Content uploaded by Eunyoung Baek
Author content
All content in this area was uploaded by Eunyoung Baek on Nov 15, 2019
Content may be subject to copyright.
EUNYOUNG BAEK Kyung Hee Cyber University
SHARON A. DEVANEY Purdue University*
How Do Families Manage Their Economic
Hardship?
Using data from the 2007 Survey of Consumer
Finances, this study examined how families
manage their economic hardship. A conceptual
model was developed based on risk management
theory and the permanent income hypothesis.
About half of families used credit and about a
third used their own savings to make up the
difference between income and spending. The
results of multinomial logit analysis showed that
families’ use of management methods differed
when they faced economic hardship, depending
on their situation.
BACKGROUND OF THE STUDY
The U.S. economy is undergoing the worst eco-
nomic crisis since the Great Depression of the
1930s (Moseley, 2009). Because of this eco-
nomic crisis, the national unemployment rate
has topped 10%, which is the highest since 1983
(Bureau of Labor Statistics, 2009). Not sur-
prisingly, unemployment is one of the events
that commonly leads to financial insecurity
in families (Sullivan, Warren, & Westbrook,
2000). The increase in unemployment has raised
Assistant Professor, Department of Asset Management,
Kyung Hee Cyber University, 1 Hoegi Dong, Dongdaemun
Gu, Seoul 130-701, Korea (eunibaek@khcu.ac.kr).
*Professor Emeritus, Department of Consumer Sciences and
Retailing, Purdue University, 80 E. Stirrup Trail, Monument,
CO 80132-7704.
Key Words: economic hardship, multinomial logit, unem-
ployment, use of credit, use of savings.
uncertainty about income, which has caused
financial problems for many families (American
Express, 2001), even middle-income families
(Sullivan et al.).
During hard times, many families might expe-
rience changes in income, especially declines
compared with a healthy economy. Most fami-
lies, however, do not anticipate an income inter-
ruption that might seriously strain their family’s
financial well-being (Carroll, Dynan, & Krane,
1999). Even when a family anticipates a decline
in income, financial planners claim that, in gen-
eral, Americans have not prepared themselves
well enough for a sudden loss of income (Welsh,
2001). Previous studies have documented that a
large proportion of families do not have the rec-
ommended level of emergency funds to prepare
for such unforeseen events (Chang, Hanna, &
Fan, 1997; Ding & DeVaney, 2000; Johnson &
Widdows, 1985).
Then, how do families manage their financial
difficulties? What if the family’s income is not
enough to cover all expenses? It is important to
know what types of methods families use to man-
age their economic distress and to know whether
their management behavior is rational. Although
the topic is important, little is known about how
families manage their finances when they face
economic hardship. Therefore, this study exam-
ines how families manage their finances when
they experience economic hardship and what
factors determine the choice of how they man-
age financially. When it comes to unexpected
financial difficulties, families usually take one
358 Family Relations 59 (October 2010): 358 – 368
DOI:10.1111/j.1741-3729.2010.00608.x
How Families Manage Economic Hardship 359
of two approaches: depleting their assets or using
credit (Eisenhauer, 1994; Vaughan, 1997). No
empirical examination, however, has been con-
ducted on which methods a family chooses for
managing their financial difficulties.
The purposes of this study are (a) to examine
families’ methods for managing economic hard-
ship, (b) to compare the characteristics of fami-
lies that are under economic hardship with those
that are not, and (c) to investigate the determi-
nants of the choice of method of managing their
finances when families face economic hardship.
The results will broaden understanding of
how families deal with their economic hard-
ship. As Gill and Ilahi (2000, p. 1) mentioned,
‘‘A systematic approach to social policy would
begin by understanding how families behave
when confronted with risk.’’ Therefore, a better
understanding of families’ behavior in response
to economic hardship will provide useful infor-
mation for policymakers, for financial advisors
and educators, and for the credit industry.
REVIEW OF LITERATURE
Theoretical Background
Although there are many theories explaining
families’ financial behaviors such as saving
and consumption, there is no single explicit
theory explaining how families behave when
they encounter economic hardship. Thus, in this
section, major concepts of financial risk man-
agement and the permanent income hypothesis
are reviewed to provide a theoretical framework
for this study.
Financial risk management. According to risk
management theory, there are two approaches
to managing financial risk: risk control and risk
financing. Risk control emphasizes minimizing
the problem by avoiding or lowering the
probability of risk. Risk financing focuses
on guaranteeing the availability of funds to
meet losses (Vaughan, 1997). Risk control is
primarily related to lowering the probability of
the occurrence of risk, whereas risk financing is
a method for dealing with risks that occur. This
study is primarily concerned with risk financing
because the focus of the study is to know which
methods would be used when dealing with
economic hardship that has already occurred.
Risk financing comes in two forms: risk
retention and risk transfer. Using risk reten-
tion, risk can be financed by accumulating funds
to meet expected loss or by allocating some
portion of the budget to meet uninsured losses.
Using risk transfer, risk can be financed by
outside resources that guarantee some funds
to meet losses. Purchasing an insurance pol-
icy is a primary approach to transferring risk
(Vaughan, 1997).
