Financial literacy gaps of the Austrian population

Abstract

Most people lack financial knowledge and have trouble dealing competently with their financial affairs, as evidenced by recent empirical research on financial literacy in various countries. We investigate financial literacy among Austrians based on a novel dataset covering about 2,000 persons. Our findings corroborate the existence of substantial knowledge gaps, which are most conspicuous among respondents with low educational attainment and low incomes, among younger and older people as well as women. We provide tentative evidence for a positive link between financial knowledge scores and financial behavior, such as setting
aside rainy day funds or using several sources of information when choosing financial products.

Full text

MONETARY POLICY & THE ECONOMY Q2/15 35
Over the last years, numerous empiri-
cal studies, including large-scale and
international ones, have consistently
shown that a considerable share of the
population has substantial knowledge
gaps in basic financial concepts, such as
interest rates or inflation (Atkinson and
Messy, 2012). At the same time, finan-
cial products and economic relation-
ships have become ever more complex
over the past decades, requiring a
higher level of financial knowledge.
In this paper, we investigate the
level of financial literacy of the Aus-
trian population as well as the relation-
ship between financial knowledge and
sound financial behavior. Our analysis
is based on a new quantitative dataset
resulting from a survey of about 2,000
Austrian individuals. This survey was
part of an OECD initiative to collect
comparable data on financial literacy
for a broad set of countries (OECD,
2013). In line with the OECD’s defini-
tion of financial literacy, the dataset
therefore covers information on finan-
cial knowledge, attitudes and behavior.
The paper is structured as follows:
Section 1 defines financial literacy and
financial knowledge and gives an over-
view of findings of previous empirical
research. Section 2 describes the new
Austrian dataset and analyzes the
determinants of financial knowledge.
Section 3 deals with respondents’ self-
assessment of their financial knowledge.
Section 4 links financial knowledge to a
set of behavior patterns and section 5
concludes.
1 Financial literacy and financial
knowledge: definitions and
empirical findings
As with many other complex terms and
concepts, there is no single generally
accepted definition of financial literacy.
Researchers and organizations have de-
fined financial literacy in many differ-
ent ways (Hung et al., 2009; Holzmann,
2010). We apply the rather broad defi-
nition used by the Organisation for
Economic Co-operation and Develop-
ment (OECD) and its International
Network on Financial Education (INFE),
which integrates financial knowledge,
skills, attitudes and behavior, as well as
their mutual relationships: Financial
literacy is defined as “a combination of
Refereed by:
Thomas Retzmann,
University of
Duisburg-Essen
Financial literacy gaps of the Austrian
population
Most people lack financial knowledge and have trouble dealing competently with their
financial affairs, as evidenced by recent empirical research on financial literacy in various
countries. We investigate financial literacy among Austrians based on a novel dataset covering
about 2,000 persons. Our f indings corroborate the existence of substantial knowledge gaps,
which are most conspicuous among respondents with low educational attainment and low
incomes, among younger and older people as well as women. We provide tentative evidence
for a positive link between f inancial knowledge scores and financial behavior, such as setting
aside rainy day funds or using several sources of information when choosing financial products.
Maria Silgoner,
Bettina
Greimel-Fuhrmann,
Rosa Weber1
JEL classification: A20, D14
Keywords: financial literacy, financial education, central banks
1 Oesterreichische Nationalbank, Foreign Research Division, maria.silgoner@oenb.at; Vienna University of Eco-
nomics and Business, bettina.fuhrmann@wu.ac.at and rosa.weber.09@gmail.com (intern, Foreign Research
Division, January–May, 2015). The views expressed in this paper are exclusively those of the authors and do not
necessarily reflect those of the OeNB, the Vienna University of Economics and Business or the Eurosystem. The
authors would like to thank Pirmin Fessler and Martin Taborsky (OeNB), Gerd Feistritzer and Teresa Schaup
(IFES) as well as the participants of an internal OeNB seminar for helpful comments and valuable suggestions.

Financial literacy gaps of the Austrian population
36 OESTERREICHISCHE NATIONALBANK
financial awareness, knowledge, skills,
attitude and behaviour necessary to
make sound financial decisions and
ultimately achieve individual financial
wellbeing” (Atkinson and Messy, 2012).
Let us turn to some of the elements
used in the OECD’s definition. Finan-
cial knowledge relates to the possession
of know-how and understanding in
particular of one’s personal financial
matters and of important everyday
financial concepts. It thus refers to a
rather basic form of financial literacy.
Furthermore, actual financial know-
ledge also influences the way people
themselves assess their understanding
of finance matters, i.e. perceived finan-
cial knowledge, as well as their finan-
cial skills (Hung et al., 2009). People’s
actions, i.e. their financial behavior, in
turn, depend on all three elements
(actual knowledge, perceived knowl-
edge, and skills). Finally, the experi-
ence people gain through financial be-
havior feeds back to both their actual
and perceived financial knowledge.
Still, the relationships are likely to be
imperfect, as each of these elements
also depends on other factors internal
and external to the individual, such as
attitudes or resources.
Even the most comprehensive defi-
nition of financial literacy leaves several
crucial questions unsolved: How are
the different elements of financial liter-
acy interrelated? Which elements con-
tribute most to effective decision mak-
ing? Is financial literacy a construct of
its own and in which way does it differ
from reading skills and numeracy (Sälzer
and Prenzel, 2014)? Is there a clear dis-
tinction between financial literacy and
the broader concept of economic liter-
acy (Retzmann et al., 2010; Retzmann
and Frühauf, 2014)?
Despite this ongoing discussion
within the scientific community, most
studies on financial literacy have focused
on measuring financial knowledge of
individual persons. By contrast, values,
attitudes and behavior as well as a
broader perspective on critical eco-
nomic thinking have played a compara-
tively minor role.
In our research, we adopt the
OECD approach to measuring financial
literacy (Atkinson and Messy, 2012)
and include not only financial knowl-
edge, but also financial behavior and
attitudes as highly relevant elements of
financial literacy. This decision is based
on research showing a relationship
between people’s attitudes toward
money and the extent to which they
face financial problems (Barry, 2014),
as well as between their money atti-
tudes and level of satisfaction with their
financial situation (Shim et al., 2009).
The empirical financial literacy re-
search can be classified into three broad
categories: One strand of literature
deals with the design and use of meth-
ods to assess financial knowledge (e.g.
Knoll and Houts, 2012). Nowadays,
most studies use survey techniques
based on questionnaires asking respon-
dents various knowledge questions,
with both open and multiple choice
answers. Some of the questions have
become well established and have been
used in several countries around the
world, providing international bench-
marks (Lusardi and Mitchell, 2011a and
2011b; Atkinson and Messy, 2012).
Nevertheless, the validity of these items
is doubtful: they possibly cover knowl-
edge that is not necessarily indicative of
a respondent’s capability to take sound
financial decisions. For instance, a per-
son’s ability to properly calculate inter-
est rates may not reveal much about his
or her success in dealing with financial
decisions and products (Greimel-Fuhr-
mann, 2014).
A second strand of literature inves-
tigates the personal characteristics that

