor-
relations between the dependent variable of the outcome equation and dierent control
variables were investigated.
The control variables in the models were selected on the basis of the previous theor-
etical and empirical literature and the analysis of correlations done for this study. Bover
et al. (2014), for example, point out that important factors of debt holdings of households
in the euro area are the age, income and education level of household members. My
analysis of correlations shows that among these three variables, age is most strongly
related to the dependent variables under observation in my paper for Estonia.
The Heckman selection equation is as follows:
Probability(HD2017i=1) =F(
a
0+
a
1RA2013i+
a
2EIP2013i+
+
a
3EHP2017i+
K
k=1
g
kx2013ik +
2
s=1
d
sz2013is +ui)
(1)
where HD2017i,RA2013i,EIP2013i, and EHP2017idenote the values of the variables HD,RA,
EIP, and EHPdescribed in Table A1 in Appendix for household iin the 2017 or 2013
survey, x2013i1,x2013i2,...,x2013ik denote the control variables and z2013i1,z2013i2
denote exclusion restrictions, and u
i
denotes the error term. The control variables
include age and education of the reference person of the household, income of the
household and a dummy indicating whether the household owns its main residence.
The parameters of the equation are
a
0,
a
1,
a
2,
a
3,
g
1,
g
2,...,
g
kand
d
1,
d
2. F(.)
denotes the cumulative distribution function for a standard normally distributed
random variable.
132 E. BRANTEN
The Heckman outcome equation is:
ln(LS2017i)=
b
0+
b
1RA2013i+
b
2EIP2013i+
b
3EHP2017i+
K
k=1
m
kx2013ik +
u
IMRi+ui(2)
where LS2017iis the outstanding balance of liabilities of household iin the 2017 survey and
IMRidenotes the inverse of the Millsratio for household i, which corrects for the selection
in the outcome equation. The parameters of the equation are
b
0,
b
1,
b
2,
b
3,
m
1,
m
2,
,
m
kand
u
.
The Heckman selection and outcome equations described above are estimated for the
total debt, covering both mortgage and non-mortgage loans, and for mortgage loans and
non-mortgage loans separately. The Heckman model is also used for modelling the debt
service-to-income ratio of the household.
It should be noted that the models with the dependent variable for whether the
household had applied for credit are estimated for the panel of households that partici-
pated in both the 2013 and 2017 HFCS. The values of the dependent variable of the
models are taken from the 2017 HFCS, while the values of the explanatory variables
are taken from the 2013 HFCS. The same applies for the models with the dependent
variables for whether the household received credit and for whether the household
did not apply for credit because of a perceived credit constraint, and also for the
Heckman models estimated. The rationale behind this is the assumption that the situ-
ation for the loan liabilities of households in 2017 was aected by the decisions taken
previously, which are assumed to be aected by the past values of dierent explana-
tory variables. The variable for house price expectation is only available in the 2017
survey and so this is used. In the probit models estimated with expectations and risk
aversion as dependent variables, the values of both the dependent variables and the
explanatory variables are taken from the 2017 HFCS and these models are estimated
using the whole sample of the 2017 HFCS.
The models describing the factors of expectations and risk aversion are estimated so
that the household member-level variables take the values of the interview reference
person. The other models are estimated so that the household member-level variables
take the values of the Canberra denition reference person. The interview reference
person was in the large majority of cases the person who answered the household
questionnaire, including the questions about the risk attitudes and expectations of
the household. This means it may be informative to use the characteristics of this
person to analyse the factors of those risk attitudes and expectations. The character-
istics of the Canberra denition reference person may, however, provide more infor-
mation for the households borrowing decisions. As a robustness check, the models
are also estimated so that the household member-level variables take the values of
the Canberra denition reference person for models describing the factors of risk aver-
sion and expectations and of the interview reference person for the other models.
Robustness checks were also carried out using dierent variables for risk attitudes
and expectations. The main results of this paper are robust to changes in the choice
of the reference person and the choice of the explanatory variables included in the
models.
BALTIC JOURNAL OF ECONOMICS 133
5. Results
5.1. Factors of optimistic expectations and risk aversion
Before the analysis of the relationships between expectations, risk attitudes and the bor-
rowing-related variables is conducted, the factors related to these expectations and risk
attitudes are claried. Table 2 gives an indication of the main variables that are related
to positive expectations and risk averseness. Income, age, education, employment
status, gender, and the perceived ability to get nancial assistance from friends or rela-
tives can be noted as particularly important factors.
In the 2017 survey, 73% of households were not willing to take any nancial risk.
Income, education, being employed, and perceiving the ability to get nancial assistance
from friends or relatives were positively related to and age negatively related to a willing-
ness to take risk. The magnitude of the average marginal eect is largest for age. House-
holds where the reference person is younger than 35 are 29.3 percentage points less likely
to be risk averse than households where the reference person is more than 54 years old.
Table 2. Estimated probit models for predicting expectations and risk aversion.
Dependent variable: EIP
(2017)
Dependent variable: EHP
(2017)
Dependent variable: RA
(2017)
Income group (reference group:
INC1) (2017)
INC2 0.0140
(0.0252)
0.0258
(0.0274)
0.0369
(0.0405)
INC3 0.000174
(0.0269)
0.0347
(0.0276)
0.0546
(0.0396)
INC4 0.00853
(0.0283)
0.0279
(0.0295)
0.114***
(0.0425)
INC5 0.0147
(0.0295)
0.0318
(0.0312)
0.181***
(0.0468)
Age (reference group: AGE>54)
(2017)
AGE<35 0.104***
(0.0200)
0.0213
(0.0185)
0.293***
(0.0346)
AGE35-54 0.0568***
(0.0155)
0.0210
(0.0194)
0.143***
(0.0266)
Education level (reference group:
EDU1) (2017)
EDU2 0.0114
(0.0151)
0.000318
(0.0231)
0.0902***
(0.0285)
EDU3 0.0289
(0.0201)
0.0164
(0.0227)
0.177***
(0.0326)
MALE (2017) 0.0292**
(0.0127)
0.0127
(0.0127)
0.0387**
(0.0189)
EMPL (2017) 0.00620
(0.0158)
0.00501
(0.0173)
0.0625**
(0.0258)
MAR (2017) 0.0139
(0.0124)
0.00333
(0.0127)
0.00959
(0.0211)
FA (2017) 0.0525***
(0.0144)
0.0275*
(0.0157)
0.117***
(0.0208)
HS (2017) 0.000364
(0.00529)
0.00209
(0.00542)
0.00311
(0.00824)
Number of observations 2,610 2,637 2,634
Pseudo R
2
0.134 0.014 0.222
Notes. Average marginal eects of the variables. Standard errors in parentheses. *, ** and *** denote signicance at the
10%, 5% and 1% levels respectively. Values of Pseudo R
2
are calculated for each implicate separately and the minimal
values across implicates are reported in the table (a similar approach has been used in Kukk (2017)). Sources: Estonian
Household Finance and Consumption Survey; authors calculations.
