The Effect of Credit on Spending Decisions: The Role of the Credit Limit and Credibility

Abstract

The objective of the present research is to study consumer decisions to utilize a line of credit. The life-cycle hypothesis from economics argues that consumers should intertemporally reallocate their incomes over their life stream to maximize lifetime utility. One form of intertemporal allocation is to use past income (in the form of savings) in the future. A second form is the use of future income in the present. This can only be done if consumers have access to a temporary pool of money that they can draw from and replenish in the future—a function performed by consumer credit.
However, our research reinforces prior findings that consumers are unable to correctly value their future incomes, and that they lack the cognitive capability to solve the intertemporal optimization problem required by the life-cycle hypothesis. Instead, we argue that consumers use information such as the credit limit as a signal of their future earnings potential. Specifically, if consumers have access to large amounts of credit, they are likely to infer that their lifetime income will be high and hence their willingness to use credit (and their spending) will also be high.

Conversely, consumers who are granted lower amounts of credit are likely to infer that their lifetime income will be low and hence their spending will be lower. However, based on research in the area of consumer skepticism and inference making, we also argue for a moderating role of the credibility associated with the credit limit. Specifically, we argue that the above effect of credit availability would be particularly strong for consumers who believe that the credit limit credibly signals their future earnings potential (i.e., a naı¨ve consumer who has limited experience with consumer credit). However, as consumers gain experience with credit, they start discounting credit availability as a predictor of their future and start questioning the validity of the process used to set the credit limit.
Hence, with experience the effect of credit limit on the willingness to use credit should be attenuated. We test these predictions in five separate studies. In the first experimental study, we manipulate credit limit and credibility and pose subjects with a hypothetical purchase opportunity. Consistent with our prediction, credit limit impacted the propensity to spend, but only when the credibility was high. In the second experimental study, we replicate these findings even when subjects were given information about their expected future salaries, and also show that the credit limit influences their expectation of future earnings potential.

In the third study, we show that the mere availability (and increase) of current liquidity cannot explain our findings. In the fourth study, we conduct a survey of consumers in which we measure a number of demographic characteristics and also ask them for their propensity to spend in a given purchase situation. In the fifth study we use the Survey of Consumer Finances (SCF) dataset, a triennial survey of U.S. families that is designed to provide detailed information on the use of financial services, spending behaviors, and selected demographic characteristics.
Results from both studies 4 and 5 provide further support for our proposed framework—credit limits influence spending to a greater extent for consumers with lower credibility: younger consumers and less-educated consumers. Across all studies we achieved triangulation by using a variety of approaches (surveys and experiments), subjects types (young students and older consumers), nature of predictor variables (manipulated and measured), dependent measures (purchase likelihood, credit card balance, new charges), and methods of analysis (ANOVA and regression), and consistently found that increasing credit limits on a credit card increases spending, especially when the credibility of the limit is high.

This paper joins a growing body of literature in marketing and behavioral decision theory that goes beyond the traditional domains of inquiry (e.g., product choice, effects of marketing mix variables) and focuses on consumer decisions relating to the appropriate use of income to finance consumption. Our framework differs from prior research on the effect of payment mechanisms on spending in two significant ways. First, we are interested in the effects of the availability of credit on spending, and not necessarily in the effect of the transaction format that is associated with each payment mechanism. Second, while prior research has studied the point-of-purchase and historic (i.e., prepurchase) effects of credit, the present research is concerned with the availability of credit in the future. Specifically, our framework is invariant to the current and prior usage of credit by the consumer. (Consumer Credit; Credit Cards; Intertemporal Choice; Mental Accounting; Self-Control )
Introduction

The provision of consumer financing has become a pervasive tool for many marketers worldwide (Glassman 1996). Simultaneously, consumer debt has soared to record levels and an alarming number of households are finding themselves in financial difficulties (Andelman 1998, Monthly Review 2000). Given these trends in the consumer marketplace, a surprisingly sparse amount of research has attempted to study a consumer’s decision to use credit. Why should consumers use credit to finance purchases? The life-cycle hypothesis (Modigliani and Brumberg 1954) posits that consumers attempt to maintain their lifestyle and consumption baskets over their lifetime even though their income and wealth may fluctuate over time. Specifically, older consumers can borrow from their past savings and consume at levels beyond their current incomes. Conversely, young consumers who expect future incomes to be higher than their present income can ‘‘borrow from their future income’’ to support their present lifestyle. These processes have been referred to as consumption smoothing (Shefrin and Thaler 1988). The availability of credit facilitates consumption smoothing and hence a rational consumer can use credit to intertemporally maximize lifetime utility.