Financial literacy gaps of the Austrian population
MONETARY POLICY & THE ECONOMY Q2/15 37
influence financial knowledge (see e.g.
Atkinson and Messy, 2012). Lusardi
and Mitchell (2014) summarize the ex-
isting empirical literature. The sur-
veyed studies consistently identify im-
portant knowledge gaps, especially in
more sophisticated financial topics,
such as real interest, compound interest
or the effect of exchange rate changes.
They moreover show that financial
knowledge correlates with socio-demo-
graphic characteristics: knowledge in-
creases with education level and in-
come. Men tend to achieve higher
scores in financial knowledge tests than
women. This gender gap remains even
after correcting for differences in edu-
cation, profession or income and may
be partly explained by women’s differ-
ent response behavior in survey tests.
Studies examining whether this gender
gap is ascribable to couples’ division of
financial responsibilities, proxied by
marital status and budget responsibil-
ity, have yielded rather inconclusive
results.
Respondents with a migration back-
ground tend to show relatively weak
test results. The link between age and
knowledge appears to be U-shaped,
with younger and older participants
performing particularly poorly. In the
case of younger respondents, this may
be due to missing or scant experience
of work life and in financial markets.
The elderly, on the other hand, may not
be familiar with new financial products
and/or lack the skills and technological
tools necessary to understand them.
Respondents from large cities tend to
attain better scores than respondents
from rural areas.
The third main strand of literature
investigates the link between financial
knowledge and behavior. Atkinson and
Messy (2012) for example find that re-
spondents with greater financial knowl-
edge show more competent behavior.
According to Gerardi et al. (2010),
limited financial literacy was one of the
drivers of mortgage delinquency and
default in the U.S. subprime mortgage
market. Lusardi and Tufano (2009)
confirm that the least financially savvy
use high-cost borrowing and pay higher
fees. Brown and Graf (2013) show that
financial literacy is positively related to
retirement planning in Switzerland.
Klapper et al. (2013) point out that
Russians with elevated financial liter-
acy participate more actively in formal
financial markets and are more likely to
have unspent income left, which enables
them to better deal with macroeco-
nomic shocks. Beckmann and Stix
(2015) find that a majority of respon-
dents is aware that depreciations in-
crease loan installments and that knowl-
edge about exchange rate risk greatly
impacts on the choice of the loan cur-
rency.
Few papers describe field experi-
ments corroborating the effectiveness
of financial training measures. Lühr-
mann et al. (2012) for example show
that financial training increases not
only German teenagers’ actual finan-
cial knowledge, but also their interest
in financial matters as well as their
self-assessed knowledge. They are
moreover less prone to make impulse
purchases and are better at identifying
the riskiness of assets.
2 A novel dataset on Austrians’
financial knowledge
The financial literacy survey this article
focuses on was carried out among about

Financial literacy gaps of the Austrian population
38 OESTERREICHISCHE NATIONALBANK
2,000 individuals in Austria aged 15
and above in the fall of 2014.2 It con-
sisted of face-to-face interviews cover-
ing questions on financial knowledge,
several aspects of financial behavior and
attitudes as well as a rich set of so-
cio-demographic variables. The survey
was part of an initiative of the OECD’s
INFE working group to run a unified
and extended survey on financial liter-
acy, behavior and attitudes in a wide
range of countries. A pilot study had al-
ready been conducted in 14 countries
in 2010 and 2011, in which Austria had
not participated, however.
The questions of the OECD survey
were based on previous international
research and were designed to be of rel-
evance for people from very different
backgrounds in a wide range of coun-
tries (Atkinson and Messy, 2012). The
questionnaire consisted of open and
multiple choice questions. In addition
to the full set of mandatory question
and a wide choice of optional items
(OECD, 2013), the Austrian survey3
also included some supplementary
questions that we considered to be of
special relevance in the Austrian con-
text. For details on the socio-demo-
graphic distribution of the sample, see
table A1 in the annex.
2.1 Survey identifies important
knowledge gaps
In the survey, 11 questions covered var-
ious aspects of financial knowledge.
They test the understanding of basic
and more sophisticated economic and
financial concepts, such as inflation, in-
terest rates, the risk-return link and ex-
change rates. None of the questions re-
quired expert knowledge. Box 1 shows
the 11 questions in detail.4
2 The survey comprised 1,994 computer-assisted personal interviews (CAPIs) from October to November, 2014. The
non-response rate was about 30%. If not indicated differently, we use survey weights to produce descriptive
population statistics throughout the paper. The weights consist of a combination of sampling/design weights and
poststratification weights based on external population statistics on age and gender at the province level.
3 The full questionnaire is available from the authors upon request.
4 While most of these survey questions have been repeatedly used in research and are thus especially suited for
international comparison, they share a general validity caveat: Is the ability to perform simple calculations and
understand basic concepts, such as interest or inflation, really the relevant prerequisite for responsible and sound
financial behavior? Or are general attributes, e.g. being a well-organized and forward-looking person, or interest
in financial issues more relevant for financial success?
Box 1
Questions on financial knowledge in the survey
The survey of financial literacy covers 11 questions on financial knowledge, which we classif ied
– for later use – into basic questions (“B”) and sophisticated questions (“S”). The split follows
intuition, but is also supported by the share of correct answers shown in chart2. The correct
answers are given in bold at the end of each question.
B “Division”: Imagine that five brothers are given a gift of EUR1,000. If the brothers
have to share the money equally, how much does each one get? (EUR200)
S “Inflation”: Now imagine that the brothers have to wait for one year to get their share
of the EUR 1,000 and inflation stays at 2%. In one year’s time, will they be able to buy (a)
more with their share of the money than they could today, (b) as much as today, (c) or less
than they could buy today? (c)1
S “Real interest rate”: Imagine that the interest rate on your savings account was 1% per
year and inflation was 2% per year: After one year, how much would you be able to buy with
the money in this account? (a) More than today, (b) exactly the same, (c) less than today. (c)
1 If respondents came up with the alternative answer “It depends on the types of things they want to buy” (not part of
the set of options), this answer was counted as correct.