134 E. BRANTEN
The magnitude of the eect is also quite large for income and education. Households in
the quintile with the highest income are about 18.1 percentage points less likely to be risk
averse than households in the lowest income quintile. The probability of those with ter-
tiary-level education being risk averse is 17.7 percentage points lower than the probability
for those with less than upper-secondary education. The results also indicate that men are
less risk averse than women, but that eect is relatively small next to the eects of age,
income and education.
My results that income and education are negatively related to risk averseness are in
line with previous studies (examples of these are provided in Van de Venter et al.
(2012) and Tavor (2019)). Previous studies have also found a positive relationship
between age and risk averseness, and have found that women are more risk averse
than men are (examples of these are provided in Van de Venter et al. (2012) and Tavor
(2019), while some examples for gender dierences are also provided in Fisher and Yao
(2017) for example). Banks et al. (2020) show that an increase in risk averseness at
older ages can largely be explained by health changes and other life events, such as retire-
ment, widowhood or marital change. Fisher and Yao (2017, p. 191) point out that gender
dierences in nancial risk tolerance are explained by gender dierences in the individual
determinants of nancial risk tolerance, and that the disparity does not result from gender
in and of itself. Their study nds income uncertainty to be a variable that intermediates
the eect of gender on nancial risk tolerance.
As can be seen from Table 2, most of the main characteristics of a household are not
signicantly related to its expectations about its future nancial situation. The house-
holds perceived ability to get nancial assistance from friends or relatives seems to con-
tribute to optimism in its expectations, though it is not impossible that optimism
encourages the expectation that assistance is available from friends or relatives. It is
notable that younger people and men may be more optimistic in their real income expec-
tations, though the magnitude of the eect of gender is not very large.
5.2. Factors of borrowing decisions of households
This subsection focuses on the analysis of factors that aect the demand for and the
supply of loans. Models (1) and (2) in Table 3, which include no control variables,
show that optimistic expectations are statistically signicantly and positively related to
the decision of the household to apply for credit. The estimated average marginal
eects are signicant and positive for real income expectations as well as house price
expectations. At the same time, model (3) shows that risk averse households are less
likely to apply for credit. However, when control variables are added, the variables
describing expectations become insignicant (model (4) in Table 3). It seems that expec-
tations are on their own related to the decision to apply for a loan, but in this setting
they do not contain any relevant additional information beyond the main economic
and sociodemographic characteristics of the household. These results could be
viewed in the context of the general state of the economy in Estonia in 20132017,
which was a time of economic expansion. Ahmed and Cassou (2016, p. 86) point out
that during economic expansions, consumer condence shocks likely reect news,
while during economic contractions, consumer condence shocks are consistent with
animal spirits. It might in consequence be that the condence indicators of households
BALTIC JOURNAL OF ECONOMICS 135
did not contain much of the animal spirits
7
, which are states that cannot be explained
by economic rationality.
In the 2017 survey, 31% of households had applied for credit within the previous three
years. Model (4) in Table 3 shows the reluctance of the household to take any nancial risk
is, on average, related to the probability of the household applying for a loan being 5.7
percentage points lower. The magnitude of the eect is not very large, but still sizeable
compared to several other control variables in the model.
Table 3. Estimated probit models for predicting households applying for credit, receiving credit and
not applying for credit due to perceived credit constraints.
Dependent variable: APPL (2017)
Dependent
variable: REC
(2017)
Dependent
variable: PC (2017)
(1) (2) (3) (4) (5) (6)
EIP (2013) 0.0960*
(0.0511)
0.00728
(0.0466)
0.0630
(0.0470)
0.0333
(0.0239)
EHP (2017) 0.0868**
(0.0433)
0.0579
(0.0552)
0.0521
(0.0367)
0.0276
(0.0486)
RA (2013) 0.149***
(0.0301)
0.0573*
(0.0331)
0.0491
(0.0434)
0.0215
(0.0238)
Income group (reference
group: INC1) (2013)
INC2 0.0174
(0.0661)
0.0198
(0.134)
0.000197
(0.0454)
INC3 0.0288
(0.0611)
0.0737
(0.109)
0.0141
(0.0424)
INC4 0.0364
(0.0595)
0.0421
(0.119)
0.0379
(0.0400)
INC5 0.0341
(0.0629)
0.166
(0.106)
0.0362
(0.0399)
Age (reference group:
AGE50-64) (2013)
AGE20-34 0.221***
(0.0446)
0.0131 (0.0618) 0.0181
(0.0273)
AGE35-49 0.123***
(0.0357)
0.0575
(0.0585)
0.00196
(0.0219)
Education level (reference
group: EDU1) (2013)
EDU2 0.0706
(0.0497)
0.122*
(0.0712)
0.0810**
(0.0413)
EDU3 0.0916*
(0.0519)
0.0841
(0.0769)
0.0830*
(0.0426)
EMPL (2013) 0.113***
(0.0405)
0.0624
(0.0439)
0.0150
(0.0268)
Household type (reference
group: HT1) (2013)
HT2 0.0226
(0.0434)
0.00742
(0.0562)
0.00718
(0.0245)
HT3 0.00582
(0.0369)
0.0324
(0.0546)
0.00944
(0.0222)
HD (2013) 0.132***
(0.0310)
0.0512
(0.0416)
0.0200
(0.0201)
Number of observations 1,318 2,278 1,321 1,318 416 1,318
Pseudo R
2
0.003 0.002 0.019 0.094 0.06 0.038
Notes. Average marginal eects of the variables. Standard errors in parentheses. *, ** and *** denote signicance at the
10%, 5% and 1% levels respectively. Values of Pseudo R
2
are calculated for each implicate separately and the minimal
values across implicates are reported in the table. Sources: Estonian Household Finance and Consumption Survey;
authors calculations.
136 E. BRANTEN
In an alternative specication, which is otherwise similar to model (4) but where risk
attitudes are measured by a dummy taking the value 1 if the household is willing to
take substantial or above average nancial risks, the variable for risk willingness turns
out to be signicant and has a positive and of larger magnitude eect than the eect
of being risk averse. Households willing to take substantial or above average nancial
risks apply for a loan with about 9.2 percentage points higher probability, compared to
other households. This indicates that being willing to take signicant nancial risks
encourages applying for credit more than being risk averse discourages it.
