Financial Services Review, 33(2) 124 The Association of Cryptocurrency and the Use of Alternative Financial Services Gary Curnutt1 and David Smith2 Abstract Alternative financial services (AFS) have been studied in recent years in terms of how these financial markets are utilized. The products and services include check cashing, pawnshop loans, payday advance loans, electronic cash transmissions, tax refund anticipation arrangements, rent- to-own contracts, prepaid debit cards, gift cards, and loans collateralized by automobile titles. Cryptocurrency has become part of this AFS ecology. The 2023 Survey of Household Economics and Decisionmaking collected information on AFS use, including the use of cryptocurrency as an AFS. This research answered the questions: (a) Do users of cryptocurrencies for AFS also tend to use them for investments; (b) do users of cryptocurrencies to make payments tend to use them for other AFS purposes, and (c) do users of cryptocurrencies to send money to friends and family tend to use them for other AFS purposes? Creative Commons License This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License. Recommended Citation Curnutt, G., & Smith, D. (2025). The association of cryptocurrency and the use of alternative financial services. Financial Services Review, 33(2), 124-143. Introduction The rate at which households are considered banked, that is, having access to traditional financial services such as savings and checking accounts, has steadily increased in the United States from 2011 to 2021 (Federal Deposit Insurance Corporation, 2021). The increase was from 91.8% to 95.5% of households banked. There are still unbanked households. Underbanked households also exist, with some access to but not full use of traditional sources of savings, investments, payment methods, and credit (Birkenmaier & Fu, 2023). The unbanked and underbanked use alternative financial services that are generally less efficient, harder to access, and more expensive than traditional 1 Western Carolina University, Cullowhee,, NC, USA. 2 Corresponding author (smithdavid@wcu.edu). Western Carolina University, Cullowhee, NC, USA. services from banks, credit unions, and other financial institutions (CFPB 2016). Since almost 5% of U.S. households face these challenges, it is interesting to understand their use of cryptocurrencies as alternative financial instruments in services and as investments. Alternative financial services (AFS) have been studied in recent years regarding how these financial markets are utilized (Barcellos & Zamarro, 2021; Birkenmaier & Fu, 2023; Fan et al., 2024). The products and services include check cashing, pawnshop loans, payday advance loans, electronic cash transmissions, tax refund anticipation arrangements, rent-to-own contracts, prepaid debit cards, gift cards, and loans collateralized by automobile titles (CFPB 2016). https://creativecommons.org/licenses/by-nc/4.0/ https://creativecommons.org/licenses/by-nc/4.0/ Curnutt and Smith 125 Cryptocurrency has become part of this AFS ecology (Board of Governors of the Federal Reserve System, 2022). For background on the alternative financial service providers before cryptocurrency use, see Prager’s 2014 article “Determinants of the Locations of Alternative Financial Service Providers.” Investing in cryptocurrency has two distinct sets of market participants. Investment professionals, both institutional and sophisticated individuals, use cryptocurrency as an alternative asset class as a portfolio enhancement to hedge against currency volatility, stock market risk, geopolitical instability, and global economic policy stratification (Almeida & Gonçalves, 2023). The other set of participants consists of individual investors acting more as gamblers (or speculators) attempting to anticipate where the other investors will peg their prices either in a rising market or a downward-trending environment, thus potentially profiting from the movement of the cryptocurrency market prices (Roza et al., 2023). The cryptocurrency market capitalization exceeds $1.5 trillion, with over 14,000 separate coins tracked that are available through over 1,000 exchanges worldwide, providing ample room for trading (CoinGecko, 2024). Cryptocurrency investing activity exists both in traditional financial service institutions and in an ecosystem that is an alternative to mainstream investment services. Important to the growth of cryptocurrency as an investment (as well as other uses) is the access provided to the public with the creation of crypto exchanges. These are businesses that facilitate exchanging fiat currencies for cryptocurrencies, where anyone can open an account, deposit dollars, buy cryptocurrency, and do the reverse. Exchanges allow trade against other cryptocurrencies and fiat currencies, such as the dollar or yen. Fees range from .1% to .5% of transactions. Traditional brokerage firms also now provide these services (Giudici et al., 2020). The retail public can also access cryptocurrency through ATMs and other kiosk-type arrangements, paying either a fee (generally 1%) or purchasing at rates higher than exchange rates to incorporate a fee (Brown, 2023). Hypotheses Using 2022 data from the Survey of Household Economics and Decisionmaking we formulate the following hypotheses: H1. Holding cryptocurrencies as an investment is positively associated with using other AFSs. H2. Using cryptocurrencies to make payments is positively associated with using other AFSs. H3. Using cryptocurrencies to send money to friends or family is positively associated with using other AFSs. By exploring these hypotheses, pathways toward increasing access to financial services for banked and underbanked households can be supported through policy decisions and financial institution actions. This understanding will assist in relieving some of the economic burden felt by those having to use AFS services and products because of current barriers and behaviors. Similarly, greater coherence in the investment use of cryptocurrency could be the result of understanding the potentially suboptimal investment decisions (Kim et al., 2023). The Use of Cryptocurrency as an AFS Cryptocurrency has several attributes that make it a compelling alternative financial services product. First, cryptocurrency allows for transaction privacy (Houy et al., 2024). Privacy is possible because the value represented by the cryptocurrency is stored on a decentralized ledger, generally a blockchain, without identifiable ownership data. This storage unit is called a wallet. Value is transferred via anonymous instructions to transfer value from the sender’s wallet to the receiver’s. A second reason cryptocurrency is used is its perceived safety, provided the user selects a safe option suitable to their needs (Houy et al., 2024). Third, transferring value is fast and inexpensive. Fourth, cryptocurrency is not part of the traditional banking system and is thus more acceptable to those distrusting banks and bank regulators. For these and other reasons, cryptocurrency may be the financial instrument of choice for both the Financial Services Review, 33(2) 126 sender and receiver, including for illegal transactions such as money laundering (Wronka, 2022). Besides direct transfers of cryptocurrency between senders and receivers, cryptocurrency can be used via an intermediated transaction, such as buying an Amazon gift card with cryptocurrency, which can then be used exclusively at Amazon online for purchases. Amazon does not accept cryptocurrencies directly, but this method is essentially the same. Services such as Coinsbee.com provide this service for Amazon and many other retailers, as well as prepaid MasterCard and Visa cards (Coinsbee, 2024). Coinsbee.com will sell you a Hotels.com gift card if you want to book a hotel stay. Buy a DoorDash card to pay for your late- night meal delivery. Exchanges are available where people can buy and sell cryptocurrencies using fiat currencies such as the U.S. dollar. Behavioral Finance Aspects of the Use of Cryptocurrencies as an AFS and Investment Using cryptocurrency as an alternative to traditional financial instruments