Financial literacy gaps of the Austrian population
MONETARY POLICY & THE ECONOMY Q2/15 39
Chart 1 shows that 4% of the re-
spondents answered all 11 questions
correctly and a total of 36% were able
to give the correct answer to at least
9 questions. Furthermore, about 11%
of the respondents answered at least
5 questions incorrectly (not including
“don’t know” answers).
Taking a closer look at the responses
to the 11 knowledge questions, we find
comparatively low levels of financial
knowledge for some of the questions.
B “Zero interest”: You lend EUR25 to a friend one evening and he gives you EUR25
back the next day. How much interest has he paid on this loan? (0)
S “Interest after 1 year”: Suppose you put EUR100 into a no-fee savings account with
a guaranteed interest rate of 2% per year. You do not make any further payments into this
account and you do not withdraw any money. How much would be in the account at the end
of the first year, once the interest payment is made? (102 EUR)
S “Interest after 5 years”: And how much would be in the account at the end of five
years? (a) More than EUR110, (b) exactly EUR110, (c) less than EUR110, (d) It is impossible
to tell from the information given. (a)
S “Exchange rate”: Suppose you have taken out a loan in Swiss francs. Then the
exchange rate of the euro depreciates against the Swiss franc. What happens to the euro
amount you owe? (a) It increases, (b) it stays exactly the same, (c) it decreases. (a)
S “Interest – bond”: If interest rates rise, what will typically happen to bond prices? (a)
They will rise, (b) they will fall, (c) they will stay the same, (d) there is no relationship between
bond prices and the interest rate. (b)
B “Risk – return”: Is the following statement (a) true or (b) false? An investment with a
high return is likely to be high risk. (a)
B “Cost of living”: Is the following statement (a) true or (b) false? High inflation means
that the cost of living is increasing rapidly. (a)
S “Risk diversification”: Is the following statement (a) true or (b) false? It is usually
possible to reduce the risk of investing in the stock market by buying a wide range of stocks
and shares. (a)
%
18
16
14
12
10
8
6
4
2
0
0 1 2 3 4 5 6 7 8 9 10 11
Percentage of respondents answering x financial knowledge questions correctly
Chart 1
Source: OeNB.
Note: N = 1,994. The bars add up to 100%.
Correct answers per respondent

Financial literacy gaps of the Austrian population
40 OESTERREICHISCHE NATIONALBANK
Very basic questions that require an un-
derstanding of division, interest rates,
risk diversification or inflation were
answered correctly by a vast majority
of respondents. The portion of correct
answers exceeded 80% and the net
number of correct answers (i.e. correct
minus false answers) was also high. For
example, 94% of the respondents man-
aged to correctly divide 1,000 by five
(“Division”). 85% of them knew that
the interest rate is zero when you lend
someone EUR25 and get back the same
amount the next day (“Zero interest”).
The same fraction understood that
higher return is usually associated with
higher risk (“Risk – return”). And 84%
of the respondents understood the con-
cept of inflation (“Cost of living”).
However, with slightly more com-
plex questions, we observe that the
number of correct responses declined
markedly, to well below 80%. Also,
the number of wrong answers increased
sharply, and more people stated that
they did not know the answer:
• The concept of real interest rates, i.e.
interest minus inflation, was under-
stood by 71% of the respondents,
while 22% failed this question (“Real
interest rate”).
• 67% correctly answered the question
about the amount of money you are
left with after setting EUR100 aside
for one year in an account with an
interest rate of 2% (“Interest after
1year”).
• 65% of the respondents understood
that inflation implies that over time
one can buy less with a fixed amount
of money (“Inflation”).
• 61% of the respondents understood
that investment risk can be reduced
by buying a wide range of stocks and
shares (“Risk diversification”).
• 54% of the respondents understood
the effect of exchange rate fluctua-
tions on a foreign currency loan (“Ex-
change rate”). Note that this share in-
creases to 89% if we only consider
those 22 respondents that indicated
that they have a foreign currency
loan.
For questions with an even more
advanced level of sophistication, the net
correct answers, i.e. the number of
correct answers minus the number of
false responses, is actually negative:
43% of the respondents grasped the
concept of compound interest, i.e. the
fact that leaving an amount of money in
your account for five years usually gives
you interest not only on the principal,
but also on the interest gained in the
past. At the same time, 44% failed on
that question (“Interest after 5 years”).5
%
Division
Zero interest
Risk – return
Cost of living
Real interest rate
Interest after 1 year
Inflation
Risk diversification
Exchange rate
Interest after 5 years
Interest – bond
0 10 20 30 40 50 60 70 80 90 100
Answers to 11 financial knowledge questions
Char t 2
Source: OeNB.
Note: N = 1,994.
Correct answer Incorrect answer Don’t know
94
85
85
84
71
67
65
61
54
43
21
3
6
3
6
22
16
29
14
28
44
40
3
9
12
10
8
17
7
26
18
12
39
5 The exact wording of the question on “Interest after 5 years,” taken from the OECD questionnaire, does not
unambiguously specify that annual interest payments are credited to the same account. Yet, it is not uncommon
that interest is credited to a different account. Hence, we cannot rule out that this has biased the results.