Therefore, families can purchase insurance
or accumulate savings to deal with unexpected
financial difficulty. Not all types of financial
risk such as an unexpected loss of income are,
however, insurable through market insurance, so
open-ended credit is an alternative for managing
such risk. In this case, using credit is one of the
risk transfer methods to manage such financial
risk. Using credit as a form of risk management
has gained popularity because the use of credit to
finance risk does not incur any expenses unless
a loss occurs (Eisenhauer, 1994).
Permanent income hypothesis. The permanent
income hypothesis is one of the major theories
that explains household consumption and sav-
ing. According to risk management theory, the
major methods of managing financial difficulty
of families involve using either savings or credit.
The permanent income hypothesis provides an
explanation when saving and dissaving (e.g.,
using credit) occur.
The permanent income hypothesis posits that
a family maximizes its utility in proportion to
its total lifetime resources (i.e., ‘‘permanent
income’’; Freidman, 1957). That is, a family’s
decisions about consumption depend on the fam-
ily’s permanent income, which is its expected
total lifetime income, rather than its current
income. Consequently, resources may need to
be transferred from periods of high income
to periods of low income to smooth the ability to
consume. In other words, this means that fami-
lies borrow in times of low income to retain the
desired level of consumption, and they will repay
past debts and save in times of high income to
provide for future consumption (Bryant, 1990).
The permanent income hypothesis explains that
the level of consumption is determined by the
permanent income stream. This assumes that
the gap between permanent income (total life-
time income) and temporary income (current
income) determines the changes in saving and
dissaving (e.g., borrowing; Freidman). Accord-
ingly, saving is positive if the gap is positive, and
saving is negative if the gap is negative (Bryant).
360 Family Relations
Conceptual framework. On the basis of the
concepts of risk management theory and the per-
manent income hypothesis, it is expected that
families will dissave either by depleting their
assets or by borrowing when they are faced with
unexpected financial difficulty. Depleting assets
can take the form of depleting liquid assets that
the family has accumulated and/or liquidating
nonliquid assets. Some families might choose to
increase debt by using credit rather than deplet-
ing their assets because of (a) the high transaction
costs frequently associated with liquidation of
assets or (b) the opportunity costs associated with
holding liquid ‘‘emergency funds’’ (Eisenhauer,
1994). To avoid transaction and opportunity
costs in pursuit of short-term liquidity, families
may see that using unsecured credit, such as that
offered by credit cards, is an attractive option
(Brito & Hartley, 1995). According to their sit-
uation and preferences, some families choose to
borrow and hold a lower level of emergency
funds, whereas others choose to accumulate
assets to prepare for unforeseen events.
Therefore, on the basis of these theories, the
following conceptual model was developed.
Yi=f(S
f,P
f)
Yirepresents a family’s method of managing
financial difficulty. According to risk manage-
ment theories, it is expected that using savings
(risk retention) or using credit (risk transfer)
could play a role to manage unexpected finan-
cial difficulty. In addition, the choice of method
of managing financial difficulty will depend on
the family’s situation (Sf,) including past unem-
ployment, financial needs and resources, and
preferences (Pf).
Related Literature
Regarding economic hardship, previous studies
have examined the relationship between eco-
nomic hardship and marital quality (Conger
et al., 1990; Johnson & Booth, 1990) and the
effect of economic hardship on the family mem-
ber’s adjustment (Conger et al., 1992). Dew
(2008) examined the relationship between debt
as an economic stressor and marital satisfaction
change. Very few studies, however, have exam-
ined how families manage their finances when
they face economic hardship and what factors
are related to the choice of financial management
methods. As explained in the theoretical back-
ground section, the use of savings or the use
of credit might be major methods of managing
unexpected financial difficulty. Therefore, this
section reviews research on emergency funds
and open-ended credit as financing methods.
The focus of previous studies on emergency
funds was on measuring adequacy of emergency
funds and identifying families who were
most likely to be financially unprepared for
emergencies (e.g., Ding & DeVaney, 2000;
Huston & Chang, 1997; Johnson & Widdows,
1985). No research was located that investigated
emergency funds as a financing method when a
household encounters financial difficulty.
The previous literature on emergency funds
has questioned why so many families do not
hold adequate amounts of emergency funds and
whether such behavior is rational (Chang et al.,
1997; Hatcher, 2000). More recently, studies
have suggested that the concept of emergency
funds should be reevaluated to consider other
available resources to meet unexpected events
when families encounter economic hardship (Bi
& Montalto, 2004; Huston & Chang, 1997;
Worthington, 2004). Bi and Montalto included
several sources of credit as alternative forms
of emergency funds. Using an Australian sam-
ple, Worthington examined potential sources of
emergency funds and factors related to those
sources.
As Eisenhauer (1994) suggested, using credit
is one alternative available to families that need
to finance risk. The use of credit becomes
a critical resource when unexpected financial
difficulties occur if a family has not accumu-
lated adequate savings to prepare for this risk.