may be a logical, efficient, and effective transaction method and investment instrument. It also may be used based on sub-optimal decisionmaking due to behavioral finance factors. These factors describe potentially irrational financial decisions. For example, one reason for using cryptocurrency is distrust of banks (Board of Governors, 2022). If the distrust is without rational support, perhaps this decision to use cryptocurrency is irrational, and thus, the behavior of using cryptocurrency for transactions is suboptimal. Since people are generally loss- averse, a behavioral finance concept, this distrust may be based on that behavioral finance factor (Sokol-Hessner & Rutledge, 2018). The lived experiences of the unbanked and underbanked households include high overdraft fees at traditional banks. Having had to pay these fees creates avoidance behaviors that result in utilizing these AFSs even if traditional services are available (Dlugosz et al., 2021). Yes, the distrust of banks may be rational, perhaps experientially based, so it is essential not to be rash when declaring irrationality. Herding behavior is a behavioral finance concept prevalent in investing in and using cryptocurrencies (Omane-Adjepong et al., 2021). Herding is when a person follows the crowd, taking cues from others, with no leader necessary. One reason for herding is a natural tendency to fear missing out, sometimes made into the acronym FOMO. Media attention on cryptocurrencies is intense and constant, with thirty or more articles published daily being common (Lee & Jeong, 2023). This attention attracts new participants in this market, both cryptocurrency investors and users of cryptocurrencies as an AFS. The entry of new investors, likely less informed, coupled with the information and misinformation in the media reports, contributes to the volatility of cryptocurrency markets (Lee & Jeong, 2023). In auction markets like cryptocurrency exchanges, good news drives prices higher than fundamentals suggest, while bad news has the opposite effect. This mispricing can be attributed to behavioral finance factors such as overconfidence, recency bias, anchoring, confirmation bias, disposition effect, loss aversion, and risk aversion (Thampanya et al., 2020). Media attention can trigger another emotional response that may drive irrational behavior: narrative influence. This is a powerful influence associated with storytelling. The stories in the media encompass not only what the market is doing in terms of pricing but also the personal stories of individual winners and losers. This type of information can affect the readers’ emotions and result in sub-optimal decisionmaking (Shaffer et al., 2018). Confirmation bias, where a person pays attention only to the stories that fit with their current beliefs, can lead them to make financial decisions without considering all the alternatives and consequences. Behavioral Finance and Literacy Financial literacy is an important variable regarding behavioral finance results. There is much discussion on how to measure financial literacy, including questions to be asked, the number of questions needed for a valid answer, and the saliency of the questions in context, among others (Ouachani et al., 2021). The SHED study used what is commonly called The Big Three questionnaire developed in 2008 (Lusardi & Mitchell, 2014) and used by many large group Curnutt and Smith 127 studies (Ouachani et al., 2021). Financial literacy is positively correlated with financial decisions in that the more literate the individual is, the better their decisions. Better decision-making was based on more rational considerations, while worse decisions were based on less rational behaviors (Kumari, 2020). In other words, the more literate individuals were less susceptible to the behavioral finance concepts that would lead them to less optimal financial decisions. Risk Tolerance The volatility of cryptocurrency is high compared to many other investments (Baur & Dimpfl, 2021), so by definition, it is risky, with risk being deviance from the mean. As an asset, price fluctuations are affected by size premium, attention-driven overreaction momentum, and familiarity (Liu et al., 2022). It would seem, then, that risk tolerance, that is, an individual's susceptibility to making decisions during volatility downwards, would play a large role in participation in cryptocurrency as an investment. Risk tolerance is measured in many ways, from self-reported (willingness to take the risk) to objective and subjective lengthy questionnaires (Omanovic & Zaimovic, 2024). Risk tolerance measurements are designed to understand how an individual views risk and risk’s potentiality. Three categories of measuring instruments are prevalent: a) propensity measures, b) stated preferences, and c) revealed preferences, with all three having validated results. (Eun & Grable, 2024). However, due to behavioral finance concepts such as loss aversion, emotional responses, overconfidence, confirmation bias, recency bias, and others, what people confirm a priori is not constant with actual decisions when losses mount, or gains are excessive (Guillemette & Finke, 2014). The SHED study used self- reporting for stated preferences. To mitigate the effects of behavioral finance factors on decisionmaking, an individual needs to take proactive steps to understand their risk perception and identify biases they may have, such as overconfidence and anchoring. This can be done through education (financial literacy) and consulting with professional financial advisors (coaching) (Almansour et al., 2023). Formal coursework has been shown to reduce behaviors such as the disposition effect, which is the tendency to hold losing investments too long and sell winning investments too early (Paraboni & da Costa, 2021). Relationship Between AFS Use and Cryptocurrency To discuss the relationship between the use of AFS and the use of cryptocurrency in any manner requires a dissection of the reasons people use AFS in the first place. Then, it is possible to associate the perceived benefits of cryptocurrency with the current uses of AFS. Traditional financial products and services have several drawbacks for the unbanked and underbanked, making AFSs more attractive. First, there may not be traditional banks physically or virtually convenient to the person. A prepaid debit card can be purchased and used in many locations at many times of the day or night, if not constantly. Second, the person may not be able to carry the minimum balance required at a bank for an ATM/Debit card. The person can deposit to a prepaid card when possible and carry a zero balance until they have cash. Third, the traditional payment systems contain lags that inconvenience those needing immediate access to their paychecks, thus check-cashing services. This is also true for others who need money now but will not be paid for work until next week, thus payday loans. Waiting for a tax refund may not be acceptable, so the refund anticipation loan scheme exists. It is likely that except for those transactions where the counterparty requires cryptocurrency, these AFS users will not be cryptocurrency users either for transactions or investments. They don’t have money to invest, and their current AFS use is working for them. (Carmona, 2022). Another narrative regarding AFS use is not based on inconvenience or lack of wherewithal. Instead, there is a cohort that uses AFS because, for some, many, or all transactions, they do not want to use traditional providers and instruments. Some people do not trust banks for various reasons. Others may want to avoid banks, not out of distrust, but perhaps to avoid reporting requirements or paper trails. Cryptocurrency may be an added means for transactions and investing in this case due to the privacy available and Financial Services Review, 33(2) 128 perceived efficiency in terms of speed and cost of the transaction. Investment may be due to either intellectually sound investment allocation purposes or speculation aspects that appeal to those interested in quick wealth accumulation (Houy et al., 2024). Combining the elements of a person