This paper presents a qualitative study of Reddit comment data, investigating how individuals leverage consumption to protect themselves from criminal threats. Our analysis finds that people consume to serve three strategic functional categories of protective consumption behaviors: threat-detecting, vulnerability-reducing and severity-reducing strategies. Within these overarching strategies, our analysis of the field data also supports consumers’ use of four particular tactical functions: withdrawal, deterrence, resistance, and recovery. We leverage protection motivation theory as an enabling lens to analyze the data; in doing so, we offer contributions to the theory as well. We discuss these strategies/tactics by introducing what we term the Crime-Protective Consumption Framework. A deeper understanding of how consumers derive value from consumption activities to protect themselves, others, and their property from crime can help guide new product development and marketing messages, and more broadly, illuminate how criminal hazards influence consumer decision-making.
Author(s): Robert Arias, Dallas Novakowski, Miranda Yin
This study analyzes how consumer overconfidence influences sequential behaviors related to financial influencers. Data were drawn from the 2024 Fund Investor Survey conducted by the Korea Financial Consumers Protection Foundation. Overconfidence was categorized into three subtypes: overestimation, overplacement, and overprecision. Dependent variables were defined across four stages of finfluencer engagement: viewing, subscribing, investing, and experiencing harm. Logistic regression was employed as the primary method, complemented by decision tree and random forest models to enhance explanatory power and predictive accuracy. The results demonstrate that overconfidence exerts stage-specific effects. Overplacement increases the likelihood of viewing finfluencer content, while overprecision consistently shows negative effects at the subscription and investment stages. Overestimation, in turn, significantly elevates the likelihood of consumer harm. By contrast, consumer resources and competencies—including financial literacy, risk tolerance, financial planning horizon, assets, and age—emerged as the most influential predictors in the decision tree and random forest models, underscoring their critical role in shaping overall engagement. These findings yield two key implications. First, distinct types of overconfidence biases act as significant determinants of consumer engagement at different stages. Second, consumer protection strategies should extend beyond correcting psychological biases to strengthening financial literacy and addressing age-specific differences in risk tolerance and financial planning horizon. Overall, the results provide meaningful insights for the design of educational and policy measures aimed at protecting retail investors in the era of financial influencers.
A significant body of literature examines the influence of financial confidence, overconfidence, and/or underconfidence on a variety of financial behaviours and outcomes. Yet researchers have used a host of different methodologies to produce this work. This paper describes the results of reviewing 66 articles and reports the focus concepts (confidence, over/underconfidence), the operational definitions, the data used, and the various ways in which the key variables were constructed in existing research. It concludes by identifying decision points for scholars conducting future research in this area.
This research, “Beyond Financial Education and Knowledge,” explores how objective financial knowledge and past exposure to financial education shape individuals’ future long-term investment motivation. Using a quantitative, correlational study design, the research uses existing data from the 2021 United States Financial Industry Regulatory Authority’s (FINRA) National Financial Capability Study (NFCS), analysing a nationally representative sample of 2,824 respondents. Results from regression analysis reveal an upbeat and statistically positive relationship between objective financial knowledge and long-term investment motivation, confirming that objective financial knowledge is a strong predictor. Financial education did not demonstrate a significant impact on long-term investment motivation. Demographic variables of gender, ethnicity, education, and income were significantly associated with motivation, while age was not. These findings provide insights for policymakers and financial educators to improve the finance education topics offered in the K-12 curriculum and other financial education interventions to strengthen this foundational offering and enhance future financial decision-making.
When faced with a financial question, a person may answer correctly, answer incorrectly, or state that they don’t know the answer. The purpose of this study is first to explore the significance of answering a financial question incorrectly in contrast to answering “don’t know”. Second, to investigate how taking a financial education course may be associated with the responses a person provides. We will look both at the behaviors associated with answering don’t know compared to answering incorrectly, as well the relationship between taking a financial education course and what kind of responses a person makes to financial knowledge questions. Our analysis suggests that answering incorrectly rather than answering “don’t know” is associated with an increase in problematic financial behavior. Therefore, we should be cautious in suggesting people trust they know the answer to financial questions, as if it pushes them to answer incorrectly it could have harmful consequences. We also find that taking a course in financial education is associated with an increase in the number of incorrect responses a person provides. This suggests there are some previously under-appreciated risks to financial education as it is currently being provided.
