In June 2016, a multidisciplinary team of researchers at the Financial Wellness company, Payoff, developed a nationally representative study from which we can quantify and understand the physical, emotional and psychological effects of how Americans experience financial stress. Establishing health and financial baselines, we are examining the link between psychological and medical outcomes and financial stress, and determining how to identify and protect people from a cascade of negative health outcomes. In previous studies, we have seen that financial stress is linked to mental health issues, and this study goes further by assessing personal finances, physical and psychological health.
Systematic study of the impact of the poor financial health of Americans is sparse. Our team released the Payoff Financial Stress Index in September, 2016, the first study of its kind that evaluates the psychological impact of financial stress on a representative sample of 2,286 Americans, as surveyed by Yougov.com.
Our data shows comorbidity with Acute Financial Stress (AFS) and a range of physical and mental health issues. From here, we will continue this study to assess how long-term financial stress affects people across health spectrums as well as how it affects micro and macro financial indices.
We began this study after a two-year course of research conducted by the authors to support product development at Payoff, a science, financial wellness and lending company. Concluding validated psychological study of financial behaviors and personality, it became clear that millions of Americans experience stress related to their finances at clinically relevant levels, leading to study and analysis of what we have termed Acute Financial Stress, which was evident in 23% of Americans over 17 and in 36% of Millennials. We contend that AFS is Financial Post-Traumatic Stress Disorder (PTSD), and long-term stress at this level is known to impact all of the body’s systems, potentially leading to illness.
Research History
Financial Personality and Factor Fear: April, 2014 - April, 2015
The research in support of Payoff began with an iterative project that developed the Financial Personality Quiz: based on the Five-Factor Model (FFM) of personality. Classical psychometric methods supported the FFM as a model to understand financial behaviors, attitudes and values, yet exploratory factor analyses of the item set developed to cover the intended domain consistently resulted in an un-hypothesized factor as well: ‘Factor Fear’, as we called it.
This factor contained items suggesting broad and abstract fear of financial failure, avoidance of an individual’s finances (including not knowing how much money one had), denial of the consequences of this avoidance and isolation from others concerning one’s financial problems. As a maladaptive approach to protecting oneself psychologically, avoidance is particularly destructive to one’s financial well-being and leads to compounding problems. We noted how quickly this approach cascaded into full-blown pathological stress as bills pile up, creditors circle, isolation increases and anxieties and related symptoms escalate.
While certainly not a new area of study, the effects of chronic stress, often as defined by the myriad problems and deficiencies confronted by those who live in poverty has recently led to a very productive theory on how chronic stress leads to medical issues. This theory, termed Allostatic Load by the molecular biologist Bruce McEwen, frames the inability to bring the stress response back to normal, or allostasis, wherein a whole range of psychological and physical dysfunctions develop and/or become exacerbated.
Acute Financial Stress: Financial Post-Traumatic Stress Disorder - April, 2016
‘Factor Fear’ was later hypothesized to be consistent with the symptoms seen among those with trauma-induced PTSD. Crucially: the reported symptoms were consistent with those required for a diagnosis of trauma-related PTSD. 23.2% of those surveyed in our National Study met the gold-standard criteria for PTSD, stemming solely from symptoms related to financial stress using the complete 17 item PTSD Checklist-Civilian version (PCL-C). At this time we termed financial PTSD as AFS.
Financial Stress Index (FSI) - September, 2016
Our studies of Financial Personality and AFS led to the development of the Financial Stress Index (FSI), as an effort to both provide a metric to follow the level of financial stress present across America and to track the percentage of people meeting criteria for AFS. This survey also includes a range of demographic and lifestyle information, extensive information of present financial status, several psychometric instruments and an extensive review of medical conditions and medications prescribed.
This study included a six item scale derived from the PCL-C (See Appendix A) and has been validated to provide a continuous index of the financial stress that is evident in the United States. Additionally, it has been re-validated to identify a level of dysfunction that is consistent with the standard diagnosis of PTSD, allowing for an estimate of the ongoing percentage of Americans with the dysfunction labeled Acute Financial Stress. This diagnosis has been altered for the current study to better reflect the scoring system used in the original PCL, which required the presence of significant symptoms in three areas: thoughts, emotions and behaviors. The questions used to assess symptoms in these areas are shown in Appendix A.