Although using credit to finance such risk is
becoming more popular (Eisenhauer), research
on this topic is limited. A few theoretical dis-
cussions have shown that the use of credit could
play a significant role in smoothing a transitory
income shock (Brito & Hartley, 1995; Laibson,
Repetto, & Tobacman, 2000). A few empirical
studies have examined the use of credit as a
safety net against income loss (Bird, Hagstrom,
& Wild, 1997; Castellani & DeVaney, 2001).
Little empirical research, however, has exam-
ined the extent of families’ use of credit to
manage their economic hardship.
Worthington’s (2004) study identified the use
of savings and of loans from banks, credit cards,
or family or friends as potential alternative
sources of emergency funds. This study was
How Families Manage Economic Hardship 361
the first examination of how families actually
manage financial difficulties using different risk
management methods. Worthington, however,
focused on emergency funds and the family’s
ability to raise emergency funds, rather than
families’ financial risk management behavior.
These results suggested that ethnicity of the
head of household, the number of dependents,
income, home ownership, and wealth were
significantly related to willingness to use
savings. As disposable income increased, the
willingness to use savings increased. In addition,
renters were less likely to use savings than were
homeowners. Age, number of dependents, home
ownership, income, and wealth were significant
in predicting the willingness to use credit card
loans. Families who rented their dwelling and
those older than 65 were less likely to use credit
card loans. As wealth increased, the willingness
to use credit card loans increased. In contrast,
the level of income was negatively related to the
willingness to use credit card loans.
Worthington’s exploratory study was notable
because it examined the effects of several demo-
graphic factors on different alternatives of emer-
gency funds, but it did not examine the choice
among different alternatives and did not inves-
tigate the effects of attitudes and expectation
factors on these alternatives.
METHOD
Data and Sample
This study used data from the 2007 Survey of
Consumer Finances (SCF). The survey, which
has been collected every 3 years since 1982,
employs a dual frame sampling design to pro-
vide adequate coverage of items that are not
widely distributed in the population, such as
ownership of corporate stocks and tax-exempt
bonds, as well as broadly distributed items in
the population, such as home ownership. Part
of the survey data is a multistage area proba-
bility random sample that ensures the data are
nationally representative, and another part is an
oversample of high-income families from Inter-
nal Revenue Service tax files (Fries & Johnson,
2000). Weights are used to combine information
from the two samples (Bucks, Kennickell, Mach,
& Moore, 2009). The SCF provides detailed
information on households’ assets and liabilities.
Because the sensitivity of financial informa-
tion results in some missing data, the multiple
imputation technique is used, resulting in five
implicates in the SCF. (See Montalto & Sung,
1996, for more detail on multiple imputation.)
This study used all five implicates of the 2007
SCF and applied the Repeated Imputation Infer-
ence (RII) technique to use information from all
five implicates. Weight variables were used to
accommodate the dual frame sampling design of
these data.
The SCF data were selected for this study
because they contain information on whether
families experienced any financial difficulty
such as income shortfall during the past year,
and if they did, information on how families
managed their financial situation. The survey
also included detailed information on assets and
debts, information on attitudes and expectations,
as well as unemployment experiences during the
past year.
The sample for this study was limited to
the economically active population; thus, the
families whose heads were retirees, students,
homemakers, disabled persons, or unpaid volun-
teers were excluded from the sample. To analyze
how families manage their financial difficulty,
this study defined a family with financial diffi-
culty as a family that did not have enough income
to cover all the expenses. The SCF did not pro-
vide specific information on which method a
family would use if they encountered a loss of
income, but the SCF did provide information on
which method was used if a family’s spending
exceeded its income. Therefore, to analyze how
families manage their financial difficulty, the
final sample for this study included only those
families whose spending exceeded their income,
which represents 11.75% of the 2007 SCF.
Model and Measures
Model. On the basis of the theoretical back-
ground and the previous research, the empirical
model was drawn as:
Prob(Yi/Y3)=f[Sf,(UE,CI,FNf,R
f),
Pf(AEf)],
where UE represents families’ unemployment
experiences, CI represents the current status of
the income stream, FNfis families’ financial
needs, Rfis human and financial resources in the
families representing families’ situations, and
AEfrepresents attitudes and expectations repre-
senting preferences. That is, this model suggests
that the probability of choosing to use savings or
362 Family Relations
miscellaneous methods (Yi)over credit (Y3)to
manage economic hardship is a function of fami-
lies’ situations (Sf) and preferences (Pf), specifi-
cally, past unemployment experiences (UE), the
current status of the income stream (CI), finan-
cial need (FNh) and resources (Rh), and attitudes
and expectations (AEh).
Dependent variables. The major research ques-
tion addresses how families manage their
economic hardship. The SCF contains a ques-
tion that asks which methods families used
when their spending exceeded their income.
For this question, a respondent could choose
one of the nine answers: (1) borrowed money
(including using credit cards), (2) spent out
of savings/investment, (3) got behind on pay-
ments, (4) got help from others, (5) cut back on
expenses, (6) got additional income, (7) did noth-
ing, (8) declared bankruptcy, and (9) sold assets.