Although risk averseness is related to a smaller probability of applying for a loan, it is
statistically insignicant in predicting whether the loan application was satised, as
shown in model (5) in Table 3. It is also irrelevant for the probability of not applying for
credit because of a perceived credit constraint, as shown in model (6) in Table 3. Risk
averseness is insignicant in models with these dependent variables even when other
controls for household characteristics are not included.
5.3. Factors of the size of debt and debt burden of households
Tables 46present the estimated Heckman models. Table 4 covers the total debt of the
household, Table 5 presents mortgage debt and non-mortgage debt separately, and
Table 6 shows the debt service burden. Table 4 shows that the reluctance of a household
to take any nancial risk is statistically signicantly and negatively related to the size of
households debt even when other main characteristics of the household and selection
are controlled for. It is interesting that income is a statistically signicant factor in predict-
ing whether or not the household has debt, meaning it is important in the Heckman selec-
tion equation, but after the selection is controlled for, income is insignicantly related
with the size of total debt in the Heckman outcome equation. Table 5 shows that
income is still important for predicting the size of mortgage debt.
Table 5 shows that the estimated eects for mortgage loans and non-mortgage loans
dier by some relevant aspects. For mortgage loans, risk aversion is negatively related to
the probability of the household having a loan, as risk aversion is statistically signicant in
the Heckman selection equation, whereas for non-mortgage loans, risk aversion is nega-
tively related to the size of the outstanding liabilities since risk aversion is statistically sig-
nicant in the Heckman outcome equation. This result seems intuitively logical. Since a
mortgage loan is a large liability for a long period of time, the decision on whether or
not to take on such a liability depends on how risk averse the household is. Non-mortgage
loans are usually smaller and have a shorter term, so the decision of whether to take on
this type of liability is easier to take and less demanding of risk-tolerance. It is, however,
risk-tolerant households that take larger non-mortgage loans. Interestingly, Brown et al.
(2013)nd based on data on US households a negative relationship between risk aversion
and the level of debt for both mortgage and non-mortgage debt.
It is worth noting from Tables 4 and 5that the magnitude of the eect that the risk
aversion has on loan size is relatively large. The relatively large eect of risk attitudes
on the level of debt is also pointed out by Brown et al. (2013). Tables 4 and 5show
that the outstanding balance of total liabilities is on average 73.6% smaller for households
reluctant to take any nancial risk and the balance of non-mortgage debt is 38.8% smaller
ceteris paribus.
BALTIC JOURNAL OF ECONOMICS 137
As can be seen from Tables 45, the variables on expectations are insignicant in pre-
dicting the sizes of debt. Brown et al. (2005) and Brown et al. (2008), however, nd from
British microdata that positive nancial expectations
8
are signicantly and positively
related to the sizes of non-mortgage debt and mortgage debt, respectively, and the
eects of expectations are relatively large. There can be dierent reasons why my
results are dierent, for example dierent macroeconomic or cultural context, or the
way expectations are measured.
Table 6 presents the results of the Heckman model describing the factors of the debt
service burden of the household. It shows that the expectations and risk aversion of the
household are not signicantly related to the households debt service-to-income ratio. In
an alternative model specication, which is otherwise similar to that presented in Table 6,
but where risk attitudes are measured by a dummy taking the value 1 if the household is
willing to take substantial or above average nancial risks, the variable for risk willingness
turns out to be signicant. Households willing to take substantial or above average
nancial risks have, on average, by about 3 percentage point higher debt service-to-
income ratio.
Table 4. Estimated Heckman model for predicting the size of total debt.
Selection
equation:HD (2017)
Outcome equation:Logarithm
value of LS (2017)
EIP (2013) 0.0426
(0.0491)
0.0486
(0.274)
EHP (2017) 0.00425
(0.0539)
0.307
(0.343)
RA (2013) 0.0431
(0.0331)
0.736***
(0.177)
AINC (2013) 0.0283**
(0.0125)
0.0649
(0.116)
Age (reference group: AGE50-64) (2013)
AGE20-34 0.256***
(0.0423)
0.928*
(0.505)
AGE35-49 0.220***
(0.0393)
0.0601
(0.466)
Education level (reference group: EDU1) (2013)
EDU2 0.0159
(0.0512)
0.421
(0.278)
EDU3 0.0157
(0.0527)
0.559*
(0.294)
OWN (2013) 0.0321
(0.0413)
0.825***
(0.268)
EMPL (2013) 0.123***
(0.0451)
HS (2013) 0.0328**
(0.0129)
Constant 10.55***
(2.680)
Inverse of Millsratio 8.825***
(3.265)
Number of observations 1,318 830
Pseudo R
2
0.112
Adjusted R
2
0.219
Notes. Average marginal eects of the variables in the selection equation and coecients of the variables in the outcome
equation. Standard errors in parentheses. *, ** and *** denote signicance at the 10%, 5% and 1% levels respectively.
Annual income is transformed by inverse hyperbolic sine. Values of Pseudo R
2
and Adjusted R
2
are calculated for each
implicate separately and the minimal values across implicates are reported in the table. Sources: Estonian Household
Finance and Consumption Survey; authors calculations.
138 E. BRANTEN
A relevant conclusion that can be drawn from Table 6 is that risk aversion is not related
to the level of the debt service burden of the household, but the willingness to take sub-
stantial or above average nancial risks is related to a higher debt service burden.
However, it should be noted here that the debt service-to-income ratio is not fully con-
trolled by households themselves and may change over time through increases in interest
rates or falls in incomes for example.
6. Conclusion
The results of this study highlight the importance of householdsrisk attitudes in borrow-
ing-related decisions. The analysis shows that risk-tolerant households apply for loans
more often than risk-averse households do, and their loans are typically larger. Optimistic
expectations in the household on its real income and its house price are on their own
Table 5. Estimated Heckman models for predicting the sizes of mortgage and non-mortgage debt.
