Identity Theft Insurance Undermines Risk Perceptions and Increases Risky Behavioral Intentions

Abstract

A survey was conducted to examine the impact of identity theft insurance on consumer risk perceptions and intentions to engage into risky behavior. The results show that identity theft insurance, as a curative remedy, does undermine individual risk perceptions of being potential victims of identity theft. In addition, we also demonstrate that having identity theft insurance surprisingly increases consumer intentions to involve into activities that could put personal identity at risk. Identity theft insurance, which is believed by consumers to protect their identity, does have unintended impacts with negative consequences for consumer welfare.

Introduction

Identity theft, as the act of taking someone’s personal information to gain financial advantages,[footnoteRef:1] has become one of the most ubiquitous worldwide crimes in recent years. In 2017 alone, United States has more than 15.4 million victims of a total cost of $16.7 billion[footnoteRef:2] loss. The concerns and problems caused by identity theft to consumers go far beyond the direct money losses they bear. For instance, identify theft may upset the victim’s ability to rent an apartment, apply for a new loan, or even get a job. In some extreme cases, the stolen identity might even be used by the thief during a police arrest. [1: Examples include but not limited to stealing from bank accounts, obtaining unauthorized bank loans, establishing insurance policies, and opening unauthorized credit cards.] [2: Source: Insurance Information Institute (I.I.I.)]

As one of the unfortunate facts of our modern life, identity theft does happen. There are some methods can be used to reduce the chance to be the victims of identity theft. As suggested by the Federal Trade Commission, consumers should develop a routine to check their credit reports from three bureaus and statements from banks or credit card companies.[footnoteRef:3] Meanwhile, many consumers turn to identity theft service by purchasing identity theft insurance from the market. The cost of identity theft insurance typically ranges from $25 to $60 per year and may provide benefit limits up to $15, 000.[footnoteRef:4] While identity theft policies are in fact insurance products just like other insurance policies on the market, they often don’t align with what consumers traditionally perceive as insurance. [3: Source: FTC Consumer Information: https://www.consumer.ftc.gov/topics/identity-theft] [4: Source: National Association of Insurance Commissioners (NAIC).]

Unlike traditional insurance policies such as health insurance or homeowner insurance, identity theft insurance policies don’t cover any direct monetary losses due to identity theft activities. The policies only provide coverage to pay for the incurred costs to legally restore a victim’s identity. For example, the policies may cover expenses such as attorney fees, lost wages, notary cost, and even credit monitoring services. In spite of that, consumers commonly misinterpret the true coverage and believe that all future financial losses caused by identity theft activities would and should be covered. Such perceived misconception of identity theft coverage may not only create complaints or disputes between consumers and the insurance companies after financial losses occur but may also affect consumers’ perceptions of the identity theft risk they may face and further their individual behaviors toward such risk. From consumer welfare perspective, we are curious to know whether consumers with identity theft policies tend to exercise less care to protect their own personal information and more likely to engage into risky activities to jeopardize their identity.

The purpose of this research is to investigate the impact of identity theft insurance on individual risk perceptions and risky behavioral intentions. To our knowledge, the literature of identity theft insurance is not mature. Most paper only discuss the prevention or mitigation of identity theft. There has been little systematic effort in consumer behavior or insurance decision-making to study identity theft policy. This paper is the first one to address such gap and contribute to the current literature by discussing identity theft policy as a remedy insurance product with the boomerang consequences on risk perceptions and, in turn, consumer behavioral intentions.

Even though the traditional method to reducing risk mostly emphasizes the promotion of curative remedies which address the incurred risk by reducing the severity of the outcomes, this research examines whether identity theft insurance, as a curative remedy promoted by the insurance industry, has any unintended consequences that may reduce risk avoidance by consumers. That is, does identity theft insurance negatively affect consumers’ risk perceptions toward the risk and boost risky behavior among consumers?

We address this research question by conducting a consumer survey. Participants of the survey were introduced to the identity theft insurance product and their hypothetical purchase decisions were made. All participants were required to response to a list of survey questions reflecting their overall personal risk perceptions and shopping behavior toward the identity theft. By using the data collected from this survey, we were able to elicit and estimate subjects’ risk perceptions toward identity theft and also their intentions to engage into risky online shopping activities which may endanger their personal identity.