Financial literacy gaps of the Austrian population
MONETARY POLICY & THE ECONOMY Q2/15 41
The question on the link between in-
terest rates and bond prices produced
even fewer correct answers (21%),
while 40% failed (“Interest – bond”).
Furthermore, the number of respon-
dents who said that they did not know
the answer was as high as 39%, which
may in part have been due to their
unfamiliarity with the term bond. If
we only consider the bondholders
among the respondents, the fraction
knowing the correct answer increases
to 41% (as compared to 21% for the
entire sample).
Overall, this first evaluation shows
that there are remarkable knowledge
gaps as regards most of the questions.
While the majority of respondents are
familiar with basic concepts such as in-
terest rates or inflation, other import-
ant concepts such as real interest, com-
pound interest, the link between inter-
est rates and bond prices or the effect of
exchange rate fluctuations on foreign
currency loans are not understood by at
least one-quarter of the population.
This is a source of concern, given that
most Austrians are – either directly or
indirectly – consumers of financial
products that involve these concepts.
Table 1 compares the results of our
survey and of the OECD pilot study
(Atkinson and Messy, 2012) for the
eight survey questions that were used
in both questionnaires.6 Austrian re-
spondents appear to have relatively
good financial knowledge in compari-
son with the respondents (mean) from
the 14 countries that participated in the
OECD pilot study. 67% of the Austrian
respondents answered six of the eight
questions correctly compared with
51% in the OECD sample average. Be-
sides, the average number of correctly
answered questions is 73% in Austria,
versus 66% in the OECD sample aver-
age. However, these results should be
interpreted against the background of
the selection of countries surveyed by
the OECD: most of them are less
developed than Austria. When com-
paring the Austrian survey results with
countries of a similar level of economic
development, we find that the results
are more mixed: Austria’s score, for
instance, broadly matches the German
results of the pilot study, but is outper-
formed by Hungary.
2.2 The impact of socio-demographic
characteristics on financial
knowledge
The new dataset clearly points to knowl-
edge gaps among the Austrian popula-
tion. These vary, however, substantially
among different socio-demographic
groups. Overall, we are able to confirm
many of the empirical findings summa-
rized by Lusardi and Mitchell (2014).
We use the question on interest after
one year for a detailed breakdown of
respondents’ scores based on socio-
demographic criteria. Chart3, hence,
6 Questions 1, 2, 4, 5, 6, 9, 10 and 11 listed in box 1.
Tab le 1
Comparison of Austrian data with data from other OECD
countries1
Mean of correct answers to
the 8 knowledge questions Correct answers to
6 or more questions
%
Austria 73 67
OECD sample average266 51
Germany 71 58
United Kingdom 69 53
Norway359 40
Hungary 77 69
Source: OECD, authors’ calculations.
1 See Atkinson and Messy (2012).
2 T he 14 countries participating in the OECD pilot study were: Alb ania, Armenia, Czech Republic, Estonia,
Germany, Hungary, Ireland, Malaysia, Norway, Peru, Poland, South Africa, United Kingdom, British Virgin
Islands.
3 Norway reformulated the questions slightly.

Financial literacy gaps of the Austrian population
42 OESTERREICHISCHE NATIONALBANK
reveals that men score better than
women and that the level of education
and income appears to be of high rele-
vance for scores. More specifically,
83% of university or college graduates
know the correct answer to the ques-
tion, while the fraction is as low as 46%
for those having only completed com-
pulsory education. In this subgroup,
the share of respondents who do not
know the answer is high, too (34%).
The respective charts for the ten other
knowledge questions broadly confirm
this picture.
To assess whether these differences
are statistically significant while hold-
ing other variables constant, we con-
struct a financial knowledge score by
counting the number of correct re-
sponses of each respondent, following
Atkinson and Messy’s approach (2012).
We then regress this knowledge score
(as dependent variable) on the follow-
ing independent variables: a dummy for
gender; age and age squared (testing
the U-shaped link suggested by the lit-
erature); dummies for low education,
medium/vocational education and gen-
eral/high education; dummies for the
income level (low/medium/high); a
dummy for the employment status (em-
ployed full time/unemployed/not
working/employed part time/self-em-
ployed); and a dummy for town size
(small/medium/large = Vienna).
The regression results are shown in
table 2. The asterisks indicate the level
of statistical significance. The first
three columns use different subsets of
the 11 questions: The first column
shows the regression results when the
dependent variable is the number of
correct answers to the total set of
11 knowledge questions presented in
box1. The second column focuses ex-
clusively on the correct answers to the
four most basic questions, namely the
first four questions in chart2 or those
indicated with a “B” in box 1. The
dependent variable in the third column
is the number of correct answers to the
seven more sophisticated knowledge
questions, marked with an “S” in box1.
The coefficients in most rows need to
be interpreted against a benchmark,
which is shown in italics. For example,
a positive coefficient for males indicates
that test scores of men are higher than
those of the benchmark “female.”
Overall, the results are largely in
line with the findings of previous
empirical research. Men achieve scores
%
TOTAL
GENDER
Male
Female
AGE
15–29 years
30–44 years
45–59 years
60 years and older
EDUCATION
Compulsory education
Apprenticeship
Professional education1
Secondary education2
University or college degree
HOUSEHOLD INCOME
Below EUR 2,000
EUR 2,000–3,400
Over EUR 3,400
0 10 20 30 40 50 60 70 80 90 100
Answers to the question on interest after one year
Chart 3
Source: OeNB.
Note: N = 1,994.
1 Technical, commercial or vocational education not qualifying for university.
2 General or vocational secondar y education qualifiying for univer sity.
Correct answer Incorrect answer Don't know
67
69
65
69
67
66
66
46
66
73
72
83
63
72
77
16
17
15
16
20
17
13
20
18
12
18
11
15
15
14
17
14
20
15
14
17
22
34
16
16
11
7
22
13
9