For the multivariate analysis, the nine categories
were collapsed into three, on the basis of the
frequency of the answers: (1) used credit (if they
answered # 1), (2) used savings (if they answered
# 2), and (3) used miscellaneous methods (if they
answered#3to#9).Themiscellaneousmethods
included all other categories except borrowing
money and spending out of savings. Table 1
presents the measurement of the dependent vari-
ables, and Table 2 presents the frequency of the
answers for each category.
Independent variables. In order to examine the
effects of unemployment, two variables were
included: a variable measuring whether a house-
hold head was unemployed during the past
year and the length of unemployment. Current
employment status and a dual earner variable
were included to represent the current status
of the income stream. The variables repre-
senting resources included financial resources
(e.g., income, net worth, and home ownership),
including credit (e.g., credit limit and credit
constraints) and human resources such as educa-
tion and health status of the head of household.
Among the variables, income, net worth, and
credit limit were transformed to their natural
logarithms. The health constraints variable was
measured by a subjective question that asked
about the health status of the head of the house-
hold; if the respondent answered that his/her
health status was poor, the answer was coded 1
(having a health constraint) and 0 otherwise (no
health constraint).
Age of the head of household, marital status,
and the number of children were included to
represent the families’ lifecycle financial needs.
Attitudes and expectations factors included
attitude toward credit, expectations about future
interest rates, and expectations for future
income. In addition, a variable for the race of
the head of household was included, coded as
1 if the head of household was White and 0
otherwise. (Thus, it did not represent ethnicity.)
Table 1 presents the detailed measurements of
the variables.
Analysis
Descriptive statistics such as frequency and
univariate analyses were conducted to profile the
sample. The data were weighted using weight
variables, and the RII technique (Rubin, 1987)
was used to provide an average value of all
five implicates (Montalto & Sung, 1996). Next,
multinomial logit analysis was conducted to
investigate factors related to the use of different
methods of managing economic hardship.
In this study, multinomial logit analysis is
appropriate because it allows the researcher
to examine which methods are selected over
other methods when the dependent variable
has more than two categories (Borooah, 2001).
The dependent variable is a categorical variable
with three categories—using credit, savings, or
miscellaneous methods. The dependent variable
of multinomial logit analysis is the log-odds
ratio. For this study, two log-odds ratios
were estimated. Where P1=the probability
of using savings, P2=the probability of
using miscellaneous methods, and P3=the
probability of using credit, ln(P1/P3) estimates
the probability of using saving over using credit
and ln(P2/P3) estimates the probability of using
miscellaneous methods over credit.
RESULTS
Descriptive Statistics
Methods of managing economic hardship.
According to the 2007 SCF, most of the families
used credit or their own savings to finance their
income shortfall, about half of the families used
credit (50.46%), and about a third (34.12%)
used their own savings or investments. Some
(6.08%) found additional sources of income and
some (3.89%) got help from others. In order to
How Families Manage Economic Hardship 363
Table 1. Measurement of Variables
Measurement
Dependent variable
Management methods 1. Using savings; if respondents’ choice is (2)
2. Using miscellaneous methods; if respondents’ choices are (3) to (9)
3. Using credit; if respondents’ choice is (1)
Independent variables
Experience of job loss ‘‘At any time during the past twelve months, were you unemployed and
looking for work?’’ 1 if yes; 0 otherwise
Unemployment period Number of weeks of unemployed; continuous
Employment status 1 if employed; 0 otherwise
Dual earner 1 if both the household head and the spouse were working for pay, 0 otherwise
Health constraints 1 if subjective health status is poor; 0 otherwise
Income Before tax annual income in 2006; continuous; natural log of income
Net worth Total assets −total liability, continuous; natural log of net worth
Home ownership 1 if owner; 0 otherwise
Enough emergency fund 1 if liquid asset to income ratio =>3; 0 otherwise
Credit limit The maximum amount that households could borrow using credit cards;
continuous; natural log of credit limit
Credit constraints 1 if turn down for a loan application; 0 otherwise
Age The respondent’s age, continuous
Marital status 1 if married; 0 otherwise
1 if divorced, separated, and widowed; 0 otherwise
1 if single; 0 otherwise
Number of children Number of children in the households; continuous
Education 1 if the years of education <12; 0 otherwise (less than high school)
1 if the years of education =12; 0 otherwise (high school)
1 if the years of education =13 – 16; 0 otherwise (college degree)
1 if the years of education >16; 0 otherwise (more than college)
Race 1 if White; 0 otherwise
Attitude toward credit Whether the respondent felt it was all right to borrow money to cover living
expenses when income was cut
1 if yes; 0 otherwise
Expectation for future interest rate 1 if high; 0 otherwise
1 if same; 0 otherwise
1 if low; 0 otherwise
Expectation for future income 1 if up more than prices; 0 otherwise
1 if about the same as prices; 0 otherwise
1 if be less than prices; 0 otherwise
manage shortage of income, some of the fami-
lies used passive methods such as cutting back
on expenses (1.78%) or falling behind on pay-
ments (1.92%). No one in the sample declared
bankruptcy because of shortage of income. The
results confirm that using credit or depleting
their own assets were the major methods families
used to manage when they encounter economic
hardship.