The wide range of self-reported items on current financial conditions in the FSI include: total amount of debt and assets (not including real estate), household income, amount of credit card debt, months of living expenses in liquid savings available, percentage of bills paid every month, percentage of credit cards paid monthly over the past year, and critically: an estimate of an individual’s cash-flow, which combines debt and income.
This current abstract focuses on the financial and psychological metrics of the FSI, and the categorization of participants into levels of AFS including No/Minimal financial stress, Mild Financial Stress and Acute Financial Stress.
Results:
There were a total of 2,286 participants with all data needed to estimate propensity weights:
1,314 (57.5%) of the participants reported no financial stress
562 (24.6%) met the criteria for Mild AFS
411 (18.0%) met the Full AFS diagnosis.
National FSI Metric:
Our long term goal is to establish the FSI as an Index that has meaning through its own properties and in terms of how it relates to other financial indices such as the Gross Domestic Product. While it is too early to estimate relationships between the FSI and other financial indices, we do present descriptive statistics. There appears to be a slight increase in the overall level of financial stress as the mean FSI is now 12.1 (sd = 6.6, n = 2,286) as compared to 11.5 (sd = 5.8, n = 2,041) from April, 2015. Again, it is too soon to interpret this difference.
Acute Financial Stress Classification:
Given our use of an abbreviated measure of FSI and a revised, more conservative system to diagnose AFS (See Appendix A), we do not think it is appropriate to compare the percentage of Americans diagnosed presently as compared to a year ago. However, our current classification of AFS where we include a Mild group along with Full AFS and No AFS appears meaningful and will be the classification system as we move forward.
18% (411) of the 2,286 participants with complete demographic data, which allows for the application of a propensity weight to better reflect the entire population of participants in the full YouGov national panel met the criteria for Full AFS in the current sample, while 562 (24.6%) were diagnosed with Mild AFS. Thus, 1314 (57.5%) did not report any signs at all of experiencing financial stress. The significance of the frequencies of these variables across the three levels of AFS were tested with Pearson’s chi-square.
Financial Information:
Annual Family Income:
The overall chi-square was extremely significant for the overall association of Income (<$30k, $30-79,999k, $80-$149,999k, $150k+) and AFS (p < .001).
Only having an income <$30k (26.2%) was associated with an increased frequency of Full AFS.
The lowest frequency (10.8%) was found for those making $80k-$149,999.
Among the Mild AFS group, there were elevated frequencies for those making less than $80k, suggesting Mild AFS may be more sensitive to income than Full AFS.
Total Debt - The chi-square between Total Debt (All debt except mortgage) was highly significant (P < .001).
For Full AFS, the cutoff where the level of debt increased risk was $5,000. All groups above $5,000 in debt showed elevated percentages of AFS.
With Mild AFS, debt elevated risk beginning at $1,000.
Total Assets - The chi-square between Total Assets (All assets excluding real estate) was again highly significant (p < .001).
For Full AFS, the risk was clearly lower for those with Assets of $50k and above.
For those with Mild AFS, the risk did not clearly decline until participants had $100,000 or above in Assets.
FICO - FICO (<620, 620-659, 660-719, 720-759, 760+) was highly associated with AFS category (p < .001)
For Full AFS, the frequency was elevated until scores were at 720 or above. Further, those above 760 had a frequency in the single digits (7.9%).
It should be noted that rates were extremely high for Full AFS when FICO was < 620 (30.4%).
Mild AFS had an even higher frequency among those with FICO <620 (40.9%).
Mild AFS did not diminish significantly until FICO scores were 760 or greater.
Global Financials Summary: All of these Global Financial variables, when considered independently, are highly statistically significant. Interestingly, each variable had a generally clearly identifiable cut-point separating risk groups, with Income only less than $30k clearly elevating risk for AFS. Above $30k, there was a general equivalence of risk below the average, with the lowest risk being between $80-$149,999k.