already disposed to using AFS for transactions with the potential investment gains, it is reasonable to consider that a person would hold wealth in cryptocurrency and use it as needed as an AFS, similar to a stock portfolio occasionally partially liquidated into their checking account for spending. Methods Data This paper uses data from 2022 Survey of Household Economics and Decisionmaking (SHED). SHED is a nationally representative, when properly weighted, survey conducted by the Federal Reserve Board which measures the economic well-being of U.S. households and can help identify potential risks to their finances. We used the 2022 wave due to a more robust set of questions surveying cryptocurrencies compared to older waves of the survey. However, this limited the scope of the survey to just a cross- section. Future research would benefit from exploring longitudinal data spanning several future waves of SHED to observe how the association of cryptocurrency adoption is associated with the use of existing Alternative Financial Services over time. The full sample of the 2022 survey has a total of 11,667 respondents, of which our analysis included 9,326 respondents. Table 1 shows the summary statistics for five demographic control variables for both the full sample and our analysis sample. A two-sample T- test does show statistically significant differences between our analysis sample and the full sample. Notably, our analysis sample has more respondents who report being: male, married, white, having a four-year degree, and an older age. Table 1. Summary Statistics for 2022 Survey of Household Decisionmaking and Analysis Samples 2022 SHED Full Sample Analysis Sample Mean SE Mean SE Female 0.4892 0.0046 0.4700 0.0051** Married 0.5771 0.0045 0.6047 0.0050 *** White 0.6908 0.0042 0.7066 0.0047 ** Four-Year Degree 0.4304 0.0045 0.4529 0.0021*** Age 51.9671 0.1626 52.823 0.1772 *** Note: N of 11,667 for SHED sample & 9,326 for Analysis Sample. Samples are weighted using weights provided by the SHED. Two-Sample T-test: ** denotes statistical difference from the SHED sample mean at the 5% level of significance. *** denotes statistical difference from the SHED sample mean at the 1% level of significance. “SE” denotes Standard Error. Starting with a full sample of 11,667 respondents, observations were dropped for missing or incomplete responses to survey questions that were used in the model. Specifically, observations were dropped from such responses to: “What is the approximate total amount of your household's savings and investments?” (1,841 dropped), and “Where do you think your credit score falls?” (500 dropped). This left us with an analysis sample of 9,326 observations. Table 2 shows descriptive statistics for our analysis sample. In our analysis sample, only about 8.7%, 811 respondents, report holding cryptocurrency as an investment. While about 1.4%, 129 respondents, report using cryptocurrency to make a purchase and about 1.3%, 118 respondents, report using cryptocurrency to send money to friends or family. These numbers illustrate the significant difference in adoption of cryptocurrency as an alternative to conventional financial infrastructure. About 96.6% of respondents report having a checking/savings/ or money market account. However, cryptocurrency adoption seems to be somewhat comparable to other Alternative Financial Services. About 7.7% of respondents in our analysis sample have purchased a money order from a service provided that wasn’t a bank. Curnutt and Smith 129 About 5% of respondents have cashed a check from a service provider that wasn’t a bank. About 2.2% of respondents have taken out a payday loan, about 1.6% have taken out a pawnshop or auto title loan, about 0.8% have obtained a cash advance on their tax-refund and about 9.6% have paid an overdraft fee on a bank account. Appendix A contains a table detailing each variable and how they were constructed. Table 2. Descriptive Statistics of Explanatory Variables - Analysis Samples Mean SE Bought or Held Cryptocurrency as an Investment 0.08696 0.0029 Used Cryptocurrency to Make a Purchase 0.0138 0.0012 Used Cryptocurrency to Send Money 0.0127 0.0012 Has a Checking/Savings/Money Market Account 0.9661 0.0019 Purchased Money Order Outside of Bank 0.0772 0.0028 Cashed Check Outside of Bank 0.0507 0.0023 Taken Out a Payday Loan 0.0216 0.0015 Taken Out a Pawnshop/Title Loan 0.0157 0.0013 Obtained a Tax Refund Advance 0.0080 0.0009 Paid an Overdraft Fee on a Bank Account 0.0959 0.0031 Willing to take Financial Risk 0 - Not At All 0.1629 0.0038 1 0.0587 0.0024 2 0.0833 0.0029 3 0.1092 0.0032 4 0.0937 0.0030 5 0.1855 0.0040 6 0.1047 0.0032 7 0.1079 0.0032 8 0.0579 0.0024 9 0.0123 0.0011 10 – Very Willing 0.0240 0.0015 Race/Ethnicity White, Non-Hispanic 0.7066 0.0047 Black, Non-Hispanic 0.0986 0.0031 Other, Non-Hispanic 0.0455 0.0022 Hispanic 0.1189 0.0033 Two or More Races, Non-Hispanic 0.0303 0.0018 Female 0.4701 0.0052 Married 0.6048 0.0051 Total Savings & Investments Financial Services Review, 33(2) 130 Under $50,000 0.4175 0.0051 $50,000 - $99,999 0.1284 0.0035 $100,000 - $249,999 0.1403 0.0036 $250,000 - $499,999 0.1050 0.0032 $500,000 - $999,999 0.0967 0.0031 $1,000,000 or more 0.1122 0.0033 Where do You Think Your Credit Score Falls? Very Poor 0.0272 0.0017 Poor 0.0454 0.0022 Fair 0.1067 0.0032 Good 0.2351 0.0042 Excellent 0.5856 0.0051 Educational Attainment Less than High School 0.0390 0.0019 High School Diploma 0.1885 0.0039 Some College 0.3195 0.0047 Bachelor's Degree 0.4529 0.0051 Report At Doing Least Okay Financially 0.7596 0.0044 Number of Financial Literacy Questions Answered Correctly 0 0.0714 0.0027 1 0.1327 0.0035 2 0.2630 0.0046 3 0.5328 0.0052 Employed 0.6262 0.0050 Age 52.8235 0.1771 Note: N of 9,326. Sample is weighted using weights provided by the 2022 Survey of Household Decision Making. “SE” denotes Standard Error. Dependent Variables The dependent variables used in our model are variables that ask the respondent about their use or ownership of cryptocurrency in the previous twelve months. First, survey question “S16_a” is a dichotomous variable that asks the question, “In the past year, have you done the following with cryptocurrency, such as Bitcoin or Ethereum? (Bought or Held as an investment.)” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. Our second dependent variable is survey question “S16_b” which asks, “In the past year, have you done the following with cryptocurrency, such as Bitcoin or Ethereum? (Used to buy something or make a payment.)” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. Our final dependent variable is survey question “S16_c” which asks, “In the past year, have you done the following with cryptocurrency, such as Bitcoin or Ethereum? (Used to send money to friends or family.)” Likewise, responses of Yes are coded as a 1, while responses of “No” are coded as a 0. Curnutt and Smith 131 Explanatory Variables The explanatory variables consist of several variables that capture the effects of other Alternative Financial Services as well as several variables that control for other effects that may otherwise impact the association of holding cryptocurrencies and the use of AFSs. • Survey question “BK1” asks the respondent, “Do you (and/or your spouse or partner) have a checking, savings, or money market account.)” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. • Survey question “BK2_a” asks the respondent, “In the past 12 months, did you (and or your spouse or partner) purchase a money order from a place other than a bank?” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. • Survey question “BK2_b” asks the respondent, “In the past 12 months, did you (and or your spouse or partner) cash a check at a place other than a bank?” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. • Survey question “BK2_c” asks the respondent, “In the past 12 months, did you (and or your spouse or partner) Take out a payday loan or payday advance?” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. • Survey question “BK2_d” asks the respondent, “In the past 12 months, did you (and or your spouse or partner) Take out a pawnshop loan or an auto title loan?” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. • Survey question “BK2_e” asks the respondent, “In the past 12 months, did you (and or your spouse or partner) obtain a tax refund advance to receive your refund faster?” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. • Survey question “BK2_f” asks the respondent, “In the past 12 months, did you (and or your spouse or partner) pay an overdraft fee on a bank account?” Responses of Yes are coded as a 1, while responses of “No” are coded as a 0. • Survey question “FL0” is included in the model to proxy for the respondent’s risk tolerance. FL0 is an ordinary variable ranging from 0 to 10. The question asks, “On a scale of zero to ten, where zero is not at all willing to take risk and ten is very willing to take risks, what number would you be on the scale?” • Survey question “ppethm” is a categorical variable that asks the respondent to provide their self-reported “Race/Ethnicity”. We include this in our model as a proxy to control for the cultural effects that differences in race and ethnicity may have on the adoption of cryptocurrency. Respondents can answer 1 for “ White, Non-Hispanic”, 2 for “Black, Non-Hispanic”, 3 for “Other, Non-Hispanic”, 4 for “Hispanic”, or 5 for “2+ Races, Non-Hispanic”. • Survey question “ppgender” is a dichotomous variable that asks the respondent to self-report their gender. Respondents can answer 1 for “Male” or 2 for “Female”. We recode this variable as 0 for “Male” respondents and 1 for “Female” respondents. • Survey question “ppmarit5” is an ordinal variable ranging from 1 to 5. Respondents can answer 1 for “Now Married”, 2 for “Widowed”, 3 for “Divorced”, 4 for “Separated”, or 5 for “Never Married”. We recoded this variable as 0 for responses 2, 3, 4, or 5. This resulted in a dichotomous variable where a value of 0 is not married and 1 is married. • Survey question “pps0596” is an ordinal variable where respondents are asked to self-report their savings and investments on a range from 1 to 7. Respondents can answer 1 for “Less than $50,000”, 2 for “$50,000 to $99,999”, 3 for “$100,000 to $249,999” 4 for “$250,000 to $499,999”, 5 for “$500,000 to $999,999”, 6 for “$1,000,000 or more” or 7 for “Not Sure”. Financial Services Review, 33(2) 132 Responses of 7 are dropped from the sample. • Survey question “ppfs1482” is an ordinal variable that asks the respondent to self- report where they think their credit score falls on a range from 1 to 6. Respondents can answer 1 for “very Poor”, 2 for “Poor”, 3 for “Fair”, 4 for “Good”, 5 for “Excellent”, 6 for Don’t Know”. Responses of 6 are dropped from the sample. • Survey question “educ_4cat” is an ordinal variable that asks the respondent to self-report their level of educational attainment. Respondents can answer 1 for “Less than a high school degree”, 2 for “High school degree or GED”, 3 for “Some college/technical or associate’s degree”, or 4 for “Bachelor’s degree or more”. • Survey question “at_least_okay” is a dichotomous variable that asks the respondent to self-report their subjective financial wellbeing. The respondent can answer 1 for “Yes” or 0 for "No”. SHED includes three questions about basic financial concepts. “FL2” asks the question, “Do you think the following statement is true or false? ‘Buying a single company’s stock usually provides a safer return than a stock mutual fund.’” The respondent can answer 1 for “True”, 2 for “False”, or -2 for "Don’t Know.” “FL4” asks the question “Imagine that the interest rate on your savings account was 1% per year and inflation was 2% per year. After 1 year, how much would you be able to buy with the money in this account?” the respondent can answer 1 for “More than today”, 2 for “Exactly the same”, or 3 for “Less than today.” “FL5” asks the respondent “Suppose you had $100 in a savings account and the interest rate was 2% per year. After 5 years, how much do you think you would have in the account if you left the money to grow?” the respondent can answer 1 for “More than $102”, 2 for “Exactly $102”, 3 for “Less than $102” or -2 for “Don’t Know”. Using these three questions, we construct an ordinal variable that measures the respondent’s total number of correct answers for all the financial literacy questions above (FL2, FL4, FL5). This variable is 0 if the respondent answered 0 questions correct, 7.14% of respondents, 1 if the respondent answered 1 question correctly, 13.27% of respondents, 2 if the respondent answered 2 questions correctly, 26.3% of respondents, and 3 if the respondent answered all three questions correctly, 53.28% of respondents. Survey question “ppemploy” is an ordinal variable that asks the respondent to report their current employment status. Respondents can answer 1 for “Working full-time”, 2 for “Working part-time”, or 3 for “Not Working”. We create a dichotomous variable that is coded as 1 if the respondent reported 1 on ppemoly and 0 otherwise. Survey question “ppage” is a continuous variable that allows the respondent to enter current age. Model This paper examines the association between the use of alternative financial services and the intention of holding or using cryptocurrency for: investment purposes, making payments, and for using to send to friends or family. To accomplish this, we use three separate logistical regression analyses; the three resulting models are as follow: where the latent variable, 𝐼𝑖; is the unobserved net benefit of holding cryptocurrency as an investment for individual 𝑖. The variable 𝐼𝑖 is the observed dichotomous decision of individual i to hold cryptocurrency as an investment. Likewise, the latent variable, 𝑃𝑖; is the unobserved net benefit of using cryptocurrency to buy something or make a payment for individual 𝑖. The variable 𝑃𝑖 is the observed dichotomous decision of individual i to use cryptocurrency to buy 𝐼𝑖 = 𝛼0 + 𝛼1𝐴𝐹𝑆𝑖 + 𝛼𝑘𝑋𝑖 +𝜀𝑖 [1.1] 𝑃𝑖 = 𝛽0 + 𝛽1𝐴𝐹𝑆𝑖 + 𝛽𝑘𝑋𝑖 +𝜃𝑖 [1.2] 𝑅𝑖 = 𝛾0 + 𝛾1𝐴𝐹𝑆𝑖 + 𝛾𝑘𝑋𝑖 +𝜆𝑖 [1.3] 𝐼𝑖 = { 1 if 𝐼𝑖 1 > 0 0 𝑖f 𝐼 ≤ 0 } 𝑃𝑖 = { 1 if 𝑃𝑖 > 0 0 if 𝑃𝑖 ≤ 0 } 𝑅𝑖 = { 1 if 𝑅𝑖 > 0 0 if 𝑅𝑖 ≤ 0 } Curnutt and Smith 133 something or make a payment, for individual 𝑖. The latent variable, 𝑅𝑖; is the unobserved net benefit of using cryptocurrency to send money to friends or family for individual 𝑖. The variable 𝑃𝑖 is the observed dichotomous decision of individual i to use cryptocurrency to send money to friends or family, for individual 𝑖. AFS is a matrix of the seven variables that indicate the use of alternative financial services. These variables include having a bank account, purchasing money orders from places other than a bank, cash check at places other than a bank, use of a payday loan, use of a pawnshop or title loan, use of a tax refund advance, and pay overdraft fees on a bank account. X is a matrix of other explanatory variables, which include financial literacy, credit score, risk tolerance, subjective financial wellbeing, marital status, employment status, total wealth, education, age, ethnicity, and gender. For model 1.1, 𝛼1 is a vector of parameters to be estimated for key explanatory variables, AFS, and 𝛼𝑘 is a vector of parameters to be estimated for the other explanatory variables. The error term 𝜀𝑖 follows a normal distribution. Likewise, in model 1.2, 𝛽1 is a vector of parameters to be estimated for key explanatory variables, AFS, and 𝛽𝑘 is a vector of parameters to be estimated for the other explanatory variables. The error term 𝜃𝑖 follows a normal distribution. In model 1.3, 𝛾1 is a vector of parameters to be estimated for key explanatory variables, AFS, and 𝛾𝑘 is a vector of parameters to be estimated for the other explanatory variables. The error term 𝜆𝑖 follows a normal distribution. For each model, marginal effects are estimated to show associations between the explanatory variables and the observed dependent variables. Results This paper provides empirical evidence on how the adoption of cryptocurrency, or investment therein, is associated with using other AFSs. Table 3 summarizes the marginal effects for each logistical regression model. Holding Cryptocurrencies as an Investment Our results show that holding cryptocurrencies as an investment is associated