Financial literacy gaps of the Austrian population
MONETARY POLICY & THE ECONOMY Q2/15 43
that are significantly higher than those
of women, as was the case in almost all
of the 14 countries of the OECD pilot
study (Atkinson and Messy, 2012). As
we control for a comprehensive set of
socio-demographic characteristics, this
difference cannot be attributed to
gender differences in education, in-
come or labor force status. Further-
more, an alternative regression (not
shown here) using the number of incor-
rect answers as dependent variable like-
wise indicates that men outperform
women. In other words, men show a
lower likelihood of giving an incorrect
answer. By contrast, women admit sig-
nificantly more often than men that
they do not know the answer. Overall,
we find that women are less likely to
answer the financial knowledge ques-
tions correctly and that they are more
likely to answer them incorrectly or
with “don’t know.” This gender gap
is sub ject to future investigation in
Greimel-Fuhrmann et al. (2015a).
Our study also confirms the non-
linear shape of the link between knowl-
edge and age, which indicates that the
youngest and the oldest respondents
perform worst. Knowledge scores are
maximized at an age of 51. The effect of
the highest educational attainment on
knowledge is also highly significant.
Respondents who have (at most) only
finished compulsory education score
significantly worse than those having
completed medium-level vocational ed-
ucation. Graduates from general higher
schools (including tertiary education),
by contrast, perform significantly better
Tab le 2
OLS regression on financial literacy
All 11 questions Basic questions Sophisticated
questions Self-assessment
Male 0.49*** 0.10 * 0.40*** 0.06
Female
Age 0.06*** 0.03*** 0.04** 0.02**
Age squared –0.00*** –0.00** –0.00** –0.00*
Education – low1–0.99*** –0.38*** –0.60*** –0.41***
Education – vocational training2
Education – high30.68*** 0.18*** 0.51*** 0.22***
Income below EUR 2,000 –0 .32** – 0.07 –0. 25** –0.19***
Income EUR 2,000–3,400
Income over EUR 3,400 0.19 0.03 0.17 0.20**
Employed (more than 35 hours per week)
Unemployed 0.26 0.02 0.24 0.06
Not working 0.27 – 0.03 0.30* –0.04
Employed part time (max. 35 hours per week) 0.12 0.02 0.10 0.11
Self-employed 0.43* 0.09 0.34* 0.12
Town size (up to 5,000 inhabitants) –0.13 0.15** – 0.29** 0.06
Town size (5,000 to 1 million inhabitants) 0.13 0.24*** – 0.11 – 0.12
Town size (Vienna)
Constant 5.65*** 2.71*** 2.95*** – 0.26
N 1994 1994 1994 1994
R20.11 0.08 0.10 0.09
Source: Authors’ calculations.
1 At most completed compulsory education.
2 Apprenticeship; technical, commercial or vocational education not qualifying for university.
3 Secondary academic or vocational education qualifying for university education; other education qualifying for university education; university of applied sciences; technical or vocational
college; university: bachelor’s, master’s and doctoral degrees.
* p < 0.05, ** p < 0.01, *** p < 0.001
Note: Variables in italic s are benchmark s for dummy variables. The regression is based on unweighted survey data.

Financial literacy gaps of the Austrian population
44 OESTERREICHISCHE NATIONALBANK
than the intermediate benchmark group.
This result suggests that the payoff of
offering financial education measures
to those with the lowest education lev-
els is likely to be especially high.
There is furthermore some evi-
dence that lowest-income respondents
perform worse than those with me-
dium or high income. The former may
have less flexibility to “learn by doing”
as they cannot afford to make mistakes
(Atkinson and Messy, 2012). There is
no significant difference between me-
dium- and high-income earners. The
employment status is not significant
throughout all regressions with the ex-
ception of self-employment, which
appears to be associated with higher
financial knowledge.
The evidence on the town size for
the entire set of 11 questions is insignif-
icant. However, when we split the set
into basic and more sophisticated ques-
tions, we find that respondents from
Vienna outperform those from smaller
towns or rural areas in the more so-
phisticated questions. By contrast, the
former perform worse in the more
basic questions.
Throughout all regression models,
the value of R2 is rather low. This means
that we can only explain about 10% of
the variation in financial knowledge
scores by the socio-demographic vari-
ables that we use. This is a common
feature of microdata studies on finan-
cial knowledge (see e.g. Brown and
Graf, 2013; Bucher-Koenen and Lamla,
2014; Fonseca et al., 2014).7 It is not
the purpose of our model to predict lit-
eracy scores, however. Instead, we aim
at identifying the least knowledgeable
portions of the population to be able
to target them especially via new tailor-
made financial education measures.
3 Respondents’ self-assessment
of financial knowledge
The level of self-confidence in financial
matters may shape financial behavior,
and also how confidently people answer
knowledge questions, thus avoiding the
“don’t know” option. In our survey, we
explicitly ask people to self-assess their
knowledge:
“Self-assessment”: How would you rate
your level of financial knowledge on a scale
of 1 to 5? (1 = not at all knowledgeable,
5 = very knowledgeable).
The answers to this question reveal that
respondents are broadly aware of exist-
ing knowledge gaps. The average
self-assessment is 3.3, which is only a
little higher than the “neutral” value 3.
The mean for men is slightly higher
than that for women (3.31 versus 3.21).
At the same time, the self-assessment
scores also increase with the financial
knowledge measure, as also found by
Van Rooij et al. (2011). This correlation
may be the result of causality that runs
both ways, as higher knowledge can be
assumed to support self-confidence in
one’s own financial capabilities, while
self-confidence, in turn, may give re-
spondents the courage to answer diffi-
cult questions instead of refusing,
which potentially raises the test score.
Overall, women appear to have a more
realistic perception of their own finan-
cial capabilities, as indicated by the
higher correlation coefficient compared
with men (0.32 versus 0.28).
Chart 4 confirms the positive rela-
tionship between perceived financial
7 We experimented with several additional explanatory variables that are common in the literature, such as migra-
tion background, marital status or a proxy for the risk aversion of the respondent. The coefficients are, however,
not significant, while the major results remain unchanged.