Similarly for Debt: $5,000 in debts delineated higher than average from lower than average risk of AFS, with some group fluctuations.
Assets above $50k was the general cutoff point where the prevalence of AFS decreased.
FICO above 720 was the general cut-point where lower prevalence was apparent, although when considering only above 760, the frequency of AFS was extremely low.
This suggests that people with Total Debt below $5K, Income above $30k, Assets above $50k and FICO above 720 will have an extremely low frequency of AFS.
Monthly Income/Expenses: This self report question asked subjects to rate whether their monthly expenditures are much higher than their income or slightly higher, as well as income slightly higher than expenses or much higher. Again, there was an extremely significant (p < .001) association with AFS.
With Full AFS, almost 50% (47.8%) of those who reported expenditures were “much higher” than income were diagnosed with AFS. This is the most predictive factor seen among these financial variables.
Those who reported expenditures were “slightly higher” had a 22% frequency of AFS.
When expenditures were about even with income, the frequency rate of AFS declined to 12%.
However, when income is “slightly higher” (5.9%) and even more so when it is “much higher” (1.7%), AFS decreases almost to zero. This finding could have profound ramifications as we further consider how to treat this condition.
With Mild AFS, there was not a decline below expected rates when income and expenditures were even, as there was with Full AFS.
While there were clear declines in Mild AFS when income slightly or greatly exceeded expenditures these in no way approached zero as with Full AFS.
In sum, this simple question seems to be one of the most predictive of AFS that we have yet found.
Percentage of Credit Card and Bills Paid: We asked two separate questions about the percentage of debt typically paid (<50%, 50-75%, 76-99% and 100%) at the end of each month, separate questions for credit card debt and all other bills. Both questions showed highly significant associations with AFS (p < .001).
The pattern was the same for both credit card and other bill payments.
The pattern was also the same for Full and Mild AFS.
Fundamentally, there is no decrease in the rates of Full or Mild AFS unless 100% of all bills and credit cards are paid every month.
This is perhaps the most compelling empirical support yet for the supposition that Debt equals Stress.
Months of Expenses Available: There was an extremely high level of significance in the association between the number of months of expenses the person could cover without tapping into long-term savings vehicles.
With Full AFS, there was a marked decline provided there was 1 month or more of savings available.
With Mild AFS, there was a decline in the rates provided there were 6-9 months of expenses that could be covered from savings.
For both Full and Mild AFS, there was also a dramatic decrease in frequency rates provided there were more than 12 months of expenses available. Full AFS drops to 6.2% and Mild to 6.7%, indicating a powerful protective effect in having more than 12 months of expenses available.
Summary of Self-Reported Cash-flow: The self-reports that people give regarding their Cash-flow and ability to pay their bills is perhaps the most revealing information about AFS from this survey. Simply put: if a person feels they are bringing in more each month than is going out, Full AFS drops to negligible rates. Further, if all bills, including credit cards, are paid each month in full, AFS again drops to single digits. Finally, if the goal is to achieve the elimination of AFS, people should be encouraged to have at least 12 months of living expenses in their savings account. If someone meets all of these conditions, it appears that for all intents and purposes, AFS becomes a non-issue.
Types of Debt
Type | Overall p value | No Stress | Mild Freq | Full AFS |
Overall Freq | | 57.5% | 24.6% | 18.0% |
Mortgage Y (n=682) | < .001 | 64.4% | 21.6% | 14.1% |
Student Loan Y (n=460) | < .001 | 42.4% | 35.4% | 22.2% |
Credit Card Y (n=1002) | Ns | 55.7% | 25.7% | 18.6% |
Payday Y (n=84) | <.001 | 22.6% | 36.9% | 40.5% |
Other Unsecured Y (n=225) | <.001 | 40.9% | 32.4% | 26.7% |
Other Secured Y (n=309) | Ns | 57.3% | 25.2% | 17.5% |
Types of Debt Summary: When considering Types of Debt, it is evident that all debt is not created equal. These Types were simple Yes/No indicating the presence of the types of debt, not the amount.