Selection
equation:HMD
(2017)
Outcome equation:
Logarithm value of MDS
(2017)
Selection
equation:HNMD
(2017)
Outcome equation:
Logarithm value of
NMDS (2017)
EIP (2013) 0.0313
(0.0462)
0.109
(0.220)
0.00621
(0.0574)
0.133
(0.279)
EHP (2017) 0.0268
(0.0490)
0.185
(0.204)
0.0214
(0.0590)
0.160
(0.345)
RA (2013) 0.122***
(0.0316)
0.0310
(0.272)
0.000741
(0.0355)
0.388**
(0.169)
AINC (2013) 0.0216
(0.0202)
0.181**
(0.0853)
0.0296**
(0.0124)
0.0344
(0.126)
Age (reference group:
AGE50-64) (2013)
AGE20-34 0.326***
(0.0417)
0.114
(0.670)
0.140***
(0.0448)
0.521
(0.370)
AGE35-49 0.194***
(0.0302)
0.180
(0.458)
0.150***
(0.0389)
0.114
(0.390)
Education level
(reference group:
EDU1) (2013)
EDU2 0.0629
(0.0445)
0.301
(0.362)
0.0360
(0.0537)
0.0353
(0.282)
EDU3 0.0926*
(0.0474)
0.556
(0.400)
0.0888
(0.0555)
0.113
(0.344)
OWN (2013) 0.163***
(0.0342)
0.871*
(0.451)
0.0242
(0.0411)
0.523**
(0.256)
EMPL (2013) 0.120***
(0.0375)
0.118***
(0.0455)
HS (2013) 0.00644
(0.0103)
0.0357***
(0.0130)
Constant 11.19***
(3.318)
10.51***
(2.892)
Inverse of Millsratio 4.416
(3.086)
7.146**
(3.548)
Number of
observations
1,318 380 1,318 720
Pseudo R
2
0.178 0.061
Adjusted R
2
0.194 0.067
Notes. Average marginal eects of the variables in the selection equation and coecients of the variables in the outcome
equation. Standard errors in parentheses. *, ** and *** denote signicance at the 10%, 5% and 1% levels respectively.
Annual income is transformed by inverse hyperbolic sine. Values of Pseudo R
2
and Adjusted R
2
are calculated for each
implicate separately and the minimal values across implicates are reported in the table. Sources: Estonian Household
Finance and Consumption Survey; authors calculations.
BALTIC JOURNAL OF ECONOMICS 139
positively related to the decision of the household to apply for credit, but they do not
contain any relevant additional information beyond the households main economic
and sociodemographic characteristics.
It is shown that the eect of risk aversion on debt size is quite large. The size of the
total debt is, on average, by about 73.6% smaller for households unwilling to take on
any nancial risk, and the size of the non-mortgage debt is by about 38.8% smaller.
Risk aversion is not related to the level of the debt service burden measured by the
debt service-to-income ratio of the household, but the willingness to take substantial
or above average nancial risks is related to a 3 percentage point higher debt
service-to-income ratio. All of these results are relevant for nancial stability analysis,
since if risk-tolerant households do not increase their debt service burden by too
much more than risk-averse households do, the risks for households themselves and
for the nancial system are reduced. However, the absolute size of debt, which my
analysis shows to be related to risk averseness, is relevant for the extent of the possible
problems that households and the nancial system could face if the incomes of house-
holds were to fall or interest rates to rise, and loan repayment diculties were then to
Table 6. Estimated Heckman model for predicting the debt service-to-income ratio.
Selection equation:
HD (2017)
Outcome equation:
DSTI (2017)
EIP (2013) 0.0426
(0.0491)
0.0139
(0.0100)
EHP (2017) 0.00425
(0.0539)
0.0142
(0.0164)
RA (2013) 0.0431
(0.0331)
0.0147
(0.0107)
AINC (2013) 0.0283**
(0.0125)
0.00501
(0.00632)
Age (reference group: AGE50-64) (2013)
AGE20-34 0.256***
(0.0423)
0.0368
(0.0277)
AGE35-49 0.220***
(0.0393)
0.0408
(0.0266)
Education level (reference group: EDU1) (2013)
EDU2 0.0159
(0.0512)
0.00642
(0.0206)
EDU3 0.0157
(0.0527)
0.0245
(0.0201)
OWN (2013) 0.0321
(0.0413)
0.00382
(0.0139)
EMPL (2013) 0.123***
(0.0451)
HS (2013) 0.0328**
(0.0129)
Constant 0.118
(0.161)
Inverse of Millsratio 0.372*
(0.209)
Number of observations 1,318 663
Pseudo R
2
0.112
Adjusted R
2
0.017
Notes. Average marginal eects of the variables in the selection equation and coecients of the variables in the outcome
equation. Standard errors in parentheses. *, ** and *** denote signicance at the 10%, 5% and 1% levels respectively.
Annual income is transformed by inverse hyperbolic sine. Values of Pseudo R
2
and Adjusted R
2
are calculated for each
implicate separately and the minimal values across implicates are reported in the table. Sources: Estonian Household
Finance and Consumption Survey; authors calculations.
140 E. BRANTEN
arise. Also, it should be noted that although the share of households willing to take sub-
stantial or above average nancial risks is small, their higher DSTI ratios may pose a risk
to nancial stability.
Further analysis is merited for the dierences between the results for mortgage loans
and those for non-mortgage loans. My analysis shows that for mortgage loans, risk aver-
sion is negatively related to the probability of having a loan, whereas for non-mortgage
loans, risk aversion is negatively related to the size of the outstanding liabilities. This is
intuitive. Since a mortgage loan is a large liability for a long period of time, deciding to
take that loan requires a certain amount of risk tolerance. The amount of risk tolerance
needed to take a non-mortgage loan depends on the size of the loan, and larger non-
mortgage loans are taken by risk-tolerant households.
My study also assesses the characteristics of households that are not willing to take any
nancial risk. Risk aversion is negatively related to the households income, its reference
persons level of education, its reference person being employed, and it perceiving the
ability to get nancial assistance from friends or relatives, while risk aversion is positively
related to the age of the households reference person. The results also indicate that men
are less risk averse than women. These results for Estonia are in line with previous studies
by other authors from dierent countries and regions, and examples of such studies are
provided in Van de Venter et al. (2012), Tavor (2019), and Fisher and Yao (2017) among
others.
Notes
1. The views expressed are those of the author. The results published and the related obser-
vations and analysis may not correspond to the results or analysis of the data producers.
The previous version of the paper was issued as a working paper in the Eesti Pank
Working Paper Series: Branten, E. (2021). The role of risk attitudes and expectations in house-
hold borrowing in Estonia. Eesti Pank, Working Paper Series, 5/2021. https://www.eestipank.
ee/en/publications/working-papers/2021/52021-eva-branten-role-risk-attitudes-and-expecta
tions-household-borrowing-estonia
2. Bover et al. (2014) and Du Caju et al. (2016) include 11 euro area countries: Austria, Belgium,
France, Germany, Greece, Italy, Luxembourg, the Netherlands, Portugal, Slovakia, Spain.
