Vulnerability and fraud: evidence from the COVID-19 pandemic

The COVID-19 pandemic created an unprecedented global crisis, causing not only physical suffering but also significant psychological and mental health issues. This environment of vulnerability became a breeding ground for fraudulent activity, with scammers exploiting the widespread anxiety and uncertainty. Between January 1, 2020, and January 19, 2022, U.S. victims reported over 671,868 fraud complaints, resulting in losses exceeding $669.67 million. These complaints spanned various categories, including online shopping, credit cards, and government benefits, often mentioning COVID-19 related terms. Similar increases were observed in the United Kingdom and Australia. Figure 1 illustrates the parallel increase in fraud incidents and confirmed COVID-19 cases in the U.S., particularly evident from March to June 2020. Figure 2 maps the severity of consumer fraud across U.S. states, revealing significant variations. This study investigates the relationship between the spread of COVID-19 and the surge in fraud and scams, addressing three key questions: (i) the extent of pandemic-related fraud, (ii) the factors driving this increase, and (iii) policy recommendations for mitigation. Two competing hypotheses are explored: the vulnerable-to-become-victimization hypothesis, based on the opportunity model of predatory victimization, and the vulnerability-risk-aversion hypothesis, which considers the role of dopamine in decision-making under uncertainty. The opportunity model posits that victimization risk increases with exposure to offenders, proximity to them, lack of guardianship, and target attractiveness. The pandemic heightened all these factors: social distancing increased online exposure, lockdown reduced guardianship, and psychological distress made individuals more susceptible. Conversely, the vulnerability-risk-aversion hypothesis suggests that uncertainty surrounding the pandemic might lead to risk-aversion and potentially fewer fraud cases due to low dopamine levels associated with unfamiliar domains.

The study draws upon existing literature on crime victimization, including the lifestyle/exposure theory, routine activity theory, and the opportunity model of predatory victimization. The opportunity model, which considers exposure, proximity, guardianship, and target attractiveness, serves as a key framework. Prior research on consumer deception often focuses on misleading advertising, while studies on consumer fraud victimization utilize surveys or experimental data. This study distinguishes itself by applying the opportunity model to the context of the COVID-19 pandemic, an unprecedented societal shock. Existing research also highlights the impact of financial literacy on economic decision-making and the vulnerability of older adults to fraud. The study integrates these perspectives to investigate the specific dynamics of fraud during the pandemic.

The study employs a quantitative approach, using data on COVID-19 cases in the U.S. from January 4 to July 28, 2020, obtained from Our World in Data, and fraud and scam complaints filed with the Federal Trade Commission (FTC) during the same period. Two key variables are constructed: (i) the scam case daily growth rate (ASCAM_CASES), calculated using a seven-day moving average to account for reporting biases, and (ii) the COVID-19 daily new case growth rate (LAGACOVID_CASES), also calculated using a seven-day moving average. A regression model is developed to analyze the relationship between the lagged COVID-19 spread and the increase in scams and fraud cases (Equation 3). Control variables include the average daily return of the S&P 500 index, a dummy variable for non-trading days, a measure for seasonal affective disorder (SAD), and the government stringency index. Descriptive statistics (Table 1) summarize the key variables. To address the potential confounding factor of increased awareness and reporting, the study innovatively employs Google Trends data. Observations are divided into high and low pandemic stress days and high and low fraud awareness days, based on the Google Search index for relevant keywords. Regression analyses are repeated for these subsamples (Tables 3 and 4). Robustness checks are conducted using alternative measures of fraud and COVID-19 growth rates (Table 5). Finally, a state-level analysis examines the impact of financial literacy on finance-related versus non-finance-related fraud (Table 6). This analysis uses data on total reported fraud cases in each state, scaled by population, and incorporates control variables such as personal income, education level, poverty rate, and religious proportion.

The baseline regression results (Table 2) demonstrate a significantly positive association between the spread of COVID-19 and the increase in fraud cases. A 10% increase in confirmed COVID-19 cases was associated with a 1.285% daily increase in fraud cases, representing a substantial jump compared to the average daily increase. Figure 5 visually represents this relationship. Control variables reveal that stock market downturns and seasonal affective disorder are also associated with increased fraud. The Google Trends analysis (Tables 3 and 4, Figures 6 and 7) demonstrates that the increase in fraud cases is primarily driven by heightened pandemic stress and vulnerability, not increased awareness and reporting. The state-level analysis (Table 6) reveals a positive relationship between the spread of COVID-19 and finance-related fraud in states with low financial literacy, but no such relationship exists in states with high financial literacy. Robustness checks using alternative measures of fraud and COVID-19 growth rates confirm the significant positive association between the spread of COVID-19 and the increase in fraud cases.

The findings strongly support the vulnerable-to-become-victimization hypothesis. The increase in fraud during the pandemic was primarily driven by the increased vulnerability of individuals experiencing stress, isolation, and impaired cognitive function due to the pandemic’s impact. The Google Trends analysis effectively rules out the alternative explanation of increased awareness and reporting as the primary driver. The significant role of financial literacy in mitigating finance-related fraud highlights the importance of financial education and awareness programs. The finding that financial literacy did not affect non-finance-related scams suggests the need for targeted interventions.

This study provides critical evidence of the opportunistic exploitation of pandemic-induced vulnerability by fraudsters. The findings underscore the need for comprehensive fraud prevention programs that focus on improving cognitive functioning, enhancing financial literacy, and reducing social isolation, particularly for vulnerable populations like older adults. Future research could benefit from accessing more detailed data, including daily state-level data, victim characteristics, and a more comprehensive pre-pandemic baseline to quantify the impact more precisely.

The study's reliance on FTC complaint data presents limitations. Daily data from before the pandemic are unavailable, hindering a direct comparison of pre- and post-pandemic fraud rates. Detailed victim and crime characteristics are also lacking. The absence of daily state-level data limits the exploration of local government policies and interventions. Access to demographic profiles of fraud victims would strengthen the analysis.