Characteristics of the sample. The final sample
for this study consisted of 388 families whose
spending exceeded income over the past year. To
provide a clear profile of the sample, the results
of the descriptive statistics for the sample group
(n=388) and for the families that had enough
income to cover their expenses (n=2,911) are
presented in Table 3.
As Table 3 shows, the characteristics of the
sample group are markedly different from those
of the other group. About 22% of the sample
group had an unemployment experience during
the past year, with an average period of unem-
ployment of 19 weeks. Only 11% of the other
364 Family Relations
Table 2. Methods of Managing Economic Hardship
Methods %
Borrowed money including using credit cards 50.46
Spent out of savings/investment 34.12
Got behind on payments 1.92
Got help from others 3.89
Cut back on expenses 1.78
Got additional income 6.08
Did nothing 1.75
Total 100
group had been unemployed, with an average
unemployed period of 15 weeks. The average
income of the sample group was 70% of the
other group’s income, and the average net worth
of the sample group was 54% of the other group’s
net worth. About 20% of the other group had suf-
ficient emergency funds, whereas only 11% of
the sample group met the emergency funds level
guidelines (which was satisfactory if the fam-
ily’s liquid assets were greater than 3 months of
income). Thirty-one percent of the sample group
had been turned down for credit, whereas only
18% of the other group had been turned down
for credit. The sample group had a more favor-
able attitude toward using credit than the other
group. Two-thirds (67%) of the sample group
responded that it was all right to borrow money
to cover living expenses when income was cut
compared with 54% of the other group. More
respondents in the sample group replied that their
expected future interest rates would be higher
and their income would be less in the future.
The Results of Multinomial Logit Analysis
Using the final sample (n=388), multinomial
logit analysis was conducted to examine fac-
tors related to choosing the use of savings or
miscellaneous methods over the use of credit.
For this analysis, all five implicates of the SCF
were used. As Rubin (1987) suggested, the RII
technique provides the best point estimates and
estimates of variance for parameters for a mul-
tiply imputed data set and can be applied to
nonlinear models. Owing to the limitations of
the statistical analysis program, five implicates
were analyzed separately for multinomial logit
analysis. STATA is good for more advanced
statistical analysis such as multinomial logit
analysis, but the earlier version of the program
Table 3. Descriptive Statistics: Profile the Sample
Characteristics
Variables
Sample Group,
n=388 (%)
Other Group,
n=2,911 (%)
Experience of job loss 22.45 11.27
Current employment
status
91.61 95.02
Dual earner 36.40 43.29
Health constraint 28.78 31.37
Home ownership 63.89 66.52
Enough emergency
fund
11.30 19.86
Credit constraint 30.71 17.87
Marital status
Married 50.41 55.93
Divorced/
separated/
widowed
29.01 22.19
Single 20.58 21.88
Education
Less than high
school (<12)
11.76 10.92
High school
(=12)
33.68 28.42
College (=16) 25.35 24.91
More than college
(>16)
29.20 35.75
Race 67.28 71.86
Attitude toward credit 66.60 53.57
Expectation for future
interest rate
Higher 73.61 65.96
Same 19.56 26.87
Lower 6.83 7.17
Expectation for future
income
Up 22.93 25.48
Stay 38.71 43.91
Less 38.36 30.60
M(SD)M(SD)
Unemployment spella19.40 (15.35) 15.08 (14.79)
Age 42.45 (11.58) 43.15 (12.55)
Number of children 0.95 (1.19) 0.83 (1.15)
Income 58,753 (58,688) 83,043 (82,429)
Net worth 266,098
(630,642)
492,388
(1,235,214)
Credit limit 20,864 (31,286) 21,058 (29,368)
Note: All values are weighted.
aNumber of unemployed weeks for those who were
unemployed during the past year.
How Families Manage Economic Hardship 365
does not provide the covariance matrices to use
RII technique for multinomial logit analysis.
The results of all five implicates showed
nearly consistent results, indicating that factors
such as unemployment spell, education, income,
net worth, emergency funds, credit limit, num-
ber of children, race, and expectation for future
income were important. Table 4 presents the
results of the first implicate of the SCF.