with using other AFSs. Individuals who reported holding cryptocurrencies as an investment over the last 12 months had a 2.68 percent higher probability of having purchased a money order from a source other than a bank during the same duration of time than those who did not. Individuals who held cryptocurrencies as an investment also had a 3.33 percent higher probability of having taken out a payday loan in the last 12 months. Likewise, individuals who invested in cryptocurrencies also had a 4.2 percent higher probability of taking out a pawnshop or auto title loan in the last 12 months than those who did not hold cryptocurrencies as an investment. Holding cryptocurrencies as an investment was also associated with a 3.82 percent higher probability of having paid an overdraft fee on a bank account over the last 12 months. Our results also show a relationship between risk tolerance as measured by the self-reporting variable FL0 and investing in cryptocurrency. At lower levels of risk tolerance, individuals demonstrate a modestly increased likelihood of cryptocurrency investment. Specifically, at level 2, the marginal effect is 0.0436, indicating a noticeable increase compared to those unwilling to take risks (level 0). At levels 3 and 4, marginal effects are 0.0258 and 0.0315, respectively, suggesting a gradual strengthening of the relationship. The association grows stronger as risk tolerance increases further. At level 5, the marginal effect rises to 0.0422. A marked increase occurs at level 6, where the marginal effect jumps to 0.0704. This upward trend continues, with level 7 showing a marginal effect of 0.1016. At level 8, the likelihood substantially increases, reflected by a marginal effect of 0.1244. The highest associations are observed at levels 9 and 10. Level 9 exhibits the strongest relationship, with a marginal effect of 0.167. Even at the maximum level of risk tolerance (level 10), the marginal effect remains high at 0.1221, highlighting a significant propensity among highly risk-tolerant individuals to hold cryptocurrencies as investments. Using Cryptocurrencies as a Payment The results for model 1.2 show a statistically significant association for several variables measuring the use of AFSs with using cryptocurrency to make a payment over the last Financial Services Review, 33(2) 134 12 months. Using cryptocurrency to make a payment in the last 12 months was associated with a 0.9 percent higher probability of having purchased a money order from a service provider other than a bank, compared to those who didn’t use crypto to make payments in the last 12 months and ceteris paribus. Using crypto as a payment was also associated with a 1.21 percent higher probability of cashing a check from a provider outside of a bank. Using crypto to make payments was associated with a 1.32 percent higher probability of obtaining an advance on an anticipated tax refund. We also find a positive association, 1.48 percent, with using crypto to make payments and paying an overdraft fee on a bank account. We find an association with risk tolerance, although only at some of the higher levels of self- reported risk tolerance. Compared to those who report they are not at all willing to take risk, those who report a risk tolerance of 6, 8 , or 10 (very willing), were all associated with a higher probability of using crypto to make payments in the last 12 months, 1.16 percent, 2.02 percent, and 3.93 percent, respectively. Those who identify as Hispanic, had a 0.9 percent higher probability of using cryptocurrencies to make payments in the last 12 months compared to those who identified as white. Identifying as Female was associated with a 1.18 lower probability of using crypto to make payments. Those who had investments between $50,000 and $99,000 had a 1.05 percent higher probability of using crypto to make payments than those who had less than $50,000 in savings and investments, but none of the other levels of investments had a significant association. Lastly, we find age to have a negative association, 0.01 percent, and employment to have a positive association, 0.81 percent with using crypto to make payments. Using Cryptocurrencies to Send Money The results for model 1.3 show several statistically significant associations with the use of AFSs and using cryptocurrencies to send money to friends or family within the last 12 months. Compared to those who didn’t use crypto to send money to friends or family, those who did had 0.62 higher probability of purchasing a money order from a provider other than a bank in the last 12 months. Sending crypto was also positively associated with cashing a check outside a bank by 0.86 percent. Our results show that using crypto to send money is associated with a 0.92 percent higher probability of taking out a payday loan, compared to those who didn’t use crypto to send money. We find that the use crypto to send money is associated with a 2.24 percent higher probability of obtaining a tax refund advance. Further, we find sending money with crypto is associated with a 0.66 percent higher probability of having paid an overdraft fee on a bank account. Model 1.3 shows some association with self- reported risk tolerance and sending money with crypto. Compared to those who were not at all willing to take risk, those who reported a risk tolerance of 1,5,8, and 10 all had a higher probability of using crypto to send money, 1.19 percent, 0.78 percent, 1.72 percent, and 5.05 percent, respectively. Compared to those who identified as white, those who reported being in any other racial or ethnic identification were more likely to use cryptocurrencies to send money to friends or family. Reporting being Black and non-Hispanic was associated with a 1.7 percent higher probability of using crypto to send money. Reporting being Other non-Hispanic was associated with a 1.61 percent higher probability of using crypto to send money. Reporting being Hispanic was associated with a 0.63 percent higher probability of using crypto to send money, compared to those who reported being white and non-Hispanic. We also find associations with some levels of self-reported credit score and sending crypto to friends or family in the last 12 months. Compared to those who report that they have a very poor credit score, those who report having a poor credit score had a 2.14 percent lower probability of using crypto to send money to friends or family. While reporting having an excellent credit score was associated with a 1.89 percent lower probability of using crypto to send money, compared to those who report a very poor credit score. Lastly, we find that answering all three financial literacy questions correctly was associated with a 0.98 percent lower probability of sending money to friends or family. Financial Services Review, 33(2) 135 Table 3. Marginal Effects of the Three Models Independent Variable Marginal Effect (Standard Error) Crypto as an Investment [1.1] (SE) Crypto as a Payment [1.2] (SE) Crypto to Send Money [1.3] (SE) Has a Checking/Savings/Money Market Account 0.0177 0.0244 -0.0010 0.0065 -0.0077 0.0051 Purchased Money Order Outside of Bank 0.0268** 0.0123 0.009** 0.0040 0.0062* 0.0036 Cashed Check Outside of Bank 0.0182 0.0153 0.0121** 0.0043 0.0086** 0.0041 Taken Out a Payday Loan 0.0334* 0.0196 0.0035 0.0059 0.0092* 0.0053 Taken Out a Pawnshop/Title Loan 0.042* 0.0238 0.0066 0.0062 -0.0032 0.0060 Obtained a Tax Refund Advance 0.0035 0.0263 0.0132* 0.0074 0.0224*** 0.0055 Paid an Overdraft Fee on a Bank Account 0.0382*** 0.0104 0.0148*** 0.0041 0.0066* 0.0039 Willing to take Financial Risk 0 - Not At All (Reference) 1 0.0217 0.0137 0.0037 0.0057 0.0119* 0.0064 2 0.0436*** 0.0120 -0.0018 0.0044 0.0074 0.0053 3 0.0258** 0.0105 0.0033 0.0049 0.0036 0.0042 4 0.0315*** 0.0114 -0.0005 0.0043 0.0002 0.0041 5 0.0422*** 0.0093 0.0050 0.0044 0.0078* 0.0046 6 0.0704*** 0.0119 0.0116* 0.0063 0.0038 0.0051 7 0.1016*** 0.0136 0.0022 0.0046 0.0087 0.0061 8 0.1244*** 0.0168 0.0202** 0.0084 0.0172** 0.0081 9 0.167*** 0.0346 0.0376 0.0232 0.0353 0.0265 10 – Very Willing 0.1221*** 0.0249 0.0393*** 0.0132 0.0505*** 0.0167 Race/Ethnicity White, Non-Hispanic