Economic predictors of the subjective experience of financial stress
Thoughts
Connects with: @marjanovic2013
Annotations
simonse2024 - p. 1
We define financial stress as a psychological construct that reflects a state where pressing financial concerns surpass available resources, endangering well-being (Van Dijk et al., 2022). Financial stress includes subjective appraisals of the situation and affective and cognitive responses.
simonse2024 - p. 1
We incorporate two appraisals: insufficient financial resources and lack of control over one’s financial situation. The first appraisal captures the (potential) harmfulness of the situation, whereas the second refers to coping potential - the perceived ability to deal with the (potentially) harmful situation adequately. We also include affective and cognitive responses, namely financial worries and rumination, and short-term focus.
simonse2024 - p. 2
Financial stress differs from financial well-being, defined by Brüggen et al. (2017) as “the perception of being able to sustain current and anticipated desired living standards and financial freedom.” Financial stress focuses on people’s current financial situation and is the inability to meet financial demands. In contrast, financial well-being includes the current and anticipated financial situation and is understood as the ability to meet financial demands. Netemeyer et al. (2017) define financial well-being as current money stress and future financial security. The first aspect - current money stress - involves being behind with one’s finances, feeling that one’s finances control one’s life, and being obsessed with money. The second aspect of Netemeyer’s definition - future financial security - resembles Hoffmann et al., (2022) definition of financial well-being as expected financial security.
simonse2024 - p. 2
Finally, our conceptualization of financial stress encompasses financial worry, defined as “repeated and negative thinking about the uncertainty of one’s (future) financial situation,” and financial rumination, defined as “repetitive, passive, and pessimistic thinking about the possible causes and consequences of one’s financial concerns”
simonse2024 - p. 2
This definition resembles Xiao and Kim’s (2022, p. 139) definition of financial stress as a “psychological state worrying about personal finance.”
simonse2024 - p. 2
It is similar to financial anxiety (Kim et al., 2023), defined as worrying and anxiety about current and future financial situations.
simonse2024 - p. 2
The literature about the relationship between income and financial stress is ambiguous, suggesting that other economic factors may also play a role. Research in mental health psychology and other fields, for example, indicates that mental well-being and stress are not only associated with income but also with economic factors such as savings, debts, income volatility, and employment.
simonse2024 - p. 2
he current research examines the relative importance of five aspects of one’s economic situation - income, savings, debts, income volatility, and employment status - in predicting financial stress. Also, it examines whether the associations differ between lower- and higher-income households. Finally, we statistically control for well-established confounders, such as age, education level, gender, and personality traits.
simonse2024 - p. 4
There is no coherent picture of how different elements - in conjunction - correlate with financial stress. The current research, therefore, takes a more integrative perspective on households’ economic situations by including five aspects: income, savings, debts, income volatility, and employment
simonse2024 - p. 5
We used the 12-item Psychological Inventory of Financial Scarcity (PIFS) developed by Van Dijk et al. (2022) to measure financial stress (M = 1.96, SD = 1.12, Cronbach’s α =.93). Their psychometric evaluation shows that the PIFS is a reliable and valid measure. It combines scarcity theory with frameworks of financial stress.
simonse2024 - p. 6
come. Centerdata measures net monthly household income in euros. We corrected for household size because the needs of a household grow with each additional member. To consider economies of scale, we adjusted household income by dividing it by the square root of household size, according to OECD (2013) guidelines.
simonse2024 - p. 6
Savings may serve as buffers against unexpected expenditures and income shocks. Ruberton et al. (2016) stressed the importance of a minimal buffer in the form of liquid wealth for well-being. We defined buffer as a dichotomous variable equaling one if a household’s liquid assets exceeded a threshold depending on income and household size and zero otherwise. We argue that higher-income families need a higher buffer because they have more fixed expenditures and own more property. Based on the Buffer Calculator provided by Nibud (n.d.) (National Institute for Family Finance Information), we used the following formula to define the threshold for having sufficient buffer: € 600 + [monthly income] + € 400 * [household size].
simonse2024 - p. 6
Debts. Given that the number of debts is more predictive of financial stress than the total debt amount (Ariely et al., 2009; Ong et al., 2019), we included the number of debts as an independent variable in our analysis. We also argue that, for most households, having a mortgage contributes less to financial stress than other types of debt since the home’s value usually amply compensates the mortgage loan’s value.
simonse2024 - p. 6
Income volatility. Two possible indices of income volatility are the relative size and the number of adverse income shocks in a given period. Prause et al. (2009) found that the latter was a better predictor of psychological depression than the former; an income loss results in the need to cut expenditures and may cause difficulty paying the bills.
simonse2024 - p. 6
Employment. Centerdata asks respondents to select their primary occupation from 14 options.
simonse2024 - p. 7
Our model included nine control variables: gender, age, education level, household size, and five personality traits. Our theoretical framework suggests that household size, gender, and personality traits may be associated with financial stress. Income tends to have an inverseU relationship with age and rise with education level.
simonse2024 - p. 7
We included Goldberg’s (1992) Big Five personality traits: openness to experience, conscientiousness, agreeableness, extraversion, and emotional stability (α =.78,.78,.81,.88, and.89, respectively).
simonse2024 - p. 7
We found no multicollinearity between the independent variables in our model (see supplemental materials, Tables S3 and S4).
simonse2024 - p. 7
Deleting observations with missing values on one or more variables would leave 49% of the data unused, resulting in inflated standard errors (Van Buuren, 2018). The preferred methods for dealing with missing data fall into two broad groups: maximum likelihood estimation and multiple imputation (Allison, 2002). A test run with 20 imputations resulted in a maximum lambda of.64. Based on Von Hippel’s (2020) guidance, we set the number of imputations at 93, corresponding with lambda =.05. We, therefore, created 93 imputed data sets, each representing a plausible completion of the missing values. These 93 imputed data sets gave us 93 different versions of the complete data, accounting for uncertainty in the missing data
simonse2024 - p. 7
It is well established that ordinary least squares (OLS) estimation can give highly unreliable outcomes in the presence of influential observations. OLS minimizes the sum of the squared residuals, which gives “unusual” observations an unduly large weight. Because our data contained many outliers and heavy tails, we applied the MM-estimator developed by Yohai (1987), which goes through three stages to estimate a regression model.
simonse2024 - p. 7
We performed robust regression for each imputed dataset, resulting in 93 regression analyses. Next, we applied Rubin’s (1987) rules to pool the results of these individual regressions. We averaged the estimates of the 93 individual regressions to obtain the parameter estimates.
simonse2024 - p. 8
The fit for Model 3 (R2 =.36) was significantly higher compared to Model 2 (W = 2.97, p =.019). In this model, the relative contribution of buffer savings and income was comparable. The number of debts had a smaller but significant contribution to predicting financial stress. On average, the results did not show employment contributes to financial stress. However, we did find an interaction between income and employment. Results showed a negative association between employment and financial stress for an income level two standard deviations below the mean; for all other income levels, results did not show an association between employment and financial stress.
simonse2024 - p. 8
The control variables education level, age, gender, and household size were significant covariates, whereas psychological traits were not. In line with previous findings, age and education level had a negative association with financial stress. Other things being equal, males experienced more financial stress than females, contrasting with earlier findings. Household size was negatively associated with financial stress.
simonse2024 - p. 8
Finally, we explored how the five aspects of one’s economic situation predicted each of the four aspects of financial stress (the appraisal of money shortage and lack of control, financial worries and rumination, and short-term focus, Table S13).