Financial literacy gaps of the Austrian population
MONETARY POLICY & THE ECONOMY Q2/15 45
knowledge and realized knowledge
scores. In this chart, we juxtapose the
average self-assessment of respondents
with the lowest level of financial liter-
acy (defined here as less than 4 correct
answers), those with the highest knowl-
edge score (at least 9 correct answers)
and those with exactly 6 correct an-
swers, always differentiating between
men and women.
However, in line with the findings
presented by Lusardi and Mitchell
(2014), we can also see from chart 4
that respondents with low financial
knowledge overestimated their liter-
acy, which was more pronounced in the
case of men: Respondents who were
not able to answer more than three
questions correctly still showed a rather
elevated self-assessment, with an aver-
age score of 2.7 (men) and 2.4 (women),
only slightly below the neutral value 3.8
Unrealistic perceptions of one’s finan-
cial knowledge may motivate people to
choose complex products such as for-
eign currency loans or repayment vehi-
cles without properly understanding
their properties and risks.
Column 4 in table2 shows that per-
ceived financial knowledge depends on
roughly the same variables as the
knowledge score itself: People with
higher educational attainment not only
have greater financial knowledge but
are also very much aware of this. Vice
versa, lower-income groups also have a
lower level of self-confidence in finan-
cial matters. We also observe similar
age effects. Most interestingly, how-
ever, the gender gap disappears when it
comes to the self-assessment; the coef-
ficient is no longer significant. Further
research would be needed to under-
stand this absence of a gender gap.
4 The link between financial
knowledge and financial
behavior
Financial education measures usually
rely on the assumption that people with
a higher level of financial knowledge
take better investment and financing
decisions. Investigating the relationship
between knowledge and behavior via
regression analysis is complicated given
the possibility of reverse causalities,
which could potentially introduce an
endogeneity bias into regression analy-
ses. For example, we may find it desir-
able that people carefully check several
options before taking financial deci-
sions. A higher level of financial knowl-
edge may indeed induce people to com-
pare more financial products and to
consider a variety of sources of infor-
mation before arriving at a decision.
On the other hand, the process of com-
paring products and using information
may in turn contribute to financial
knowledge.
A thorough analysis of the link be-
tween financial knowledge and behav-
ior would be beyond the scope of this
Mean (scale from 1 to 5)
4.0
3.5
3.0
2.5
2.0
1.5
1.0
Self-assessment versus realized
financial knowledge
Chart 4
Source: OeNB.
Male Female
2.7
3.4 3.6
2.4
3.3
3.6
Less than 4 correct answers
N(male)= 56
N(female)= 94
Exactly 6 correct answers
N(male)= 117
N(female)= 148
At least 9 correct answers
N(male)= 398
N(female)= 330
8 Note that in the survey the question on self-assessment was asked before the 11 knowledge questions. Respondents
thus assessed their knowledge before answering the test questions.

Financial literacy gaps of the Austrian population
46 OESTERREICHISCHE NATIONALBANK
paper and is subject to a special investi-
gation in Greimel-Fuhrmann et al.
(2015b).9 In charts 5 and 6, we never-
theless show tentative evidence corrob-
orating that greater financial knowl-
edge may indeed have an impact on fi-
nancial behavior. Our survey
questionnaire includes several ques-
tions that may be used to assess whether
respondents show sound financial be-
havior. The full list of questions is
shown in box2. This list includes ques-
tions on how people select financial
products, whether they put enough
money aside for tough economic times
and retirement, and whether they take
out a loan for economically sound rea-
sons.
9 To our knowledge only a few studies try to solve this endogeneity problem via instrumental variable estimation.
Lusardi and Mitchell (2014) conclude that the link between financial knowledge and behavior tends to get
stronger if instruments are used, implying that the noninstrumented estimates underestimate the true effect.
Box 2
Questions on financial behavior
The survey includes six questions that we use to investigate financial behavior. At the end of
each question we define what we consider to be positive f inancial behavior, i.e. behavior that
reflects well-thought-out decisions on f inancial products and keeps people from falling into
debt traps in difficult economic circumstances.
“Knows many financial products”: Can you tell me whether you have heard of any of
these types of financial products? (a) Current account, (b) savings book, (c) building loan con-
tract, (d) personal pension fund, (e) investment fund, (f) stocks and shares, (g) bonds, (h)
financial derivatives, (i) insurance policies, (j) mobile phone payment account, (k) prepaid
payment card, (l) microf inancing/crowd financing (Positive: knows several financial
products, measure: number of financial products known)
“Pension funds”: Do you currently hold a personal pension fund? (Positive: yes, mea-
sure: dummy = 1 if yes, 0 otherwise)
“Shopping around”: Which of the following statements best describes how you last
chose a product? (a) I considered several products from different companies before making
my decision, (b) I considered the various products from one company, (c) I didn’t consider any
other products at all, (d) I looked around but there were no other products to consider. (Pos-
itive: considers other products, measure: dummy = 1 if (a), (b) or (d), 0 otherwise)
“Sources of information”: Which of the following sources of information had the
greatest inf luence on your most recent financial product choice? (a) Unsolicited information
sent through the mail, (b) Information picked up in a branch, (c) Product-specif ic information
found on the Internet, (d) Information from sales staff of the firm providing the products, (e)
Best-buy tables in f inancial pages of newspapers/magazines, (f) Best-buy information found
on the Internet, (g) Specialist magazines/publications, (h) Recommendation from independent
financial adviser or broker, (i) Advice from friends/relatives not working in the financial services
industry, (j) Advice from friends/relatives working in the financial services industry, (k) Employ-
er’s advice, (l) Newspaper articles, (m) Television or radio programs, (n) Newspaper advertise-
ments, (o) Television advertisements, (p) Other advertising, (q) Own experience, (r) Other
source (Positive: uses various sources of information, measure: number of sources
of information used)
“Reason for taking out a loan”: Have you personally used credit (e.g. loan, leasing,
installments or overdraft) in the past 12 months for any of the following purposes and paid
interest on the balance? (a) To pay regular bills, (b) For everyday spending, (c) To help support
family or friends outside your immediate household, (d) To buy something on impulse, (e) To
buy a gift for someone, (f) To buy a house or pay for renovation, (g) To pay for holidays, (h) To