Black, Non-Hispanic 0.0061 0.0110 0.0139 0.0055 0.017*** 0.0056 Financial Services Review, 33(2) 136 Other, Non-Hispanic 0.0236* 0.0138 0.0064 0.0063 0.0161** 0.0080 Hispanic 0.0154 0.0103 0.009** 0.0042 0.0063* 0.0037 Two or More Races, Non- Hispanic 0.0236 0.0180 -0.0007 0.0051 0.0053 0.0068 Female -0.0489*** 0.0067 - 0.0118*** 0.0033 -0.0044 0.0029 Married 0.0006 0.0070 -0.0046 0.0033 -0.0035 0.0033 Total Savings & Investments Under $50,000 $50,000 - $99,999 0.0064 0.0106 0.0105* 0.0058 0.0078 0.0054 $100,000 - $249,999 0.0088 0.0103 0.0018 0.0052 0.0021 0.0054 $250,000 - $499,999 -0.0167 0.0110 0.0045 0.0075 0.0013 0.0064 $500,000 - $999,999 -0.0158 0.0112 0.0018 0.0070 0.0059 0.0110 $1,000,000 or more -0.0093 0.0131 0.0027 0.0063 -0.0014 0.0052 Where do You Think Your Credit Score Falls? Very Poor Poor 0.0272 0.0209 -0.0148 0.0105 -0.0214** 0.0098 Fair 0.0054 0.0169 -0.0094 0.0103 -0.0162 0.0099 Good 0.0432** 0.0174 -0.0118 0.0109 -0.0100 0.0105 Excellent 0.0448** 0.0176 -0.0161 0.0109 -0.0189* 0.0102 Educational Attainment Less than High School High School Diploma -0.0181 0.0211 0.0016 0.0059 0.0021 0.0055 Some College 0.0296 0.0211 0.0039 0.0055 0.0035 0.0053 Bachelor's Degree 0.0307 0.0215 0.0043 0.0055 0.0023 0.0057 Report At Doing Least Okay Financially -0.0206** 0.0087 0.0018 0.0037 -0.0032 0.0038 Number of Financial Literacy Questions Answered Correctly 0 (Reference) 1 0.0014 0.0139 0.0035 0.0057 0.0024 0.0062 2 0.0108 0.0132 -0.0012 0.0054 -0.0034 0.0056 3 0.0458*** 0.0133 -0.0006 0.0053 -0.0098* 0.0055 Employed 0.0362*** 0.0098 0.0081* 0.0001 0.0010 0.0036 Curnutt and Smith 137 Age -0.0023*** 0.0003 - 0.001*** 0.0043 0.0000 0.0001 Note: 9,326 observations from the 2022 Survey of Household Decision Making. *** denotes statistical significance at the 1% level, ** denotes statistical significance at the 5% level, * denotes statistical significance at the 10% level. Discussion and Implications of Key Results The results of this analysis indicate that cryptocurrencies play a role in servicing the financial needs of some consumers, particularly those who face barriers to accessing traditional banking services. The results highlight how cryptocurrencies can function as alternative financial services, particularly for individuals relying on non-traditional banking methods. This demographic often includes the unbanked and underbanked. Users of more traditional alternative financial services may also be enticed to use cryptocurrency to fulfill a need that would otherwise be fulfilled by a traditional financial institution. Cryptocurrencies have also increasingly become popular as alternative means for financial transactions due to their decentralized nature, potential cost-effectiveness, and faster transaction times compared to traditional financial systems. Cryptocurrencies show significant potential as a cost-effective remittance solution. Baur and Dimpfl (2021) identify lower transaction fees and improved speed as key incentives driving individuals and businesses towards adopting cryptocurrencies for payments. While Yermack (2015) highlights cryptocurrencies' potential in improving payment efficiency and transparency, especially across international borders. This study indicates that minority groups, particularly Black and Hispanic individuals, are more likely to use cryptocurrencies for sending money or making payments, when compared to non-Hispanic Whites. This finding is consistent in some regards with the findings of Chatterjee and Yang (2025) who find that ethnic minorities are more inclined to utilize AFS. Traditional remittance services can be prohibitively expensive and slow, whereas cryptocurrencies can offer faster and cheaper options. The use of cryptocurrencies for remittances can reduce costs for those sending money across borders, thereby supporting the financial wellbeing of these communities. However, unlike Chatterjee and Yang (2025), who find the likelihood of the use of AFS is associated with lower objective financial literacy, we find that holding crypto as an investment is associated positively with answering all three financial literacy questions correctly. Zhang et al. (2025) find that other psychological factors may also contribute to crypto ownership as an investment. Each of the three models appears to be influenced in varying ways by individuals’ willingness to accept financial risk. Those with higher risk tolerance are more inclined to own crypto as an investment, consistent with the volatile nature of the asset class and the high uncertainty associated with potential returns. This group may be motivated by speculative opportunities or by a belief in the long-term value of digital assets. Their risk profile aligns with the behaviors of individuals seeking high-reward ventures, often without the safety nets offered by traditional investment vehicles. Three dimensions may influence the relationship between risk-taking preference and investing in a highly risky asset class. First is financial literacy in the domain of cryptocurrency investing. A highly cryptocurrency-literate person would need knowledge of the many fundamental aspects of the currency of interest: the potential volatility, the markets, liquidity, etc. Absent adequate information and the ability to use it, this investor likely exhibits an overconfidence bias. They are investing while overestimating their knowledge and abilities. This situation begs the question, does anyone have adequate information and ability as a cryptocurrency investor? If a practitioner has clients investing in cryptocurrency, it might be prudent to ask; a discussion of the investment and its market risks is in order. However, it may be that the advisor’s compliance requirements do not allow that Financial Services Review, 33(2) 138 discussion. Or, the advisor is not competent to counsel in this area due to their lack of knowledge or ability as a cryptocurrency investor. Overconfidence leads to adverse outcomes in athletics, politics, investing, and most human activities. Second, investors are susceptible to additional biases that affect decisionmaking. Anchoring bias occurs in many human activities where attention is focused on an initial figure. An investor in Bitcoin (BTC), for example, may have entered the market on 9/10/23 at $24,131, considering this a fair price. Another investor may have entered on 3/24/24 at $69,146 and considered that a fair price. If they exhibit anchoring bias, the two investors will likely make very different decisions about the price of BTC if today it was at $49,000, with both decisions not based on the reality of the market for BTC but based on their belief in a fair price. This anchoring bias has a role in another bias, the disposition effect. This effect is when an investor holds losers too long and sells winners too early compared to what a rational investment plan would dictate. What a loser or winner is in the investor’s mind can be based on that initial anchor rather than the true rational value of the asset. Third, investment decisions to buy, sell, or hold are based on the information available to the investor and the ability of the investor to use that information rationally. The investor may have a recency bias and be affected by confirmation bias. With BTC, for example, recency bias will cause the investor to pay more attention to what is being reported now versus the history of BTC prices and market movements. While it is common to say that the past is not a predictor of the future in investments, a well-informed investor considers history valuable information. Confirmation bias is when the investor filters information to pay attention to what confirms their beliefs and avoids contradictory information. If the media mostly says BTC is on an upswing, which the investor wants to hear, they will ignore warning signs from other reports. These two biases skew the information being processed towards irrationality and suboptimal decisionmaking. Other biases likely play a part in the investment decisions regarding cryptocurrencies. Financial planning practitioners should proactively discover which of their clients are cryptocurrency investors. In