This study analyzed the 2021 National Financial Capability Study data set to explore the role of perceived and actual financial education on financial wellbeing. Using both the direct financial education question and the state of residence variable, we investigated the role of financial education on Financial Well Being. Based on previous literature, financial literacy (objective and subjective), use of Alternative Financial Services, and various socio-economic factors were included as control variables.
Author(s): Peter Kreysa, Soo Hyun Cho, Hofner Rusiana
The rapid rise of artificial intelligence (AI) in finance promises to make financial advice less costly and algorithmically objective. However, little is known about how negative financial experiences, particularly traumatic ones such as fraud, influence consumers' decisions when choosing AI for financial advice. Consumer fraud is widespread, and research shows it severely damages victims' financial well-being by eroding trust and confidence (Brenner et al., 2020). This creates a paradox: those who need guidance most may be least willing to trust human advisors. This study bridges research on the consequences of fraud and fintech adoption to address a critical gap: Do fraud victims seek out AI financial advice as an alternative? We propose an "algorithmic trust hypothesis," suggesting that when human intermediaries cause harm, victims may transfer their trust to algorithms perceived as more objective and impartial. The primary objective of this paper is to investigate whether consumer fraud victimization increases the demand for AI financial advice and to examine the roles of digital fluency and eroded trust in financial institutions in this relationship. Using NFCS state-by-state survey data, we contribute to the understanding of how fraud actively drives behavioral adaptation, potentially reversing the well-documented tendency to distrust algorithms (Dietvorst et al., 2015). The findings hold significant implications for researchers, financial planners, and consumer protection advisors.
Artificial intelligence (AI) is reshaping the financial services landscape, yet little is known about which consumers are most interested in adopting AI financial advice. This study examines the factors associated with consumer interest in AI financial advice and explains heterogeneous adoption pathways. Using nationally representative data from the 2024 National Financial Capability Study, the research combines logistic regression with k-means clustering within a Financial Capability and Digital-Psychological Readiness framework. Regression results reveal that interested consumers tend to have higher objective financial knowledge, stronger financial planning habits, greater risk tolerance, and more developed digital financial habits. In contrast, higher subjective knowledge is negatively associated with interest after controlling for other factors. The cluster analysis, organized around the constructs of financial capability and readiness, identifies four distinct consumer segments with divergent motivations for adoption, clarifying counterintuitive regression findings. The results provide actionable insights for practitioners seeking to design tailored targeted AI solutions and for policymakers concerned with digital inclusion and consumer protection.
This study examines whether the breadth of mobile financial services usage is associated with consumer interest in AI-based financial advice and whether financial confidence mediates this relationship. Despite financial institutions investing nearly $97 billion in AI by 2027, fewer than 3% of U.S. households use AI-based financial advice services, revealing a critical supply-demand gap. Using nationally representative data from the 2024 National Financial Capability Study, this research employs survey-weighted structural equation modeling to test three hypotheses linking mobile finance usage breadth to AI advice interest via financial confidence.
This study examined the relationships between digital access, digital literacy, and psychological well-being among adults aged 50 and older using nationally representative data from the 2022 Health and Retirement Study. Digital literacy was modeled as a multidimensional construct encompassing functional, social-communicative, and leisure/productivity skills. Using Structural equation modeling, we found that digital literacy was closely associated with psychological well-being, reflected in purpose in life and depressive symptomatology. Digital access showed a modest direct association with psychological well-being, but its relationship through digital literacy was more pronounced, highlighting the relevance of digital engagement. These findings suggested that variations in digital literacy were linked with differences in well-being among older adults. The study contributed to consumer interest research by identifying digital literacy as an important correlate of psychological well-being and informing programs and policies that promote digital skills, inclusion, and healthy aging in a technology-driven society.