Financial literacy gaps of the Austrian population
MONETARY POLICY & THE ECONOMY Q2/15 47
Charts 5 and 6 link the various
proxies for financial behavior to the
number of correctly answered knowl-
edge questions. Chart 5 shows that the
number of financial products a person
knows increases with financial knowl-
edge. Those who answer 11 questions
correctly know on average 10 different
financial products, while those with
low knowledge have heard of only about
six products. Similarly, those with the
highest knowledge score also consider
on average two different sources of
information when selecting a financial
product. Those with the lowest liter-
acy, in contrast, rely on just one source.
Knowing and considering a variety of
different products and sources can be
assumed to lead to more informed
financial decisions. Chart 5 also shows
that people with greater knowledge
have a lower tendency to take out a loan
for short-term purposes (e.g. buying
gifts, everyday spending or impulse
purchases or financing holidays) or for
risky purposes (e.g. supporting friends
or financing financial investment). Re-
lying on credit to fund basic day-to-day
purchases or risky investment may be
the starting point of a debt trap. About
two-thirds of all respondents did not –
or would not – take out a loan for any
such short-term or risky purpose.
Those with the lowest knowledge
scores (less than 6 correct answers)
listed on average more than one adverse
loan motivation. This number declines
to 0.5 for those who answered 10 or
11 questions correctly.
Chart 6 links the number of correct
answers to the share of respondents in-
dicating that they have a voluntary per-
Mean number Mean number
2.5
2.0
1.5
1.0
0.5
0.0
11
10
9
8
7
6
Financial behavior indicators: number of ...
Chart 5
Source: OeNB.
… sources of information used when making a financial decision
… adverse reasons for taking out a loan
… financial products the respondent knows of (scaling on right-hand side)
Correct answers per respondent
0 1 2 3 4 5 6 7 8 9 1110
buy a car, (i) To pay for education or training (for you or a family member), (j) To invest in your
business or in the business of a family member, (k) For financial investment (Negative: (a),
(b), (c), (d), (e), (g), (k), measure: number of negative reasons for taking a loan)
“Rainy day funds”: If you lost your main source of income, how long could you continue
to cover living expenses, without borrowing any money or moving house? (a) Less than a week,
(b) At least a week, but not one month, (c) At least one month, but not three months, (d) At
least three months, but not six months, (e) More than six months (Positive: has at least
three months of funds set aside, measure: dummy = 1 if (d) or (e), 0 otherwise)

Financial literacy gaps of the Austrian population
48 OESTERREICHISCHE NATIONALBANK
sonal pension fund and rainy day funds
of at least three months of living ex-
penses and who compare alternative of-
fers before choosing a financial prod-
uct. The evidence is less clear here.
People with higher knowledge have
more rainy day funds: among those
with the lowest literacy scores, around
40% say they have sufficient funds set
aside to cover living expenses for at
least three months; this fraction is as
high as 65% for those with top scores.
When we factor out those with the
lowest knowledge scores – subgroups
where we only have a couple of dozens
of respondents (see chart 1) –, there
also appears to be a positive link be-
tween literacy and the pension fund
question: 17% of the top performers
have a private pension fund, which
compares with only 5% for those who
answered three to four questions cor-
rectly. The link to pension funds is less
clear cut than in the literature (e.g.
Brown and Graf, 2013), which may
relate to the fact that, given Austria’s
comprehensive public pension system,
pension funds are generally not much
sought after. There appears to be no
clear link between financial knowledge
scores and the tendency to compare
alternative financial offers.
5 Summary and implications
Overall, our results reveal that – like in
many other countries – there are re-
markable financial knowledge gaps
among the Austrian population, espe-
cially among women, younger and
older people as well as those with a low
level of education. Our analysis also
suggests that this lack of financial knowl-
edge may be conducive to risky finan-
cial behavior, such as insufficient saving
for bad times or retirement, basing finan-
cial decisions on little advice or com-
parison or taking out loans for adverse
reasons, e.g. impulse purchases or gifts.
Prudent and sound financial deci-
sions by financial market participants
support central banks in fulfilling their
financial stability mandate. Further-
more, it is easier for central banks to
communicate monetary policy deci-
sions if the population has a high level
of financial literacy. A better under-
standing of monetary policy helps pre-
dict forthcoming decisions, which in
% %
80
70
60
50
40
30
20
10
0
40
35
30
25
20
15
10
5
0
Financial behavior indicators: share of respondents who state they ...
Chart 6
Source: OeNB.
… shop around before choosing a financial product
… have sufficient rainy day funds
… have a pension fund (scaling on right-hand side)
Correct answers per respondent
0 1 2 3 4 5 6 7 8 9 1110

Financial literacy gaps of the Austrian population
MONETARY POLICY & THE ECONOMY Q2/15 49
turn speeds up the transmission to the
real economy and hence makes mone-
tary policy more effective.
It is therefore in the clear self-inter-
est of central banks to provide targeted
and tailored financial education, espe-
cially to the most vulnerable groups of
the population, either directly or via
multipliers, such as teachers, journal-
ists or financial service providers. As
independent and non-commercial ex-
pert organizations, central banks are
also natural leaders in financial educa-
tion activities. The results of the new
dataset on the knowledge gaps of the
Austrian population are an important
contribution to shaping the future finan-
cial education activities of the OeNB.
References
Atkinson, A. and F.-A. Messy. 2012. Measuring Financial Literacy: Results of the OECD/Inter-
national Network on Financial Education (INFE) Pilot Study. OECD Working Papers on Finance,
Insurance and Private Pensions 15. OECD Publishing.
Barry, D. 2014. Die Einstellung zu Geld bei jungen Erwachsenen. Eine Grundlegung aus
wirtschaftspädagogischer Sicht. Heidelberg: Springer.
Beckmann, E. and H. Stix. 2015. Foreign Currency Borrowing and Knowledge of Exchange
Rate Risk. In: Journal of Economic Behavior & Organization 112. 1–16.
Brown, M. and R. Graf. 2013. Financial Literacy and Retirement Planning in Switzerland.
Numeracy 6(2). Article 6.
Bucher-Koenen, T. and B. Lamla. 2014. The Long Shadow of Socialism: On East-West
German Differences in Financial Literacy. MEA Discussion Papers 282-2014.
Fonseca, R., K. J. Mullen, G. Zamarro and J. Zissimopoulos. 2014. What Explains the
Gender Gap in Financial Literacy? RAND Working Paper WR-762.
Gerardi, K., L. Goette and S. Meier. 2010. Financial literacy and subprime mortgage
delinquency: evidence from a survey matched to administrative data. Federal Reserve Bank of
Atlanta Working paper 2010-10.
Greimel-Fuhrmann, B. 2014. Financial Literacy – die schwierige Messung eines „Life Skill“.
Methodische und inhaltliche Überlegungen zur Messung des kompetenten Umgangs mit Geld
und Finanzen. Wissenplus Sonderheft Wissenschaft 5-13/14. 48–53.
Greimel-Fuhrmann, B., M. Silgoner and R. Weber. 2015a. Is there a gender gap in the fi-
nancial literacy of the Austrian Population? Mimeo.
Greimel-Fuhrmann, B., M. Silgoner and R. Weber. 2015b. Linking financial behavior to
financial knowledge – evidence for the Austrian population. Mimeo.
Holzmann, R. 2010. Bringing financial literacy and education to low and middle income countries:
The need to review, adjust, and extend current wisdom. Social Protection Discussion Paper
1007. Washington, DC: World Bank.
Hung, A., A. Parker and J. K. Yoong. 2009. Defining and Measuring Financial Literacy. RAND
Wor k ing pap e r.
Klapper, L., A. Lusardi and G. A. Panos. 2013. Financial literacy and its consequences:
Evidence from Russia during the financial crisis. In: Journal of Banking & Finance 37. 3904–3923.
Knoll, M. and C. Houts, 2012. The Financial Knowledge Scale: An Application of Item
Response Theory to the Assessment of FinancialLiteracy. In: Journal of Consumer Affairs 46(3).
Fall 2012. 381–410.
Lührmann, M., M. Serra-Garcia and J. Winter. 2012. The effects of financial literacy
training: Evidence from a field experiment in German high schools. Beiträge zur Jahrestagung
des Vereins für Socialpolitik 2013: Wettbewerbspolitik und Regulierung in einer globalen
Wirtschaftsordnung – Session: Household Behavior, No. C12-V2.