It seems highly relevant that having a Mortgage reduces the prevalence of AFS. Further, the simple presence of Credit Card Debt, suggests that just having some credit card debt in and of itself doesn’t cause stress, it’s when one can’t pay it off (see above).
Student Loans and Other Unsecured Loans are both associated with increases in AFS. Most remarkable, though, is the extent to which Payday Loans are associated with increased rates of AFS.
The pattern of prevalence for Mild AFS is quite similar to Full AFS.
Multinomial Logistic Regression Models: To determine the relative importance of these financial variables, we employed a multinomial logistic regression model with all of the above financial variables entered as predictor variables and the three-category AFS variable as the dependent variable.
Effect | Likelihood Ratio Tests | |
| Chi-Square | Df | Sig. |
Intercept | 0 | 0 | . |
Monthly Inc/Expense | 232.843 | 10.000 | 0.000 |
% CC Paid | 7.663 | 10.000 | 0.662 |
% Bills Paid | 56.630 | 10.000 | 0.000 |
Months Liv Expense | 34.495 | 14.000 | 0.002 |
Total Assets | 54.461 | 26.000 | 0.001 |
Annual Fam Income | 40.427 | 32.000 | 0.146 |
FICO | 52.458 | 12.000 | 0.000 |
Mortgage Debt | 5.258 | 2.000 | 0.072 |
Student Loan Debt | 10.316 | 2.000 | 0.006 |
CC Debt | 0.255 | 2.000 | 0.880 |
Payday Lender Debt | 0.519 | 2.000 | 0.771 |
Other Unsec Loan Debt | 2.616 | 2.000 | 0.270 |
Other Sec Loan | 0.104 | 2.000 | 0.949 |
The significant predictors of group membership, in order of largest to smallest chi-squares are: 1) Monthly Income to Expenses 2) Percent of bills paid on a monthly basis 3) Total assets 4) FICO score 5) Annual Family Income 6) Months of Living Expense in liquid accounts 7) Student Loan Debt.
The common thread apparent through the bulk of these variables is an adequate income and resource base to assure one’s ability to pay one’s bills for this month and several months in the future. With a clear ability to meet one’s obligations - to stay above water - as it were, AFS is rare. The one type of debt that persists in predicting AFS above all of the asset and income related variables is Student Loan Debt. This suggests it may be a source of stress even when income is adequate to pay off this debt.
Conclusions
The overall FSI scores observed for this representative sample suggest that the majority of Americans (57.5%) are not experiencing significant AFS. Conversely, 18% of Americans report the same cluster of symptoms that are required for a diagnosis of PTSD following a traumatic event. These symptoms include dysfunctional thoughts, particularly: repetitive negative thoughts about the stressful situation, emotions characterized by avoidance of the stressor as well as closeness with others and behaviors demonstrating anger and irritability. While there is an ongoing debate as to whether there is any clinical difference in PTSD if it develops subsequent to a trauma versus a chronic stressor, the biological implications of PTSD secondary to chronic stress is increasingly recognized as identical to that which develop after a trauma. Our position is that until there is clear evidence that symptoms of PTSD secondary to chronic stress are less severe than trauma-related PTSD, we as mental health professionals are obligated to treat AFS in exactly the same way we would treat PTSD. When observed, we need to recognize AFS as a significant mental health disorder that requires professional care and systematic treatment.
This study further reports that over 24% of Americans have a mild version of this disorder. Our data is compelling in suggesting people in this group are in a developmental stage between no/minimal financial stress and Full AFS. As we further explore this condition, we believe it is advisable to see if intervention with this group may be more timely, clinically and cost effective than beginning treatment once Full AFS has been observed.
In considering the full range of financial variables, there are strong associations between financial status and financial stress. As was apparent in the regression analyses: an individual’s sense of being underwater or being unsure if they can pay their bills in the near future were they to lose their income leads to the development of AFS. And even more critically: if an individual is unsure of their ability to pay all of their bills in any given month seems to be the critical factor that pushes people into the condition of AFS. That is when stress appears to tip into the cascade of cognitive, emotional and behavioral disturbances that we see as identical to PTSD.