Ampudia et al. (2016) include 14 euro area countries: Austria, Belgium, Cyprus, France,
Germany, Greece, Italy, Luxembourg, Malta, the Netherlands, Portugal, Slovenia, Slovakia,
Spain. The analysis by Terraneo (2018) includes Greece, Italy, Portugal and Spain.
3. The household reference person here is chosen according to the international standards of
the Canberra Group (UNECE, 2011), which suggests applying the following criteria in the fol-
lowing order to select a unique reference person in the household: 1) one of the partners in a
registered or de facto marriage with dependent children, 2) one of the partners in a registered
or de facto marriage without dependent children, 3) a lone parent with dependent children,
4) the person with the highest income, 5) the oldest person.
4. More information on the multiple imputation of the data can be found in the methodological
reports of the HFCS: Eesti Pank (2019); European Central Bank (2013); European Central Bank
(2016); European Central Bank (2020c).
5. The probit model is used here instead of the Heckman model because the inverse of the Mills
ratio did not appear to be statistically signicant in the Heckman outcome equation that was
estimated during the analysis process for this paper.
6. The sample selection approach has also been applied in Brown et al. (2008) to model the size
of mortgage debt and in Aristei and Gallo (2016), for example, who apply a sample selection
ordered probit in the context of the repayment diculties of households. In the rst stage, a
BALTIC JOURNAL OF ECONOMICS 141
probit model describing the likelihood of mortgage insolvency is estimated, and in the
second stage, an ordered probit model describing the intensity of arrears is estimated for
insolvent households.
7. The term rst described by Keynes (1936).
8. Expectations in their studies are measured by the following question: Looking ahead, how
do you think you will be nancially a year from now, will you be: Better o; Worse o;Or
about the same?
Disclosure statement
No potential conict of interest was reported by the author(s).
Notes on contributor
Eva Branten is a PhD student at the Department of Economics and Finance at Tallinn University of
Technology. Her research interests include the borrowing behaviour and debt burden of
households.
References
Ahmed, M. I., & Cassou, S. P. (2016). Does consumer condence aect durable goods spending
during bad and good economic times equally? Journal of Macroeconomics,50,8697. https://
doi.org/10.1016/j.jmacro.2016.08.008
Ampudia, M., van Vlokhoven, H., & Żochowski, D. (2016). Financial fragility of euro area households.
Journal of Financial Stability,27, 250262. https://doi.org/10.1016/j.jfs.2016.02.003
Aristei, D., & Gallo, M. (2016). The determinants of householdsrepayment diculties on mortgage
loans: Evidence from Italian microdata. International Journal of Consumer Studies,40(4), 453465.
https://doi.org/10.1111/ijcs.12271
Banks, J., Bassoli, E., & Mammi, I. (2020). Changing attitudes to risk at older ages: The role of health
and other life events. Journal of Economic Psychology,79, 102208. https://doi.org/10.1016/j.joep.
2019.102208
Białowolski, P. (2019). Economic sentiment as a driver for household nancial behavior. Journal of
Behavioral and Experimental Economics,80,5966. https://doi.org/10.1016/j.socec.2019.03.006
Bover, O., Casado, J. M., Costa, S., Du Caju, P., McCarthy, Y., Sierminska, E., Tzamourani, P., Villanueva,
E., & Zavadil, T. (2014). The distribution of debt across euro area countries. The role of individual
characteristics, institutions and credit conditions. European Central Bank, Working Paper Series,
No. 1639.
Brown, S., Garino, G., & Taylor, K. (2008). Mortgages and nancial expectations: A household level
analysis. Southern Economic Journal,74(3), 857878. https://doi.org/10.1002/j.2325-8012.2008.
tb00868.x
Brown, S., Garino, G., & Taylor, K. (2013). Household debt and attitudes toward risk. Review of Income
and Wealth,59(2), 283304. https://doi.org/10.1111/j.1475-4991.2012.00506.x
Brown, S., Garino, G., Taylor, K., & Price, S. W. (2005). Debt and nancial expectations: An individual-
and household-level analysis. Economic Inquiry,43(1), 100120. https://doi.org/10.1093/ei/cbi008
Colasante, A., & Riccetti, L. (2021). Financial and non-nancial risk attitudes: What does it matter?
Journal of Behavioral and Experimental Finance,30, 100494. https://doi.org/10.1016/j.jbef.2021.
100494.
Dohmen, T., Lehmann, H., & Pignatti, N. (2016). Time-varying individual risk attitudes over the great
recession: A comparison of Germany and Ukraine. Journal of Comparative Economics,44(1), 182
200. https://doi.org/10.1016/j.jce.2015.10.002
Du Caju, P., Rycx, F., & Tojerow, I. (2016). Unemployment risk and over-indebtedness. European
Central Bank, Working Paper Series, No. 1908.
142 E. BRANTEN
Eesti Pank. (2019). The Estonian Household Finance and Consumption Survey (HFCS), methodologi-
cal report.
European Central Bank. (2013). The Eurosystem Household Finance and Consumption Survey.
Methodological Report for the First Wave. Statistics Paper Series, No. 1.
European Central Bank. (2016). The Household Finance and Consumption Survey: methodological
report for the second wave. Statistics Paper Series, No. 17.
European Central Bank. (2020a). HFCS User Database Documentation. Core and derived variables.
2017 Wave.
European Central Bank. (2020b). HFCS User Database Documentation. Core and derived variables.
Wave 2.
European Central Bank. (2020c). The Household Finance and Consumption Survey: Methodological
report for the 2017 wave. Statistics Paper Series, No. 35.
Fisher, P. J., & Yao, R. (2017). Gender dierences in nancial risk tolerance. Journal of Economic
Psychology,61, 191202. https://doi.org/10.1016/j.joep.2017.03.006
Friedman, M. (1957). A theory of the consumption function. Princeton University Press.
Hyytinen, A., & Putkuri, H. (2018). Household optimism and overborrowing. Journal of Money, Credit
and Banking,50(1), 5576. https://doi.org/10.1111/jmcb.12453
Keynes, J. M. (1936). The general theory of employment, interest, and money. Macmillan.