The log likelihood of the multinomial logit
analysis showed an adequate fit of the model,
and the pseudo-R2was .121. The results
indicated that income, net worth, emergency
funds, and credit limit were significantly related
to choosing to use savings over credit when fami-
lies encounter economic hardship (e.g., expenses
were greater than income). That is, families who
had a higher level of income and net worth were
more likely to use savings over credit when
they faced economic hardship. As the level of
income increased, the likelihood of using sav-
ing increased about 1.6 times. This result is
consistent with Worthington (2004)’s study. In
particular, if a family had enough emergency
funds, the family was almost four times more
likely to use its savings over credit. As the
credit limit increased, the likelihood of using
saving decreased by 7%. Although the variables
Table 4. Results of Multinomial Logit Analysis
Use Savings ln(P1/P3) Use Miscellaneous Methods ln(P2/P3)
Variables BSE Odds Ratio BSE Odds Ratio
Experience of job loss 0.098 0.450 1.103 −0.320 0.645 0.726
Unemployment spell 0.033 0.019 1.034+0.047 0.024 1.048∗
Employment status 0.180 0.568 1.197 0.459 0.758 1.582
Dual earner −0.081 0.308 0.922 −0.608 0.456 0.544
Health constraints 0.302 0.277 1.353 0.105 0.439 1.111
Education (less than high school)
High school (=12) 0.423 0.471 1.527 −0.117 0.496 0.890
College (=16) 0.562 0.492 1.754 −0.235 0.557 0.791
More than college (>16) −0.151 0.511 0.860+−1.382 0.700 0.251∗
Income 0.467 0.183 1.595∗0.218 0.289 1.244
Net worth 0.122 0.050 1.130∗0.076 0.061 1.079
Home ownership 0.159 0.349 1.172 0.577 0.519 1.781
Enough emergency fund 1.338 0.372 3.811∗∗∗ 0.088 0.744 1.092
Credit limit −0.076 0.038 0.927∗−0.195 0.049 0.823∗∗∗
Credit constraints −0.336 0.297 0.715 −0.170 0.398 0.844
Age 0.014 0.014 1.014 −0.039 0.020 0.962+
Marital status (single)
Married −0.406 0.476 0.666 0.704 0.614 2.022
Divorced/separated/widowed −0.593 0.437 0.553 −0.397 0.579 0.672
Number of children 0.046 0.112 1.047 −0.506 0.199 0.603∗
Race −0.582 0.318 0.559 −0.804 0.404 0.448∗
Attitude toward credit −0.358 0.264 0.699 0.467 0.423 1.595
Expectation for future interest
rate (same)
Higher −0.019 0.303 0.981 0.721 0.531 2.056
Lower −0.474 0.503 0.623 0.512 0.823 1.669
Expectation for future income
(stay)
Up 0.228 0.346 1.256 0.385 0.510 1.470
Less 0.335 0.285 1.398 0.951 0.420 2.588∗
Constant −6.671 1.999 — −2.524 3.040 —
Log-likelihood ratio 630.43
aReference groups are in parentheses.
+p<.10. ∗p<.05. ∗∗ p<.01. ∗∗∗p<.001.
366 Family Relations
for the period of past unemployment were not
significant at the .05 level, they were significant
at the .10 level. If the unemployment period
increased, the family was more likely to use
savings rather than credit to manage economic
hardship.
Past unemployment spell and expectation
for future income were positively related to
using miscellaneous methods over using credit,
whereas education, credit limit, number of chil-
dren, and race were negatively related to using
miscellaneous methods when families faced eco-
nomic hardship. Specifically, for each week
the unemployment period increased, the likeli-
hood of using miscellaneous methods over credit
increased by 5%. If a head of the family expected
that the family’s future income would be less,
the family was about 2.6 times more likely to
choose miscellaneous methods over credit. If a
head of the family had a college degree or a
more advanced degree or had more children, the
probability of choosing miscellaneous methods,
however, decreased by 75 and 40%, respectively.
As credit limit increased, the probability of
choosing miscellaneous methods decreased by
18%. If a head of the family was White, the fam-
ily was less likely to choose miscellaneous meth-
ods over credit to manage its economic hardship.
DISCUSSION
The recent economic downturn has brought
increased attention to families’ financial situa-
tions and how they deal with economic hardship.
Minimal research, however, has been conducted
on how families manage their economic hard-
ship. Thus, this study examined families’ man-
agement behavior when they face economic
hardship using a sample of families that did
not have enough income to cover their expenses.
The results indicated that families use differ-
ent types of management behavior when they
encounter economic hardship. As suggested by
theory, the major financing methods to make
up the difference between income and spending
are the use of savings or investments (34.12%)
and the use of credit (50.46%). Because of
the opportunity cost of maintaining an ade-
quate emergency fund, using credit could be
considered a more efficient management method
to smooth temporary financial difficulties. It
should be noted that using credit could be a
potentially risky behavior for some families,
depending on their financial situations.
According to the results, the sample group
that did not have enough income to cover their
expenses was a more economically vulnerable
group than the families that had enough income
to cover expenses. The average income of the
sample group was 70% of the income of the
other group. The average net worth of the sam-
ple group was only 54% of that of the other
group. Only 11% of the sample group had
enough emergency funds compared with 20%
of the other group. In addition, more families in
the sample group experienced unemployment.
The attitude toward using credit for the sample
group, however, was more favorable than that
of the other group.
The results of multinomial logit analysis
showed that, depending on families’ situations
and their preference, use of management meth-
ods is different when they face economic hard-
ship. The variables representing current financial
resources, such as income, net worth, and having
sufficient emergency funds significantly affected
the choice of using savings as a financing method
to manage economic hardship. In particular, hav-
ing sufficient emergency funds was the most
influential factor to determine the use of savings
over credit. If families had enough emergency
funds, they were 3.8 times more likely to use
their own savings or investment than if they did
not have adequate emergency funds. This finding
suggests that, although there may be some oppor-
tunity costs to hold emergency funds, doing so is
essential and it should be an important strategy
to manage families’ financial difficulties. The
amount of emergency funds required could be
adjusted depending on the changes in interest
rates and the economy in general.