We observe that identity theft insurance, as a curative remedy insurance, does undermine our subjects’ risk perceptions of being the victims of identity theft. In addition, having identity theft insurance surprisingly increases the subjects’ intentions to engage into more risky behavior that could put the personal identity at risk when shopping online. The survey subjects appeared to be poorly calibrated regarding the true characteristics of the identity theft policy and were more willing to trade away some of the protection from the identity theft insurance by expressing their willingness to participate into much riskier behavior.

The findings in this paper are significant from consumer welfare perspective since identity theft insurance may harm those consumers who are most in need of help to protect their personal identity. In such case, not only consumers bear the cost of negative consequences of their riskier behavior due to identity theft insurance, society as a whole also does. Our research provides valuable and deep insights for consumers, marketers, and government agencies to understand identity theft insurance. With more identity theft policies sold on the current market, by illustrating unintended negative boomerang effects of this product on consumers, this paper raises important consumer education needs and possible regulatory issue regarding the marketing of this product. This paper proceeds as follows: Section 2 reviews the existing literature and introduces the theoretical foundations. Section 3 develops testable hypotheses. Section 4 outlines the method and the estimation procedure used to test the hypotheses. Section 5 discusses the results and section 6 offers the general conclusions.

Literature Review

Identity theft is a relatively new and serious phenomenon. As a result, a stream of literature focuses mainly on the cause and prevention of identity theft. Milne (2003) designed an exploratory study to measure the behavior self‐reported by college and non-college students on more than ten identity theft preventative activities that were recommended by the Federal Trade Commission. By analyzing the data from the survey, he examined how well consumers have adopted practices that have been suggested to reduce the risk of identity theft. Milne et al. (2004) reported findings from 3 consumer surveys that were used to indicate that consumer tendency to protect oneself from online identity theft diversifies by population. They examined factors such as attitudinal, demographic, and behavioral characteristics that predict the propensity to protect online identity and privacy. Newman and McNally (2005) suggested that research is in need of assessing the issue of reporting and recording identity theft by law enforcement. Hoofnagle (2005) indicated from a legal perspective that a thorough change to the framework of privacy is necessary when combating identity theft. A fix was proposed in the paper to address the slack credit granting practices by changing the default state of credit reports from its current liquid state to the frozen one. Sauer (2006) investigated consumer attitudes toward willingness to pay for security features due to the identity theft and found that a large number of identity theft victims are actively considering security freeze legislation. Anderson (2006) examined the correlations between a consumer's demographic characteristics and the likelihood of being the victim of any identity theft activities by using the survey data collected by the Federal Trade Commission. His paper found out that young women with relatively higher household income appear to have higher probability to experience identity theft. Using a panel data from the Federal Trade Commission, Romanosky et al. (2011) investigated whether the adoption of data breach disclosure laws may efficiently reduce identity theft caused by data breaches. Lai et al. (2012) studied conventional coping and technological coping behaviors consumers exhibited when fighting identity theft. Their results illustrated that both these two coping behaviors are effective to defend against identity theft.

Researchers since then direct their attentions to the identity theft impact on consumer welfare. Hille et al. (2015) developed a scale to measure the consumer fear of online identity theft and discussed how such fear contributes to the side effect of e-commerce. Kahn and Linares-Zegarra (2016) discussed how identity theft has affected consumers’ payment choices. Their paper showed that particular types of identity theft incidents have a positive and significant effect on the probability of consumers’ adoption of checks and online banking bill payments. Drawing on the coping literature, Li et al. (2019) studied identity theft victims’ responses and antecedents to the responses. By using a survey of 197 identity theft victims, they demonstrated that perceived victimization severity due to identity theft has a positive impact on perceived distress and, in turn, a positive impact on the behavioral responses.