Financial literacy gaps of the Austrian population
50 OESTERREICHISCHE NATIONALBANK
Lusardi, A. and O. S. Mitchell. 2011a. The Outlook for Financial Literacy. NBER Working
Paper 1707 7.
Lusardi, A. and O. S. Mitchell. 2011b. Financial Literacy around the World: An Overview.
NBER Working Paper 17107.
Lusardi, A. and O. S. Mitchell. 2014. The Economic Importance of Financial Literacy: Theory
and Evidence. In: Journal of Economic Literature 52(1). 5–44.
Lusardi, A. and P. Tufano. 2009. Debt literacy, financial experiences and overindebtedness.
NBER Working Paper 14808.
OECD. 2013. OECD/INFE toolkit to measure financial literacy and inclusion: guidance, core
questionnaire and supplementary questions. September.
Retzmann, T., G. Seeber, B. Remmele and H. C. Jongebloed. 2010. Ökonomische
Bildung an allgemeinbildenden Schulen. Essen / Lahr / Landau / Kiel.
Retzmann, T. and F. Frühauf. 2014. „Financial Fitness for Life“ – Reichweite und Grenzen
der US-amerikanischen Testreihe für die finanzielle Allgemeinbildung. In: Retzmann, T. (ed.).
Ökonomische Allgemeinbildung in der Sekundarstufe I und Primarstufe. Schwalbach/Ts.:
Wochenschau Wissenschaft. 43–56.
Sälzer, C. and M. Prenzel. 2014. Financial Literacy im Rahmen der PISA-Studie. In: Retzmann,
T. (ed.). Ökonomische Allgemeinbildung in der Sekundarstufe I und Primarstufe. Schwalbach/Ts.:
Wochenschau Wissenschaft. 15–31.
Shim, S., J. J. Xiao, B. L. Barber and A. Lyons. 2009. Pathways to life success: A conceptual
model of financial well-being for young adults. In: Journal of Applied Developmental Psychology
30(6). 708–723.
Van Rooij, M., A. Lusardi and R. Alessie. 2011. Financial literacy and stock market participa-
tion. In: Journal of Financial Economics 101. 449–472.

Financial literacy gaps of the Austrian population
MONETARY POLICY & THE ECONOMY Q2/15 51
Annex
Tab le A1
Characteristics of respondents with especially high and low financial literacy
All respondents
(unweighted) All respondents
(weighted)1
Correc t ≥9
(weighted)1
Correc t ≤3
(weighted)1
%
TOTA L 35.69 8.42
GENDER
Male 46.64 48.12 42.46 6.67
Female 53.36 51.8 8 29.40 10.05
AGE
15–29 year s 16.70 20.93 29.11 9.96
30–44 years 23.77 24.54 33.90 8.03
45–59 years 27. 03 26.65 39.48 6.32
60 years and above 32.5 27. 8 8 38. 57 9.61
EDUCATION
Education – low216.15 17. 59 18 .20 20.09
Education – vocational training357.37 55.62 34.41 6.93
Education – high426.48 26.79 49.82 3.86
HOUSEHOLD INCOME
Income below EUR 2,000 42.43 41.05 31.17 10.07
Income EUR 2,000–3,400 37.71 38.1 3 9.11 5.30
Income over EUR 3,400 14.89 14.92 46.86 6 .19
Income not indicated 4.96 5.93 16 .80 22.67
EMPLOYMENT STATUS
Employed (more than 35 hours per week) 40.87 41.78 33.64 5.36
Unemployed 3.86 4 .14 34.69 7.80
Not working 38.01 36.03 34.97 11.38
Employed part time (max. 35 hours per week) 11. 53 11.67 38 .19 7.11
Self-employed 6.67 6.67 50.82 5.93
TOWN SIZE
Town size (up to 5,000 inhabitants) 43.78 43.97 32.95 9.82
Town size (5,000 –1 million inhabitants) 36.21 35.62 34.00 7.4 6
Town size (Vienna) 20.01 20.41 44.53 7.10
N 1,994 1,994 728 150
Source: OeNB.
1 Weights based on ex ternal population statistics on age and gender at the province level.
2 At most completed compulsory education.
3 Apprenticeship; technical, commercial or vocational education not qualifying for university.
4 Secondary academic or vocational education qualifying for university education; other education qualifying for university education; university of applied sciences; technical or vocational
college; university: bachelor’s, master’s and doctoral degrees.