Kłopocka, A. M. (2017). Does consumer condence forecast household saving and borrowing behav-
ior? Evidence for Poland. Social Indicators Research,133(2), 693717. https://doi.org/10.1007/
s11205-016-1376-4
Kukk, M. (2017). How does household debt aect nancial asset holdings? Evidence from euro area
countries. Studies in Economics and Finance,34(2), 194212. https://doi.org/10.1108/SEF-02-2016-
0031
Leland, H. E. (1968). Saving and uncertainty: The precautionary demand for saving. Quarterly Journal
of Economics,82(3), 465473. https://doi.org/10.2307/1879518
Massó, M., & Abalde, N. (2020). The importance of attitudes towards risk to explain indebtedness in
southern European countries. Revista Española de Sociología,29(1), 181198. https://doi.org/10.
22325/fes/res.2020.11
Meriküll, J., & Rõõm, T. (2016). The assets, liabilities and wealth of Estonian households: Results of the
household nance and consumption survey. Eesti Pank. Occasional Paper Series,3/2016.
Meriküll, J., & Rõõm, T. (2019). Estonian Household Finance and Consumption Survey: Results from
the 2017 Wave. Eesti Pank. Occasional Paper Series, 1/2019.
Modigliani, F., & Brumberg, R. (1954). Utility analysis and the consumption function: An interpret-
ation of cross-section data. In K. K. Kurihara (Ed.), Post-Keynesian economics (pp. 388436).
Rutgers University Press.
Pattarin, F., & Cosma, S. (2012). Psychological determinants of consumer credit: The role of attitudes.
Review of Behavioral Finance,4(2), 113129. https://doi.org/10.1108/19405971211284899
Pońsko, P. M. (2018). The evolution of consumer preferences in Poland. Applied Economics Letters,25
(4), 290295. https://doi.org/10.1080/13504851.2017.1319549
Tavor, T. (2019). The theoretical attitude and actual behavior of an individual towards risk. Journal of
Behavioral and Experimental Finance,23,111. https://doi.org/10.1016/j.jbef.2019.05.005
Terraneo, M. (2018). Householdsnancial vulnerability in Southern Europe. Journal of Economic
Studies,45(3), 521542. https://doi.org/10.1108/JES-08-2016-0162
UNECE. (2011). Canberra group handbook on household income statistics. 2nd edition. United Nations
Economic Commission for Europe.
Van de Venter, G., Michayluk, D., & Davey, G. (2012). A longitudinal study of nancial risk tolerance.
Journal of Economic Psychology,33(4), 794800. https://doi.org/10.1016/j.joep.2012.03.001
Zeldes, S. P. (1989). Consumption and liquidity constraints: An empirical investigation. Journal of
Political Economy,97(2), 305346. https://doi.org/10.1086/261605
BALTIC JOURNAL OF ECONOMICS 143
Appendix
Table A1. Description of the variables used in the analysis
Variable Description of the variable
APPL Household has applied for credit within the last three years (1 = yes, 0 = no). Variable is dened based on the
variable HC1300 in the HFCS User Database. Variable APPL is given the value 1 if HC1300 = 1, the value 0 if
HC1300 = 2 and the value is missing if the value for HC1300 is missing.
REC Household received as much credit it applied for within the last three years (1 = yes, 0 = no). Variable only
dened for households that had applied for credit within the last three years. Variable is dened based on
the variable HC1310a in the HFCS User Database. Variable REC is given the value 1 if HC1310a = 3, the value
0 for other values of HC1310a and the value is missing if the value for HC1310a is missing.
PC Household has not applied for credit due to perceived credit constraints (1 = yes, 0 = no). The variable is
dened based on the variable HC1400 in the HFCS User Database. Variable PC is given the value 1 if HC1400
= 1, the value 0 if HC1400 = 2 and the value is missing if the value for HC1400 is missing.The question in the
HFCS questionnaire was: In the last three years, did you (or another member of your household) consider
applying for a loan or credit but then decided not to, thinking that the application would be rejected?
HD Household has debt (1 = yes, 0 = no). Variable DL1000i in the HFCS User Database.
HMD Household has mortgage debt (1 = yes, 0 = no). Variable DL1100i in the HFCS User Database.
HNMD Household has non-mortgage debt (1 = yes, 0 = no). Variable DL1200i in the HFCS User Database.
LS Total outstanding balance of households liabilities, in euros. Variable only dened for households with debt.
Variable DL1000 in the HFCS User Database.
MDS Outstanding balance of households mortgage debt, in euros. Variable only dened for households with
mortgage debt. Variable DL1100 in the HFCS User Database.
NMDS Outstanding balance of households non-mortgage debt, in euros. Variable only dened for households with
non-mortgage debt. Variable includes outstanding balances on credit lines or overdrafts, outstanding
balance of credit cards for which the owner of the card is charged interest, and outstanding balances on
other non-mortgage loans (including private loans from relatives, friends, etc.) and leases. Variable DL1200
in the HFCS User Database.
DSTI Share of debt payments to gross income of the household. Variable only dened for households holding
debt instruments for which payments are collected. Variable DODSTOTALp in the HFCS User Database.
EIP Household expects its total income to go up more than prices over the next year (1 = yes, 0 = no). The
variable is dened based on the variable HG0800 in the HFCS User Database. Variable EIP is given the value
1 if HG0800 = 1, the value 0 for other values of HG0800 and the value is missing if the value for HG0800 is
missing.The question in the HFCS questionnaire was: Over the next year, do you expect your (households)
total income to go up more than prices, less than prices, or about the same as prices?
EHP Household expects the price of the residence the household is living in to increase by more than 5 per cent over
the next 12 months (1 = yes, 0 = no). Variable is dened based on the variable HBZ010e in the HFCS User
Database. Variable EHP is given the value 1 if HBZ010e = 10, the value 0 for other values of HBZ010e and a
missing value if the value for HBZ010e is missing.The question in the HFCS questionnaire was: We are
interestedin knowing how you think theprice of the residence youare living in might changeover the next 12
months. Please distribute a total of 10 points among the 5 changes shown in the card below, assigning more
points to the scenarios you think are more likely and zero points if a scenario seems nearly impossible to you.
RA Household is not willing to take any nancial risk (1 = yes, 0 = no). The variable is dened based on the variable
HD1800 in the HFCS User Database. Variable RA is given the value 1 if HD1800 = 4, the value 0 for other
values of HD1800 and the value is missing if the value for HD1800 is missing.The question in the HFCS
questionnaire was: Which of the following statements comes closest to describing the amount of nancial
risk that you (and your husband/wife/partner) are willing to take when you save or make investments?The
options for answer were the following: a) take substantial nancial risks expecting to earn substantial returns
(variable HD1800 has the value 1); b) take above average nancial risks expecting to earn above average
returns (variable HD1800 has the value 2); c) take average nancial risks expecting to earn average returns
(variable HD1800 has the value 3); d) not willing to take any nancial risk (variable HD1800 has the value 4).