The findings related to the unemployment
period and expectations for future income sug-
gest that families behave rationally, correspond-
ing to their risk situations. As the unemployment
period increases, people tend to see their finan-
cial difficulties as a long term, rather than a
temporary situation. Thus, they are more likely
to use their own savings or other miscellaneous
methods such as finding additional sources of
income or getting help from others rather than
increase the use of credit. Similarly, if families
expect their future income will be lower, they
try not to incur more debt by using credit.
As suggested in Mooney (2008), longer
term strategies for managing economic hardship
could include any of the following actions:
increasing income by taking on a second or
How Families Manage Economic Hardship 367
third job, changing occupations, moving in with
family members, or reducing debt. Each of these
actions requires sacrifice to achieve and could
be difficult for families. For example, if either
spouse or a single parent worked a second or
third job, it would mean having less time for them
to take care of children, do household chores,
and is likely to curtail any leisure activities.
Changing occupations would probably involve
accumulating additional debt and spending time
attending class and studying. Taking online
courses would help with travel costs and would
reduce time away from home. Moving in with
family implies losing privacy and adjusting
to the host family’s routine, which could
include sharing in household work and expenses.
Reducing debt could mean forgoing medical
care, insurance, donations to church and charity,
gifts, and entertainment. It is worth noting that
these cost-saving activities could be practiced
by families who are in the sample group or in
the other group. An outcome that has not been
mentioned is the stress that families are likely to
experience in the event of economic hardship.
Stress could be increased if families need to take
on any of the activities mentioned in this section.
The results show that the probability of using
savings increases as families’ financial resources
such as income, net worth, or emergency funds
increase, regardless of their preferences or finan-
cial needs. Meanwhile, if a head of the household
has more education and is White, the family is
less likely to use miscellaneous methods over
using credit compared with the families whose
head is non-White and less educated. In fact, the
impacts of education and race are greater than
the impacts of credit limit. The results suggest
that for families with more financial resources, it
appears that using credit as a financing method
when they encounter economic hardship would
not be the first choice. Using credit would, how-
ever, be the first available financing method for
those with less financial resources. In addition,
although for some families, using credit would
be a more efficient method to deal with a short-
term economic hardship, some families might
lack knowledge or experience in regard to the
use of credit.
Therefore, the most appropriate methods and
available resources to manage economic hard-
ship should be different depending on the fam-
ilies’ situations. Thus, financial advisors need
to evaluate if the financial risk that a family
faces would be a long-term financial risk or a
short-term one, and then to evaluate the fam-
ily’s financial situation such as the level of
emergency funds and assets, any debts, and any
other available financial resources. If the finan-
cial difficulties are expected to last for a longer
period, a family needs to find a long-term strat-
egy rather than simply using their credit. If using
credit is the only alternative, financial advisors
or educators could help families compare avail-
able credit sources and interests rates and should
inform them how to use it wisely. In addition,
for those who want to use credit for managing
their short-term economic difficulties, the credit
industry could develop more options attached to
using credit or credit instruments that are related
to deal with such difficulty.
A direction for future research would be to
interview families to learn if they practiced
additional methods of coping with economic
hardship in addition to using savings, credit,
or the miscellaneous methods mentioned in the
SCF. Another possibility would be to include
questions about some of these activities in future
versions of the SCF. If quantitative data became
available, future research could develop a more
rigorous model to examine which methods of
managing economic hardship would be more
efficient for families under different financial
conditions and risks.
A limitation of the study is related to the
sample because the SCF does not provide infor-
mation on how families manage their economic
hardship for all families in the SCF, but only
on which methods are used when a family’s
spending exceeds its income. Therefore, this
study included only families whose spending
was greater than their income. As a result, the
sample size was relatively small in size for the
final analysis, and the characteristics of the sam-
ple may be different from those of typical U.S.
families. Thus, the results should be interpreted
with caution considering the sample character-
istics. In addition, this study analyzed the 2007
SCF; although this is the most recent available
data, it was collected prior to the recent reces-
sion. Therefore, if researchers use data collected
after 2007, the results might be different.
REFERENCES
American Express. (2001). Unemployment rate
increase causes concern about financial security.
PRNewswire. Retrieved May 30, 2009, from http:
//www.hirediversity.com/news/2001/5/7/unem
ployment rate increase causes concern about.htm
368 Family Relations
Bi, L., & Montalto, C. P. (2004). Emergency funds
and alternative forms of saving. Financial Services
Review, 13, 93 – 109.
Bird, E. J., Hagstrom, P. A., & Wild, R. (1997).
Credit cards and the poor (Discussion Paper
No. 1148-97). Institute for Research on Poverty.
Retrieved December 18, 2008, from http://www.
ssc.wisc.edu/irp/pubs/dp114897.pdf
Borooah, V. K. (2001). Logit and probit: Ordered and
multinomial models. Thousand Oaks, CA: Sage.
Brito, D. L., & Hartley, P. R. (1995). Consumer
rationality and credit cards. The Journal of
Political Economy, 103, 400 – 433.
Bryant, K. W. (1990). The economic organization of
the household. Cambridge: Cambridge University
Press.