However, an economically rational intertemporal reallocation of income requires a fair degree of cognitive complexity on the part of the consumer, an assumption that has been shown to be unrealistic (cf. Johnson et al. 1987, Kotlikoff et al. 1988). Given this cognitive handicap, how do consumers intertemporally allocate incomes? What factors influence this decision? In the present research, we focus on the decision to borrow from future income. We agree with prior behavioral research that consumers are unable to correctly value their present and future resources, and that they lack the cognitive capability to solve the intertemporal optimization problem required by the life-cycle hypothesis. Furthermore, we argue that consumers use external information such as the availability of credit to infer their future earnings.
Specifically, if consumers have easy access to large amounts of credit, they are likely to infer that their lifetime income is high and hence their willingness to use credit will also be high. Conversely, consumers who are granted lower amounts of credit are likely to infer that their lifetime income will be low, and hence be less likely to use credit. However, we also argue for a moderating role of the credibility associated with the credit limit. Specifically, we argue that the above effect of credit availability would be particularly strong for consumers who believe that the credit limit credibly represents their future earnings potential (such as a naı¨ve consumer who has limited experience with consumer credit).

However, as consumers gain experience with credit, they start discounting credit availability as a predictor of their future income. Hence, with experience the effect of credit limit on the willingness to use credit is attenuated. The rest of this paper is divided into three sections. First, we review relevant research in the area of behavioral decision theory, economics, and marketing and articulate our hypotheses about the effects of credit limits and credibility on the propensity to spend. Second, we describe a series of studies using experiments as well as survey data to test our hypotheses and rule out alternative explanations. Third, we conclude with a general discussion and propose avenues for future research.

The Effect of Payment Mechanisms on Spending

Some research in marketing has studied the effects of using various payment mechanisms on consumers’ spending behavior. Hirschman (1979) and Feinberg (1986) used actual consumer transactions to compare the spending of consumers who paid by credit cards with those who used cash or checks, and found that the former spend more in otherwise identical purchasing situations. Prelec and Simester (2001) conducted an auction in which subjects bid for tickets to a sporting event that were to be purchased by the winner on the following day by using either cash or credit card (random assignment). They replicate the finding that willingness-to-pay is significantly greater in the credit card condition as compared to the cash cannot completely explain these effects. All these articles showed a point-of-purchase effect—namely, the use of a particular payment mechanism at the time of purchasing influences spending behavior. In a recent paper, Soman (2001) took a different approach to studying the effects of payment mechanisms.
He argued that past payments influence spending decisions through their retrospective evaluation (i.e., the ‘‘pain’’ of past payments, Prelec and Loewenstein 1998) but that the past usage of a particular payment mechanism moderated this effect. In particular, payments by cash and check are both memorable and painful, while those by charge cards are more easily forgotten and painless, as a result of which people who predominantly use cards overspend relative to those who use cash or checks (Soman 2001). The present research differs from this prior research in two significant ways. First, we are interested in the effects of the availability of credit on spending, and not necessarily in the effect of the transaction format that is associated with each payment mechanism. The research cited above was more concerned with the effects of the nature of the transaction format (e.g., salience, convenience, whether it involved writing down the final price) while the framework proposed in this paper is expected to be invariant to transaction formats. Second, while the prior research has studied the point-of-purchase and historic (i.e., prepurchase) effects of credit, the present research is concerned with the availability of credit in the future.

Specifically, our framework is invariant to the current and prior usage of credit by the consumer. Despite these differences, one result from Soman (2001) is relevant for the present research. In one experiment, he finds that the propensity to spend increases as a function of the credit limit, specifically that as the credit limit increases, subjects using a credit card report a higher likelihood of making a purchase ceteris paribus. While he notes that this result is orthogonal to the main thesis of his paper and that credit limit was being manipulated only to test for the robustness of his effects (Soman 2001, p. 464), we find a similar suggestion from the popular press (Keenan 1998, Punch 1992). For instance, the realization that ‘‘to the extent you raise the credit line on the account, you can keep the cardholder in the fold and get greater usage’’ (Punch 1992, p. 48) seems to be accepted in the industry.Why does access to greater amounts of consumer credit drive the consumer to spend more? No academic research in marketing has documented this phenomenon (except the one byproduct of Soman’s 2001 article) and none has attempted to explain why it occurs, hence the present research is a first step in this direction.