AINC Annual gross income of the household, in euros. Variable DI2000 in the HFCS User Database.
INC1 Households income is in the lowest (rst) income quintile (1 = yes, 0 = no). Variable is dened based on the
variable DHIQ01 in the HFCS User Database. Variable INC1 is given the value 1 if DHIQ01 = 1, the value 0 for
other values of DHIQ01 and the value is missing if the value for DHIQ01 is missing.
INC2 Households income is in the second income quintile (1 = yes, 0 = no). Variable is dened based on the
variable DHIQ01 in the HFCS User Database. Variable INC2 is given the value 1 if DHIQ01 = 2, the value 0 for
other values of DHIQ01 and the value is missing if the value for DHIQ01 is missing.
INC3 Households income is in the third income quintile (1 = yes, 0 = no). Variable is dened based on the variable
DHIQ01 in the HFCS User Database. Variable INC3 is given the value 1 if DHIQ01 =3, the value 0 for other
values of DHIQ01 and the value is missing if the value for DHIQ01 is missing.
INC4
(Continued)
144 E. BRANTEN
Table A1. Continued.
Variable Description of the variable
Households income is in the fourth income quintile (1 = yes, 0 = no). Variable is dened based on the
variable DHIQ01 in the HFCS User Database. Variable INC4 is given the value 1 if DHIQ01 = 4, the value 0 for
other values of DHIQ01 and the value is missing if the value for DHIQ01 is missing.
INC5 Households income is in the fth income quintile (1 = yes, 0 = no). Variable is dened based on the variable
DHIQ01 in the HFCS User Database. Variable INC5 is given the value 1 if DHIQ01 =5, the value 0 for other
values of DHIQ01 and the value is missing if the value for DHIQ01 is missing.
AGE20-
34
Age of the reference person of the household is in the range of 2034 years (1 = yes, 0 = no). Variable is
dened based on the variable RA0300 in the HFCS User Database and has a missing value if RA0300 value is
missing.
AGE35-
49
Age of the reference person of the household is in the range of 3549 years (1 = yes, 0 = no). Variable is
dened based on the variable RA0300 in the HFCS User Database and has a missing value if RA0300 value is
missing.
AGE50-
64
Age of the reference person of the household is in the range of 5064 years (1 = yes, 0 = no). Variable is
dened based on the variable RA0300 in the HFCS User Database and has a missing value if RA0300 value is
missing.
AGE<35 Age of the reference person of the household is below 35 years (1 = yes, 0 = no). Variable is dened based on
the variable RA0300 in the HFCS User Database and has a missing value if RA0300 value is missing.
AGE35-
54
Age of the reference person of the household is in the range of 3554 years (1 = yes, 0 = no). Variable is
dened based on the variable RA0300 in the HFCS User Database and has a missing value if RA0300 value is
missing.
AGE>54 Age of the reference person of the household is more than 54 years (1 = yes, 0 = no). Variable is dened based
on the variable RA0300 in the HFCS User Database and has a missing value if RA0300 value is missing.
EDU1 Highest completed education of the reference person of the household is second stage of basic education of
below (1 = yes, 0 = no). Variable is dened based on the variable PA0200 in the HFCS User Database.
Variable EDU1 is given the value 1 if PA0200 =0, 1 or 2, the value 0 for other values of PA0200 and a
missing value if the value for PA0200 is missing.
EDU2 Highest completed education of the reference person of the household is upper secondary or post-secondary
non-tertiary education (1 = yes, 0 = no). Variable is dened based on the variable PA0200 in the HFCS User
Database. Variable EDU2 is given the value 1 if PA0200 = 3 or 4, the value 0 for other values of PA0200 and
a missing value if the value for PA0200 is missing.
EDU3 Highest completed education of the reference person of the household is tertiary education (1 = yes, 0 = no).
Variable is dened based on the variable PA0200 in the HFCS User Database. Variable EDU3 is given the
value 1 if PA0200 = 5, 6, 7 or 8 in the 2017 survey or if PA0200 = 5 or 6 in the 2013 survey. Variable EDU3 is
given the value 0 for other values of PA0200 and a missing value if the value for PA0200 is missing.
MALE Gender of the reference person of the household is male(1 = yes, 0 = no). Variable is dened based on the
variable RA0200 in the HFCS User Database. Variable MALE is given the value 1 if RA0200 = 1, the value 0 if
RA0200 = 2 and a missing value if the value for RA0200 is missing.
EMPL The reference person of the household is employed (including self-employed) (1 = yes, 0 = no). Variable is
dened based on the variable PE0100a in the HFCS User Database. Variable EMPL is given the value 1 if
PE0100a = 1, the value 0 for other values of PE0100a and a missing value if the value for PE0100a is missing.
MAR The reference person of the household is married (1 = yes, 0 =no). Variable is dened based on the variable
PA0100 in the HFCS User Database. The variable MAR is given the value 1 if PA0100 = 2, the value 0 if
PA0100 has any other value, and a missing value if PA0100 value is missing.
HT1 Household consists of two or more adults (1 = yes, 0 = no). Variable is dened based on the variable
DHHTYPE in the HFCS User Database. Variable HT1 is given the value 1 if DHHTYPE = 6, 7 or 8, the value 0
for other values of DHHTYPE and a missing value if the value for DHHTYPE is missing.
HT2 Household consists of one adult with or without children (1 = yes, 0 = no). Variable is dened based on the
variable DHHTYPE in the HFCS User Database. Variable HT2 is given the value 1 if DHHTYPE = 9, 51 or 52,
the value 0 for other values of DHHTYPE and a missing value if the value for DHHTYPE is missing.
HT3 Household consists of two or more adults with children. Variable is dened based on the variable DHHTYPE
in the HFCS User Database. Variable HT3 is given the value 1 if DHHTYPE = 10, 11, 12 or 13, the value 0 for
other values of DHHTYPE and a missing value if the value for DHHTYPE is missing.
HS Number of members in the household. Variable DH0001 in the HFCS User Database.
FA The household perceives that it can get nancial assistance from friends or relatives (1 = yes, 0 = no).
Variable is dened based on the variable HI0800 in the HFCS User Database. Variable FA is given the value 1
if HI0800 = 1, the value 0 if HI0800 = 2 and a missing value if the value for HI0800 is missing.The question in
the HFCS questionnaire was: In an emergency, could (you/your household) get nancial assistance of say
EUR 5,000 from friends or relatives who do not live with you?