In the field of consumer research, traditional models argue that risk remedy attempts to discourage risky behavior by amplifying the negative consequences from the perceived risks of the behavior. Floyd et al. (2000) illustrated that increases in both threat appraisal and coping appraisal apparently have positively reinforce more protective behavior. Risk appraisals as key determinants of decisions and actions were tested by Sheeran et al. (2014). They also found that heightening risk appraisals do change consumer intentions and behaviors. When anticipatory emotions or perceived severity was increased, the heightening risk perceptions had relatively larger effects.

Compared to traditional models, research in risk compensation have presented evidence to suggest that risk remedy may have unintended negative consequences that in fact harm consumer welfare (e.g., Rogers and Greenfield 1999). Bolton et al. (2006) proposed a conceptual framework to discuss how the impact of remedy on the individual risky behavior could be moderated by this individual’s current relationship to the problem domain. They explained that the marketing of remedy messages may undermine risk perceptions as consumer problem status rises while the risky behavioral intentions rise. Bolton et al. (2011) used two experiments to demonstrate the negative impact of the marketing of debt consolidation loans on consumers who have mounting debt problems. They argued that debt consolidation loans which offers a financial remedy to consumers actually overstated the short-term benefits. A financial literacy intervention was discussed in the paper to assist marketing financial remedy products.

Hypotheses

Risk Perceptions

Research on persuasive messages to curb risky behavior has been premised of the theoretical framework that curative remedies attempt to reduce risky behavior by heightening the perceived risks of a behavior. For example, drivers will be told by their insurance agents that the chance for any drivers involving into car accidents are very high and they can be personally responsible for any financial losses caused by bodily injury or property damage. Floyd et al. (2000) indicated that consumers who purchase insurance products increase their perceived effectiveness of protective behaviors. Based on this framework, it suggests that individuals with identity theft coverage should be more conservative and risk averse when evaluating the possible exposures to the identity theft.

Approach to reducing risk by using insurance products as curative remedy focuses on mitigating risk by decreasing the severity of the consequences. For example, auto insurance offers to help drivers reduce the severity of financial losses due to auto accidents. In contrast to the previous framework, risk compensation research that has analyzed aggregate behavior suggests that risk compensation exists in various remedies (e.g., Calkins and Zlatoper 2001) and the remedies may have unintended outcomes that harm consumer welfare. Individuals who are covered by insurance may have misconception that they are protected from any possible financial losses and hence become more confident when assessing their risks. According to this risk compensation theory, individuals with identity theft coverage are more likely to have their perceived risks toward the identity theft reduced or eliminated. That is, identity theft insurance may have negative impact on individual perceived risks to be the victim of the identity theft. Specifically, this leads to the following prediction for the risk perceptions: Hypothesis 1: Identity theft insurance undermines individual risk perceptions. If supported, this finding would demonstrate the unintended effect of identity theft insurance on consumers. By purchasing this insurance coverage, consumers increase their perceived effectiveness of the protection from the policy and, thereby, reduce the perceived risks associated with the identity theft.

Risky Behavioral Intentions

A vast of literature (e.g., Ganderton et al. 2000) on insurance purchase decision-making argues that risk averse individuals would purchase insurance and undertake all relevant precautions to the extent that the extra benefits from such actions exceed the marginal costs, less some risk premium in the case of risk aversion. If this argument is supported, risk averse consumers who are willing to pay for identity theft insurance should be more risk averse and behave more cautious and prudent. We then expect to see less engagement in the risky behavior.

Meanwhile, risk compensation from consumer research argues that a curative remedy, such as insurance policy, reduces the costs or risks of a target behavior and people may trade away some of this gain in safety and engage in riskier behavior (Bolton et al. 2006). To illustrate, drivers with auto policy are likely to perceive that policy may lower the risks and relevant costs of auto accidents, which could encourage risky driving behavior. From a moral hazard perspective, just as people may take less care of their health once they have health insurance (Zeckhauser 1995), some consumers appear less risk averse when remedies are available. Under such framework, identity theft insurance may encourage risky behavior that could put the consumer’s identity at a risk. Specifically, this leads to the following prediction for risky behavioral intentions: Hypothesis 2: Identity theft insurance increases individual risky behavioral intentions. Due to the misconception of identity theft insurance and the exhibition of misprocessing behavior, consumers who are protected by this insurance products becomes less care or aware of identity theft scenarios around them in the market. All incurred possible losses due to identity theft are misbelieved to be covered by their identity theft coverage. These individuals tend to be more willing to take risk compared to those who are not covered by the identity theft insurance partially due to.