Bucks, B. K., Kennickell, A. B., Mach, T. L., &
Moore, K. B. (2009). Changes in U.S. family
finances from 2004 to 2007: Evidence from the
Survey of Consumer Finances. Federal Reserve
Bulletin, 95, A1 – A55.
Bureau of Labor Statistics. (2009). Labor force statis-
tics from the current population survey. Retrieved
July 14, 2009, from http://data.bls.gov/PDQ/
servlet/SurveyOutputServlet?data tool=latest
numbers& series id=LNS14000000
Carroll, C. D., Dynan, K. E., & Krane, S. D. (1999).
Unemployment risk and precautionary wealth:
Evidence from families’ balance sheets. (Working
paper). Retrieved January 25, 2002, from http:
//www.eco.jhu.edu/papers/Carroll/unempl.pdf
Castellani, G., & DeVaney, S. A. (2001).Using credit
to cover living expenses: A profile of a potentially
risky behavior. Family Economics and Nutrition
Review, 13(2), 12 – 20.
Chang, R. Y., Hanna, S., & Fan, J. X. (1997). Emer-
gency fund levels: Is household behavior rational?
Financial Counseling and Planning, 4, 47 – 55.
Conger, R. D., Conger, K. J., Elder, G. H., Jr.,
Lorenz, F. O., Simons, R. L., & Whitbeck, L. B.
(1992). A family process model of economic hard-
ship and adjustment of early adolescent boys. Child
Development, 63, 526 – 541.
Conger, R. D., Elder, G. H., Jr., Lorenz, F. O., Con-
ger, K. J., Simons, R. L., Whitbeck, L. B., Huck,
S., & Melby, J. N. (1990). Journal of Marriage
and Family, 52, 643 – 656.
Dew, J. (2008). Debt change and marital satisfaction
change in recently married couples. Family
Relations, 57, 60 – 71.
Ding, J., & DeVaney, S. A. (2000). Predictors of
emergency funds adequacy. In J. E. Morris &
C. R. Hayhoe (Eds.), Proceedings of Association
for Financial Counseling and Planning Education
(pp. 56 – 63). Alexandria, VA: Association for
Financial Counseling and Planning Education.
Eisenhauer, J. G. (1994). Household use of open-
ended credit to finance risk. The Journal of
Consumer Affairs, 28(1), 154 – 169.
Freidman, M. (1957). A theory of the consump-
tion function. Princeton, NJ: Princeton University
Press.
Fries, G., & Johnson, B. (2000). Disclosure review
and the 1998 Survey of Consumer Finances.
Retrieved February 10, 2003, from http://www/
federalrewerve.gov/pubs/oss/oss2/papers/finalasa
2000.pdf
Gill, I. S., & Ilahi, N. (2000). Economic inse-
curity, individual behavior and social pol-
icy (Working paper). Washington, DC: The
World Bank. Retrieved February 4, 2004,
from http://wbln0018.worldbank.org/LAC/LAC
InfoClient.nsf/5996dfbf9847f67d85256736005dc
67c/3ee47810770452658525694b003cd4ea/
$FILE/Gill Ilahi.pdf
Hatcher, C. B. (2000). Should families establish
emergency funds? Financial Counseling and Plan-
ning, 11(2), 77 – 83.
Huston, S. J., & Chang, R.Y. (1997). Adequate emer-
gency fund holdings and household type Financial
Counseling and Planning, 8(1), 37 – 46.
Johnson, D. P., & Widdows, R. (1985). Emergency
fund levels of families. In K. P. Schnittgrund (Ed.),
Proceedings of American Council on Consumer
Interests (pp. 235 – 241). Evansville, IN: Ameri-
can Council on Consumer Interests.
Johnson, D. R., & Booth, A. (1990). Rural economic
decline and marital quality: A panel study of farm
marriages. Family Relations, 39, 159 – 165.
Laibson, D., Repetto, A., & Tobacman, J. (2000).
A debt puzzle. NBER Working paper (No. 7879).
Retrieved March 29, 2009, from http://papers.nber.
org/paper/w7879
Montalto, C. P., & Sung, J. (1996). Multiple impu-
tation in the 1992 Survey of Consumer
Finances. Financial Counseling and Planning, 7,
133 – 141.
Mooney, N. (2008). (Not) keeping up with our par-
ents: The decline of the professional middle class.
Boston: Beacon Press.
Moseley, F. (2009). The U.S. economic crisis: Causes
and solutions, International Socialist Review, 64,
Online Edition. Retrived June 20, 2009, from http:
//www.isreview.org/issues/64/fred-moseley.
html.
Rubin, D. B. (1987). Multiple imputation for nonre-
sponse in surveys. New York: John Wiley & Sons.
Sullivan, T. A., Warren, E., & Westbrook, J. L.
(2000). The fragile middle class: Americans in
debt. New Haven, CT: Yale University Press.
Vaughan, E. J. (1997). Risk management.NewYork:
John Wiley & Sons.
Welsh, J. B. (2001). Finding gold in pink slips. Finan-
cial Planning, 13(4), 89 – 91.
Worthington, A. C. (2004). Emergency funds in Aus-
tralian families: An empirical analysis of capacity
and sources. Financial Counseling and Planning,
15(1), 21 – 30.