The purpose of this study was to explore the link between personality profiles and the ownership of annuities to increase our understanding of the psychological phenomena that may help explain the annuity puzzle. Two weighted, inferential analyses were conducted. Personality profiles representing unique combinations of the Big Five personality trait variables were identified using latent profile analysis. These variables were then used as predictors of annuity income ownership in a binary logistic regression. Personality profiles were found to be statistically significant predictors. Implications for research, policy, and practice were discussed.
Robert O. Herrmann Outstanding Dissertation Award Winner
Financial well-being in later life is shaped by far more than income and assets. This award-winning dissertation applies Engel's (1977) Biopsychosocial (BPS) Model to financial well-being for the first time, examining how biological, psychological, and sociological factors interact to predict financial outcomes among older adults. Using five waves (2010–2018) of the Health and Retirement Study (HRS) and structural equation modeling with over 35,000 respondents, the study operationalizes physical health, mental health, and social connection as higher-order latent constructs and tests their direct, indirect, and combined effects on a multidimensional measure of financial well-being. Results reveal that the integrated BPS model explains up to 99% of variance in the financial well-being construct, with mental health emerging as the dominant predictor. Physical health exerts positive but variable direct effects and meaningful indirect effects through psychological pathways, while social connection shows a surprising negative direct association, suggesting that maintaining social ties may carry financial costs even as those ties bolster mental health. Findings challenge purely economic models of financial wellness and offer actionable intervention points for financial planners, policymakers, and health professionals working with aging populations.
The present study addresses the research needs by examining how consumer's objective and subjective financial knowledge is associated with their interest in using AI for financial planning purposes. Additionally, the study considers whether prior technology adoption further moderates this relationship. This research seeks to contribute to theory and practice in consumer financial behavior by clarifying these mechanisms. It offers insights that can inform education, policy, and the responsible development of AI-driven financial tools.
Author(s): Ravisha Chutani, Kimberly Watkins, John Grable
This study examines how smart infrastructure and services affect consumer decision-making, willingness to pay, and housing affordability. Using Van Westendorp’s Price Sensitivity Measurement (PSM) model, we surveyed consumers to identify acceptable price ranges for smart service utilization and smart space development. PSM reveals the price points at which consumers view products as too expensive, too cheap, or good value, providing insights for setting strategies when introducing new technologies. Complementing this approach, a hedonic pricing analysis of more than 100,000 housing units in South Korea evaluates how smart features and recurring service costs are capitalized into housing prices. Findings indicate that while consumers are willing to pay premiums for convenience and efficiency, these premiums also contribute to affordability challenges and regional disparities. By integrating consumer-based willingness-to-pay data with market-based housing price effects, this study offers a comprehensive understanding of how smart technologies shape consumer financial outcomes. The results highlight the need for transparent pricing, equitable housing policies, and informed consumer decision-making to ensure that technological innovation enhances household well-being without limiting access to affordable housing and essential services.
This study examines the causal impact of payday lending access restrictions on household consumption patterns using Ohio's 2018 Fairness in Lending Act as a quasi-experimental setting. We employ a difference-in-differences approach comparing employed, low-income households in West Virginia and Pennsylvania counties bordering Ohio (treatment group) with similar households in non-border counties (control group). Using NielsenIQ Home Scanner data from 2016-2019, we find that households losing cross-border payday lending access experienced significant decreases in total monthly spending. Disaggregated analysis reveals that spending reductions primarily occurred in necessities, particularly food, rather than discretionary items like alcohol and tobacco. Alcohol spending showed no significant change, while tobacco spending declined only marginally. These findings challenge the welfare rationale underlying payday lending restrictions, suggesting that such policies may force vulnerable populations to reduce essential consumption rather than eliminating harmful spending behaviors, potentially exacerbating rather than alleviating financial hardship among low-income households.