Method and Estimations

Most literature study risk behavior use either survey or experiment to collect data (Outreville 2014). In this paper, we conducted a consumer survey to test our hypotheses related to the impact of identity theft insurance on individual risk perceptions and intentions to engage in risky behavior. College students were recruited as the participants for the survey. One of the main reasons to have college students as our subjects is that they appear to be high in problem status of identity theft because of their daily internet and financial behavior according to U.S. Department of Education.[footnoteRef:5] [5: U.S. Department of Education 2004. www.ed.gov/about/offices/list/oig/misused/idtheft.html.]

The survey consisted of two phases. In the first phase, participants read written introductory information about the risk of identity theft. Then they were given the opportunity to decide whether they would like to buy the identity theft insurance which was described as the insurance coverage to reduce the consequences of identity theft at a fair price. In the second phase, all participants responded to a series of questions rating their overall personal risks on a five-point scale. They were asked to weigh in their concerns regarding the financial risks of identity theft and how likely for themselves to be the victims of identity theft activities. Next, participants were instructed to imagine themselves using their personal credit cards to shop online for a friend’s upcoming birthday gift. Participants were asked to indicate and rank their preferences on three different shopping websites that vary in terms of the amount of personal information requested and collected.

Participants who indicated their willingness to buy identity theft insurance in the first phase were initially instructed to “assume that you have identity theft insurance” and then answered the survey questions in the second phase. Meanwhile, participants who showed no interest to buy identity theft insurance answered the exactly same set of questions but without the assumption of having identity theft insurance. Individual demographic information such as age, gender, household income and smoking behavior was also collected from the survey.

To measure the impact of identity theft insurance on individual risk perceptions, survey question related to the financial risks (Financial Risk) was used as the dependent variable in our model. The Financial Risk was scaled as “Not at all concerned (1)”, “Slightly concerned (2)”, “Somewhat concerned (3)”, “Moderately concerned (4)”, and “Extremely concerned (5)”. Since ordinal responses as dependent variables were involved in the model, the ordinal response regression model provided better ways of estimating parameters using the maximum likelihood method. The general form of the model is, where is a cutpoint. Independent variables used in the model included participants’ characteristics (e.g., gender, age, income level) and their decisions to have identity theft insurance coverage from the first phase of the survey. Further, to verify the impact of identity theft insurance on risk perception, another dependent variable, Victim from the survey question of being potential victims of the identity theft activities, was also used to replace “Financial Risk” in the model. [6: Decision to buy identity theft insurance was coded as follows: Yes (1) and No (0).]

To examine the impact of identity theft insurance on risk behavioral intentions, three survey questions related to the subject’s willingness to provide social security number or mother’s maiden name and use any unsecured websites when doing online shopping were used as dependent variables respectively to elicit the risky behavior subjects might be intended to exercise. They were scaled as “Would not buy (0)”, “Might or might not buy (1)”, and “Definitely buy (2)”. Participants’ characteristics and their decisions to have identity theft insurance coverage were also used in these models as the independent variables.

Results

Model 1 captures the significant impact of having identity theft insurance on perceptions of financial risk consumers may have due to identity theft. The dependent variable Financial Risk was defined as five different levels of concerns.[footnoteRef:7] As the results from Table 1 illustrate, the variable Id Theft Policy is significantly negative at the 1% level for both logit (Model 1a) and probit (Model 1b) assumptions. Examining the results from Model 1, we find support for our hypothesis H1, suggesting that identity theft insurance does undermine our participants’ risk perceptions. When being asked to rate their personal concerns regarding the financial risks of identity theft, participants who chose to be covered by identity theft insurance in the survey were more likely to be less risk averse compared to those who chose not to have identity theft insurance. [7: The levels of concern were coded as follows: Not at all concerned (1), Slightly concerned (2), Somewhat concerned (3), Moderately concerned (4), and Extremely concerned (5).]