While this current data on the financial conditions that lead to AFS are compelling and consistent, the current survey also included a wide range of questions on lifestyle, psychological status and medical conditions. These variables are currently under analysis and will be available for presentation by October, 2016. Our goal is not only to explain the financial conditions causing AFS but to also understand the long-term consequences of this condition, both in terms of additional psychological and emotional disturbances, but also in terms of physical health variables.
While we believe we have already gone a long way in establishing the devastating effects of financial stress, by following this for the foreseeable future we also hope to understand this process so well that we can help well-intentioned financial service companies as well as determined individuals to work together to end this condition that is having such a negative effect on both the short and long-term lives of so many Americans. But on an immediate basis, we request that policy-makers and consumer advocates begin to extend their awareness campaigns from just describing the financial conditions that so many Americans face to also including information on the psychological effects of this stress. Our experience with informing people of the presence of AFS has been nearly universally positive, in that people feel affirmed by a recognition of the symptoms they typically have been aware of for some time. We also find that people often feel empowered to seek treatment when they are informed such treatment is available.
Measure of Acute Financial Stress (MAFS-6) 6 Items:
Adapted from the Post-traumatic Stress Disorder Checklist (PCL)
Instructions: Below is a list of problems and complaints that some people have in response to stressful financial experiences. Please read each one carefully, and indicate how much you have been bothered by each problem in the last month. Only answer to the extent these problems are related to financial stress.
Not at all (1) A little bit (2) Moderately (3) Quite a bit (4) Extremely (5)
Thoughts:
1. Repeated, disturbing memories, thoughts, or images of a stressful financial experience from the past, e.g., defaulting on a loan, not being able to provide a living for your family, getting calls from a collection agency.
4. Feeling very upset when something reminded you of a stressful financial experience from the past.
Feelings:
7. Avoid activities or situations because they remind you of your financial stress?
10. Feeling distant or cut off from other people.
Behaviors:
14. Feeling irritable or having angry outbursts.
15. Having difficulty concentrating.
The scoring procedure for the MAFS suggested by the original authors considered an individual to have screened positive if the sum of these items was 14 or greater. However, we found that to result in a large number of false positives because the current Wave 1 National Survey found over 31% of all subjects met this criteria. To correct for this oversensitivity, we reverted to a system similar to the scoring used in the original 17-item PCL.
The PCL requires a specified level of problems occur in each of the three sub-dimensions assessed by the PCL: Emotions, Feelings and Behaviors. To assure symptoms were substantially present in each of these areas, we required items marked as 3 (Moderately) or above to be scored as indicative of AFS. Then, each of the items in the three sub-dimensions were recoded as 0 (not symptomatic) or 1 (symptomatic) and then summed. This provided three sub-dimension scores of 0, 1 or 2. For a diagnosis of AFS to be made, a score of 1 or 2 was required in the Emotion subscale and a score of 2 was required in each sub-dimension of Feelings and Behaviors. This provides a much more specific measure of AFS and this coding system also identifies individuals with a milder stage of AFS. This Mild AFS category is defined as anyone who meets criteria in one or two of the sub-dimensions but not in all three. This group will be carefully studied to try and better understand how AFS develops and if there are ways to prevent those with mild AFS from developing the more severe type of AFS.
Lang, A.J., Stein, M.B. (2005) An abbreviated PTSD checklist for use as a screening instrument in primary care. Behaviour Research and Therapy, 43, 585-594.
Lang, A. J., Wilkins, K., Roy-Byrne, P. P., Golinelli, D., Chavira, D., Sherbourne, C., Rose, R. D., Bystritsky, A., Sullivan, G., Craske, M. G., & Stein, M. B. (2012). Abbreviated PTSD Checklist (PCL) as a Guide to Clinical Response. General Hospital Psychiatry, 34, 332-338.
Weathers, F., Litz, B., Herman, D., Huska, J., & Keane, T. (October 1993). The PTSD Checklist (PCL): Reliability, Validity, and Diagnostic Utility. Paper presented at the Annual Convention of the International Society for Traumatic Stress Studies, San Antonio, TX.