The Intertemporal Allocation of Income

In the face of a mismatch between their income stream and their desired consumption stream, the life-cycle hypothesis postulates that consumers should allocate their lifetime income over time in order to smooth their consumption (Ando and Modigliani 1963, Modigliani and Bromberg 1954; see also Friedman 1957). More formally, the hypothesis states that under certainty, an individual will choose his consumption spending over his lifetime (up to age d) to maximize a concave utility function U [U  U(C1, C2, . . . , Cd) subject to d j A1  Cj  R   H , s 1 j1 s2 R1 where Cj represents consumption in period j, Rs  1/(1  rs), rs is the interest rate at time s, A1 represents initial assets, and H1 is the present value of human wealth at age 1. This formulation specifies that the cumulative discounted consumption stream over an individual’s lifetime is exactly equal to the sum of the cumulative discounted income stream and all assets. In several empirical tests, support has been found for the simpler implications of the hypothesis that the propensity to spend (a) increases as a function of initial assets, (b) decreases as a function of expected lifetime, and (c) increases as a function of human wealth (cf. Courant et al. 1986, Johnson et al. 1987, Kotlikoff et al. 1988).

Practically, it is easy for consumers to use past income in the future because it can be stored in the form of savings and investments to be used later. However, it is practically impossible for a consumer to use his own future income in the present since the future income has not yet been realized and hence does not physically exist. To do this, a consumer needs to have access to an account that can act as an intertemporal intermediary between the future lender and the present borrower. Consumer credit plays exactly this role—it provides the consumer with additional spending power in the present in exchange for repayment (of the loan and interest) in the future.

Consumption behavior in accordance with the lifecycle model should therefore be practically feasible given the availability of consumer credit. Is it psychologically feasible and observed, however? In order to behave in an intertemporally rational manner, individuals need to correctly value their present and future resources and interest rates, perform complicated net present value calculations to compute their lifetime income, and apportion its appropriately interest- adjusted value throughout their lifetime. Basically, this involves the setting up and solution of a complex optimization problem that clearly requires a significant degree of cognitive capability and willingness to participate in this complex processing.
Consequently, research attempting to test the intertemporal optimization aspect of the hypothesis has met with limited success. In an experiment in which subjects were asked to make preferred consumption choices under hypothetical life-cycle economic conditions, Johnson et al. (1987) found that individuals repeatedly made substantial computational errors (see also Kotlikoff et al. 1988). Their experiment led them to conclude that their results ‘‘raise serious questions about the life-cycle model’s ability to describe consumption choice’’ (Johnson et al. 1987, p. 42). In a summary paper, Courant et al. (1986) similarly conclude that despite its elegance and rationality ‘‘the life-cycle model has not tested very well’’ (p. 279). An additional complication in the real world is the fact that the future is never certain for most individuals. Therefore, prior to setting up the optimization problem, the individual also needs to compute alternative income scenarios, assign probabilities to each scenario and compute a most likely income stream.

Thus, while it appears that consumers may have the right intuitions about intertemporal allocations (Shefrin and Thaler 1988), the life-cycle hypothesis is not an accurate descriptor of their behavior. In the midst of a number of seemingly ad hoc modifications proposed to account for anomalous data, a psychologically enriched version of the life-cycle model was proposed by Shefrin and Thaler (1988). The behavioral life-cycle hypothesis highlights the role of one of the individual characteristics described in Fisher’s (1930) Theory of Interest: self-control. Given that the future is uncertain and that consumers do not have the cognitive apparatus to perform the complex optimization problem, it becomes important for consumers to ensure that they do not excessively borrow from the future. Two observations make this an especially difficult problem.
First, prior research shows that immediate consumption always seems to be an attractive alternative to future consumption (cf. Ainslie and Haslam 1992, Benzion et al. 1989, Kirby 1997). Second, in the absence of certainty, ‘‘consumers are prone to magnify the importance of goods in the present and at the same time, they underestimate the importance of cash in the future’’ (Zelenak 1999, p. 154). How do consumers exert self-control? Psychologists have written about the fact that exerting selfcontrol is an effortful process and requires the individual to construct and adhere to a set of well-defined rules or precommitment devices (Elster 1979, Schelling 1984). In the case of spending, individuals can construct a number of different types of rules to constrain expenditure. For instance, individuals can use a ‘‘matching rule’’ in which the consumption in any period t is constrained by the income available in t.