this way, they can apply their planning knowledge and skills to overcome literacy gaps, counsel on risk-taking behaviors, integrate these investments into an overall plan, and correctly manage their clients’ portfolios given the cryptocurrency investing behavior. Practitioners need to become experts in both the investment side and the behavioral aspects of cryptocurrency investing. In contrast, using cryptocurrency for payment or sending money seems less associated with elevated financial risk tolerance. Instead, these behaviors may reflect practical concerns, such as access, affordability, and convenience. However, there is still a meaningful link to some levels of risk tolerance, suggesting that even among those using crypto for utility rather than speculation, a certain comfort with novel and less-regulated systems is necessary. These users might be more open to financial innovation and may perceive the trade-offs, such as price volatility, as acceptable in exchange for benefits like lower remittance fees or increased transaction speed. Importantly, there is also a notable association between cryptocurrency usage and the use of alternative financial services (AFS), such as payday loans, pawnshop loans, and money orders obtained outside of banks. This connection suggests that for many users, cryptocurrency may serve as a complement to or substitute for traditional AFS. Individuals who rely on AFS often do so because of limited access to conventional banking, and their adoption of cryptocurrency may reflect a search for more accessible, affordable, or immediate financial solutions. In this way, cryptocurrencies may function as part of an informal financial ecosystem that services the needs of the unbanked and underbanked. This connection highlights the potential for crypto to address systemic gaps in financial inclusion, while also raising concerns about the vulnerability of these populations to risk and exploitation in unregulated digital markets. The implications are far-reaching. For consumers, the connection between risk preference and crypto behavior underscores the need for Curnutt and Smith 139 personalized financial education that accounts for individual risk profiles and financial goals. For advisors, it reinforces the importance of discussing crypto investments with clients, not just in terms of returns but also in the context of behavioral and psychological readiness. Advisors must be prepared to help clients understand how their risk preferences may influence both their motivations and decision-making processes across different forms of crypto use. For policymakers and regulators, understanding the interplay between risk tolerance and crypto usage supports the design of more nuanced regulations that address the varying needs of different user groups. Those engaging in high- risk investment behavior may require stronger consumer protections, while users relying on crypto for payments or remittances might benefit more from improved infrastructure and safeguards against fraud or access issues. References Almansour, B. Y., Elkrghli, S., & Al-Mansour, A. Y. (2023). Behavioral finance factors and investment decisions: A mediating role of risk perception. Cogent Economics & Finance, 11(2), 1–20. https://doi.org/10.1080/23322039.2023. 2239032 Almeida, J., & Gonçalves, T. C. (2023). Portfolio diversification, hedge and safe-haven properties in cryptocurrency investments and financial economics: A systematic literature review. Journal of Risk and Fi- nancial Management, 16(1), 1–25. https://doi.org/10.3390/jrfm16010003 Barcellos, S. H., & Zamarro, G. (2021). Un- banked status and use of alternative fi- nancial services. Journal of Pension Eco- nomics & Finance, 20(4), 468–481. https://doi.org/10.1017/S147474721900 0052 Birkenmaier, J., & Fu, Q. (2023). Bank account access methods, household financial wellbeing, and alternative financial ser- vices. Journal of Poverty, 27(1), 33–59. https://doi.org/10.1080/10875549.2021. 2010863 Baur, D. G., & Dimpfl, T. (2021). The volatility of Bitcoin and its role as a medium of ex- change and a store of value. Empirical Economics, 61(5), 2663–2683. https://doi.org/10.1007/s00181-020- 01990-5 Brown, T. (2023). Bitcoin ATMs: What, why, and where. Business Review at Berkeley. https://businessreview.studentorg.berkel ey.edu/bitcoin-atms-what-why-and- where/ Carmona, T. (2022, May 19). Debunking the nar- ratives about cryptocurrency and finan- cial inclusion. Brookings. https://www.brookings.edu/articles/de- bunking-the-narratives-about-cryptocur- rency-and-financial-inclusion/ CFPB. (2016). Supplemental findings on payday, payday installment, and vehicle title loans, and deposit advance products. https://files.consumerfinance.gov/f/doc- uments/Supplemental_Re- port_060116.pdf Chatterjee, S., & Chang, Y. (2025). Utilization of alternative financial services and the role of financial capability. Journal of Con- sumer Affairs, 59(1), e12614. CoinGecko. (2024). 2024 Q1 crypto industry re- port. https://www.coingecko.com/re- search/publications/2024-q1-crypto-re- port Coinsbee. (2024). Buy gift cards with crypto. https://www.coinsbee.com/en/ Dlugosz, J., Melzer, B., & Morgan, D. P. (2021). Who pays the price? Overdraft fee ceilings and the unbanked. FRB of New York Staff Report 973. Federal Reserve Bank of New York. https://papers.ssrn.com/sol3/papers.cfm? abstract_id=3876646 Eun, J. K., & Grable, J. E. (2024). A comparison of financial risk-tolerance assessment methods in predicting subsequent risk tolerance and future portfolio choices. Risks, 12(11), 170. doi: 10.3390/risks12110170 https://doi.org/10.1080/23322039.2023.2239032 https://doi.org/10.1080/23322039.2023.2239032 https://doi.org/10.3390/jrfm16010003 https://doi.org/10.1080/10875549.2021.2010863 https://doi.org/10.1080/10875549.2021.2010863 https://doi.org/10.1007/s00181-020-01990-5 https://doi.org/10.1007/s00181-020-01990-5 https://businessreview.studentorg.berkeley.edu/bitcoin-atms-what-why-and-where/ https://businessreview.studentorg.berkeley.edu/bitcoin-atms-what-why-and-where/ https://businessreview.studentorg.berkeley.edu/bitcoin-atms-what-why-and-where/ https://www.brookings.edu/articles/debunking-the-narratives-about-cryptocurrency-and-financial-inclusion/ https://www.brookings.edu/articles/debunking-the-narratives-about-cryptocurrency-and-financial-inclusion/ https://www.brookings.edu/articles/debunking-the-narratives-about-cryptocurrency-and-financial-inclusion/ https://www.coingecko.com/research/publications/2024-q1-crypto-report https://www.coingecko.com/research/publications/2024-q1-crypto-report https://www.coingecko.com/research/publications/2024-q1-crypto-report https://www.coinsbee.com/en/ https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3876646 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3876646 Financial Services Review, 33(2) 140 Fan, L., Green, L., & Park, N. (2024). Financial stressors and alternative financial service use: Extending the ABC-X model of family stress. International Journal of Consumer Studies, 48(1), 1–15. https://doi.org/10.1111/ijcs.13002 Federal Deposit Insurance Corporation. (2021). FDIC national survey of unbanked and underbanked households. https://www.fdic.gov/analysis/house- hold-survey/2021report.pdf Board of Governors of the Federal Reserve System. (2022). Survey of household economics and decisionmaking. Federal Reserve. https://www.federalreserve.gov/consum erscommunities/shed.htm Giudici, G., Milne, A., & Vinogradov, D. (2020). Cryptocurrencies: Market analysis and perspectives. Journal of Industrial and Business Economics, 47, 1–18. https://doi.org/10.1007/s40812-019- 00138-6 Guillemette, M., & Finke, M. (2014). Do large swings in equity values change risk toler- ance? Journal of Financial Planning, 27(6), 44. https://www.proquest.com/trade-jour- nals/do-large-swings-equity-values- change-risk/docview/1535617637/se- 2?accountid=14968 Houy, S., Schmid, P., & Bartel, A. (2024). Secu- rity aspects of cryptocurrency wallets: A systematic literature review. ACM Com- puting Surveys, 56(1), 1–31. https://doi.org/10.1145/3596906 Kim, K. T., Hanna, S. D., & Lee, S. T. (2023). Investment literacy, overconfidence, and cryptocurrency investment. Financial Services Review, 31(2/3), 121–132. https://doi.org/10.61190/fsr.v31i2/3.353 0 Kumari, D. A. T. (2020). The impact of financial literacy on investment decisions: With special reference to undergraduates in the western province, Sri Lanka. Asian Jour- nal of Contemporary Education, 4(2), 110–126. https://doi.org/10.18488/jour- nal.137.2020.42.110.126 Lee, K., & Jeong, D. (2023). Too much is too bad: The effect of media coverage on the price volatility of cryptocurrencies. Journal of International Money and Finance, 133, n/a. Liu, Y., Tsyvinski, A., & Wu, X. (2022). Common risk factors in cryptocurrency. The Journal of Finance, 77(2), 1133– 1177. https://doi.org/10.1111/jofi.13119 Lusardi, A., & Mitchell, O. S. (2014). The eco- nomic importance of financial literacy: Theory and evidence. Journal of Eco- nomic Literature, 52(1), 5–44. Omane-Adjepong, M., Alagidede, I. P., Lyimo, A. G., & Tweneboah, G. (2021). Herding behaviour in cryptocurrency and emerging financial markets. Cogent Economics & Finance, 9(1), 1–14. https://doi.org/10.1080/23322039.2021. 