OWN Household owns its main residence (1 = yes, 0 = no). Variable DA1110i in the HFCS User Database.
Sources: Authors compilation based on: Eesti Pank, 2019; European Central Bank, 2020a; European Central Bank, 2020b.
BALTIC JOURNAL OF ECONOMICS 145
... Existing literature mainly focuses on the influencing factors of household debt and the consequences of household over-indebtedness [1]. Micro-level influencing factors of household debt include demographic characteristics, human capital characteristics [2], household risk attitudes, and financial expectations [3]. The impact of household income levels on household debt has a dual effect: most scholars believe that income and debt present a substitution relationship, while some scholars argue that high-income households face smaller financing constraints, hence income and debt exhibit a complementary relationship [4]. ...
Article
Full-text available
In recent years, the risk of rural household debt has risen rapidly, and the development and diffusion of e-commerce in rural areas has the tendency to promote the expansion of household debt. Using the sample data of the China Family Panel Survey (CFPS) from 2014 to 2020,and taking the "Comprehensive Demonstration of E-commerce in Rural Areas" project as a quasi-natural experiment, this paper analyzes the impact and mechanism of e-commerce on rural household debt by using the staggered DID model. The results show that e-commerce significantly increases the debt size and risk of rural households. Mechanism analysis shows that e-commerce mainly alleviates the financial exclusion problem, stimulates household consumption-based expenditure and agricultural production, and thus promotes household consumption and borrowing debt. Based on this, it is important to pay attention to the debt risk of low-income newly indebted households in rural areas, to strengthening the upgrading of the consumption structure in rural areas, raising the standard of living of the rural population. This paper provides a new perspective for understanding the indebtedness of rural residents and identifying the impact of e-commerce and the digital economy on rural debt, which is of some significance for further improving the development of e-commerce in rural areas.
Article
Full-text available
The aim of this paper is to analyse the socio-demographic determinants and individuals’ attitudes towards risk to explain indebtedness in Southern European countries. This analysis is done within the context of an increasing process of financialization of individual behaviour. We use the Household Finance and Consumption Survey (HFCS), drawn up by the European Central Bank between 2009 and 2011. The methodological plan consists of a combination of a multiple correspondence analyses in order to identify the various household debt models and a logistic regression analysis in order to detect the explanatory variables that best explain the differences between indebtedness models. The main contribution of this paper is the identification of those variables that explain the economic differences in household debt levels. Results are showing a relevant influence of individual attitudes to explain indebtedness in Southern Countries in detriment of sociodemographic variables which are not relevant explanatory factors.
Article
We contribute to the literature on household mortgage debt by exploring one particular influence on mortgage debt at the household level, namely the financial expectations of the individuals within the household. Our theoretical model predicts a positive association between the quantity of mortgage debt and optimistic financial expectations. Our empirical findings based on household level data provide convincing support for our theoretical priors in that optimistic financial expectations are positively associated with the level of outstanding mortgage debt.
Article
In this paper we try to: (i) study the personal features of the subjects that can influence the risk attitude in financial and non-financial contexts, (ii) understand the correspondences among some behaviors in financial and non-financial choices. We start from the questionnaire used by Colasante and Riccetti (2020), that investigates how subjects take into account the first four moments of the return distribution in making risky decisions, and that collects data from a very large and heterogeneous population. We find that age and geographical location are important determinants of risk propensity in all domains. Moreover, we find that risk attitudes in financial and non-financial contexts are correlated, but correlation is far from 1, with a larger risk aversion in non-financial contexts. Therefore, there is a common underlying risk trait, but the context is also relevant. Interestingly, the financial risk propensity is positively correlated to the propensity to perform illegal activities.
Article
This paper investigates risk attitudes at older ages in 14 European countries. Older individuals report lower willingness to take risks in all countries. Using panel data we are able to show that this relationship between financial risk attitudes and age is not due to cohort effects or selective mortality. We also show that key mechanisms driving this change with age are health changes and other life events – in our preferred specification around half of the overall evolution of risk attitudes with age can be explained by health shocks, retirement, and widowhood or marital change that occur increasingly frequently as individuals age. These life-events are a particularly important explanation of the evolution of risk attitudes for women.
Article
This paper analyzes and compares risk aversion levels in the theoretical attitude and actual behavior of individuals. Theoretical attitude is measured in two different ways, both directly and indirectly. The present research first examines differences between theoretical attitude and behavior. The findings for the entire sample and for the sociodemographic subgroups show that respondents demonstrate a lower risk aversion level in their behavior than in their indirectly measured theoretical attitude. In contrast, individuals are more risk averse in their behavior than in their directly measured theoretical attitude. Next, within the sociodemographic factors, the study compares the average risk aversion levels in directly and indirectly measured theoretical attitude, and in behavior. The sociodemographic subgroups within gender, age, and marital status differ in their risk aversion levels in theoretical attitude, but none of the sociodemographic subgroups differ in the risk aversion level for behavior. Regression results suggest that when an individual indicates he is more risk averse in both his directly and indirectly measured theoretical attitude, he is also more risk averse in his behavior. In addition, a respondent who is older, more religious, or earns a lower income is more risk averse in his behavior.
Article
The study focused on investigating influence of economic sentiments on household financial decisions. A group of Polish households participated simultaneously in a longitudinal study on financial behavior and a consumer tendency survey. Combination of responses from the two surveys was used to investigate a link between economic situation assessment and household financial decisions. With the use of latent class modeling five distinct patterns of saving behavior and six distinct patterns of borrowing behavior were identified among respondents. Patterns in both these domains were confirmed as stable in time. Subsequently, with the use of individual data, a link between indicators from the consumer tendency survey and the evolution of household financial behavior in the domains of borrowing and savings was established. Consumer confidence stimulated debt for durables and mortgages but reduced debt uptake for unexpected expenditures or consumption purposes. Lower consumer confidence increased only likelihood of saving for consumption, confirming the precautionary motive of saving. Multi-purpose and home renovation savings both correlated positively with consumer confidence.
Article
We use Finnish household-level data from 1994 to 2013 to measure how often and what kind of forecast errors households make and how the errors are linked to the households' borrowing behavior and overindebtedness. We find that those households that make the largest optimistic forecast errors have greater debt-to-income ratios. They also are more likely to report that they suffer from excessive debt loads and have problems in coping with their bills. There are no such systematic effects for the households that make pessimistic forecast errors.