1933681 Omanovic, A., & Zaimovic, A. (2024). The determinants of financial risk tolerance and portfolio allocation: Have the COVID-19 pandemic and the Ukraine war affected our risk tolerance? Borsa Istanbul Review, 24(6), 1381-1392. doi: 10.1016/j.bir.2024.11.001 Ouachani, S., Belhassine, O., & Kammoun, A. (2021). Measuring financial literacy: A literature review. Managerial Finance, 47(2), 266-281. doi: 10.1108/MF-04- 2019-0175 Paraboni, A. L., & da Costa, N. (2021). Improving the level of financial literacy and the influence of cognitive ability in this process. Journal of Behavioral and Experimental Economics, 90, 101656. https://doi.org/10.1016/j.socec.2020.101 656 Prager, R. A. (2014). Determinants of the loca- tions of alternative financial service pro- viders. Review of Industrial Organization, 45(1), 21–38. https://doi.org/10.1111/ijcs.13002 https://www.fdic.gov/analysis/household-survey/2021report.pdf https://www.fdic.gov/analysis/household-survey/2021report.pdf https://www.federalreserve.gov/consumerscommunities/shed.htm https://www.federalreserve.gov/consumerscommunities/shed.htm https://doi.org/10.1007/s40812-019-00138-6 https://doi.org/10.1007/s40812-019-00138-6 https://www.proquest.com/trade-journals/do-large-swings-equity-values-change-risk/docview/1535617637/se-2?accountid=14968 https://www.proquest.com/trade-journals/do-large-swings-equity-values-change-risk/docview/1535617637/se-2?accountid=14968 https://www.proquest.com/trade-journals/do-large-swings-equity-values-change-risk/docview/1535617637/se-2?accountid=14968 https://www.proquest.com/trade-journals/do-large-swings-equity-values-change-risk/docview/1535617637/se-2?accountid=14968 https://doi.org/10.1145/3596906 https://doi.org/10.61190/fsr.v31i2/3.3530 https://doi.org/10.61190/fsr.v31i2/3.3530 https://doi.org/10.18488/journal.137.2020.42.110.126 https://doi.org/10.18488/journal.137.2020.42.110.126 https://doi.org/10.1111/jofi.13119 https://doi.org/10.1080/23322039.2021.1933681 https://doi.org/10.1080/23322039.2021.1933681 https://doi.org/10.1016/j.socec.2020.101656 https://doi.org/10.1016/j.socec.2020.101656 Curnutt and Smith 141 Roza, T. H., Tavares, H., Kessler, F. H. P., & Pas- sos, I. C. (2023). Problematic trading: Gambling-like behavior in day trading and cryptocurrency investing. Trends in Psychiatry and Psychotherapy (pre-pub- lication). https://doi.org/10.47626/2237- 6089-2023-0623 Shaffer, V. A., Focella, E. S., Hathaway, A., Scherer, L. D., & Zikmund-Fisher, B. J. (2018). On the usefulness of narratives: An interdisciplinary review and theoreti- cal model. Annals of Behavioral Medi- cine, 52(5), 429–442. https://doi.org/10.1093/abm/kax008 Sokol-Hessner, P., & Rutledge, R. (2018). The psychological and neural basis of loss aversion. Current Directions in Psycho- logical Science: A Journal of the Ameri- can Psychological Society, 28(1), 20–27. https://doi.org/10.1177/0963721418806 510 Thampanya, N., Wu, J., Nasir, M. A., & Liu, J. (2020). Fundamental and behavioural determinants of stock return volatility in ASEAN-5 countries. Journal of Interna- tional Financial Markets, Institutions & Money, 65, 101193. https://doi.org/10.1016/j.intfin.2020.101 193 Wronka, C. (2022). Money laundering through cryptocurrencies: Analysis of the phe- nomenon and appropriate prevention measures. Journal of Money Laundering Control, 25(1), 79–94. https://doi.org/10.1108/JMLC-02-2021- 0017 Yermack, D. (2015). Is Bitcoin a real currency? An economic appraisal. In Handbook of Digital Currency (pp. 31–43). https://doi.org/10.1016/B978-0-12- 802117-0.00002-3 Zhang, Y., Naveed, K., & Qi, J. (2025). Crypto investment: The role of investment motivations, investment confidence, and risk perceptions. Financial Services Review, 33(1), 120–141. https://doi.org/10.47626/2237-6089-2023-0623 https://doi.org/10.47626/2237-6089-2023-0623 https://doi.org/10.1093/abm/kax008 https://doi.org/10.1177/0963721418806510 https://doi.org/10.1177/0963721418806510 https://doi.org/10.1016/j.intfin.2020.101193 https://doi.org/10.1016/j.intfin.2020.101193 https://doi.org/10.1016/B978-0-12-802117-0.00002-3 https://doi.org/10.1016/B978-0-12-802117-0.00002-3 Financial Services Review, 33(2) 142 Appendix A Table A1. Variable Used in the Study Variable Type Variable Name Description Dependent Variables S16_a In the past year, have you bought or held cryptocurrency as an in- vestment? (Yes = 1, No = 0) S16_b In the past year, have you used cryptocurrency to buy something or make a payment? (Yes = 1, No = 0) S16_c In the past year, have you used cryptocurrency to send money to friends or family? (Yes = 1, No = 0) Explanatory Variables BK1 Do you (and/or your spouse or partner) have a checking, savings, or money market account? (Yes = 1, No = 0) BK2_a In the past 12 months, did you purchase a money order from a place other than a bank? (Yes = 1, No = 0) BK2_b In the past 12 months, did you cash a check at a place other than a bank? (Yes = 1, No = 0) BK2_c In the past 12 months, did you take out a payday loan or payday ad- vance? (Yes = 1, No = 0) BK2_d In the past 12 months, did you take out a pawnshop loan or an auto title loan? (Yes = 1, No = 0) BK2_e In the past 12 months, did you obtain a tax refund advance to re- ceive your refund faster? (Yes = 1, No = 0) BK2_f In the past 12 months, did you pay an overdraft fee on a bank ac- count? (Yes = 1, No = 0) FL0 Risk tolerance (Scale: 0 = Not at all willing to take risks, 10 = Very willing to take risks) Financial Literacy Variables FL2 "Buying a single company’s stock usually provides a safer return than a stock mutual fund." (1 = True, 2 = False, -2 = Don’t Know) FL4 "If the interest rate on savings is 1% per year and inflation is 2%, how much would you be able to buy after one year?" (1 = More than today, 2 = Same, 3 = Less than today) FL5 "If you had $100 in a savings account at 2% interest per year, how much would you have after five years?" (1 = More than $102, 2 = Exactly $102, 3 = Less than $102, -2 = Don’t Know) FL_Score Number of correct financial literacy answers (0 = 0 correct, 1 = 1 correct, 2 = 2 correct, 3 = 3 correct) Control Variables ppethm Self-reported race/ethnicity (1 = White, Non-Hispanic; 2 = Black, Non-Hispanic; 3 = Other, Non-Hispanic; 4 = Hispanic; 5 = Two or more races, Non-Hispanic) ppgender Self-reported gender (0 = Male, 1 = Female) Curnutt and Smith 143 ppage Respondent’s age (Continuous variable) ppemploy Employment status (1 = Working full-time, 0 = Otherwise) educ_4cat Educational attainment (1 = Less than high school, 2 = High school degree/GED, 3 = Some college/technical/associate’s degree, 4 = Bachelor’s degree or higher) ppmarit5 Marital status (1 = Married, 0 = Not married) pps0596 Self-reported savings and investments (1 = Less than $50,000, 2 = $50,000 to $99,999, 3 = $100,000 to $249,999, 4 = $250,000 to $499,999, 5 = $500,000 to $999,999, 6 = $1,000,000 or more) ppfs1482 Self-reported credit score perception (1 = Very Poor, 2 = Poor, 3 = Fair, 4 = Good, 5 = Excellent) at- least_okay Self-reported financial well-being (Yes = 1, No = 0)