Similarly, consumers may adopt a rule whereby they allow themselves to have credit balances up to a specified limit (Weber and Booream 1996). For instance, the web page for Mastercard tells consumers that it is important to budget wisely, but also reminds them that ‘‘You may use credit cards to pay for many of your expenses. They allow you to have the flexibility of paying credit card bills over time.’’ (Mastercard-A 2000). This website further recommends that outstanding credit balances should not exceed 20% of total income (Mastercard-B 2000). One method of exerting self-control is to use mental accounts (Heath and Soll 1996, Shefrin and Thaler 1988, Thaler 1985, 1999).

Budgets ensure that consumers do not overspend on tempting products that they may otherwise want to buy (but that they should not). The reason that such precommitment devices are effective in exerting self-control is that people do not arbitrarily choose tempting alternatives just because they are attractive (Hsee 1995). Specifically, precommitment changes the incentives under which people will make their future choice such that giving in to temptation will be costly (cf. Wertenbroch 1998). However, recent research also shows that if there is any degree of ambiguity in the structure of the mental accounts, consumers can exploit this to justify choosing a tempting course of action (Soman and Gourville 2001, Soman and Cheema 2001).
For instance, Soman and Cheema (2001) showed that subjects were willing to use an unbudgeted windfall gain to purchase a tempting product by using the windfall money to augment the spending budget. In the present research, we propose that a line of available credit plays a role similar to the windfall gain in Soman and Cheema’s (2001) experiments.

The Credit Limit, Consumer Inferences, and Skepticism

Previous research suggests that consumers understand the intuition behind intertemporal allocation of income but do not have the cognitive sophistication to follow the life-cycle hypothesis predictions. Furthermore, consumers employ self-controlmechanisms (like mental accounts), but if they can find rational ways of succumbing to tempting purchases, they will do so. And consumers seem to use their own personal rules of thumb—heuristics—to determine the extent to which they can utilize available credit. In the present section, we suggest that consumers may also use the size of the available credit (i.e., the credit limit) as a heuristic in determining their future income potential and, consequently, their propensity to use credit.

Specifically, we propose that consumers with access to larger amounts of credit will infer that they will have a larger future income and will display a larger propensity to spend than consumers with smaller credit limits. The rationale behind this expectation is relatively straightforward. The credit limit in most consumer credit situations (e.g., credit cards, overdraft loans) is exogenously set by the lender. Ideally, the lender should utilize all available information to determine the future earnings potential of the consumer (Garcia 1980), and hence we speculate that in the absence of all contrary evidence, consumers do expect the credit limit to represent the future well.
Norton (1993) notes that credit is seen as a credible source of alternate income—typically by younger consumers and novice users of credit—and that such consumers tend to integrate available credit with their spendable income. We use the term ‘‘credibility of the limit’’ (or simply ‘‘credibility’’) to represent the degree to which consumers expect their credit limit to truly capture their future earnings potential. Since borrowing from the future is rational if the credit amount represents a true reflection of future income potential, we argue that consumers will readily integrate the available credit with their ‘‘spendingmoney’’ if the credibility is high. Consequently, their propensity to spend will be greater (Ferber 1973, Tobin 1972) with larger credit limits conditional on high credibility. We note that the credibility of the limit has two antecedents.

The first antecedent is the credibility of the process used to set the credit limit. As discussed above, a lender should utilize all available information to determine the future earnings potential of the consumer (Garcia 1980). A search of the banking literature as well as an interview with a consumer loan officer in a credit union revealed, however, that there exist no formal guidelines for banks to use in setting credit limits. In fact, banks and other lenders typically use histories of past credit as inputs (Tressler 1998) and there seems to be no explicit use of future income projections in making credit-limit decisions. 1 We thank an anonymous reviewer for stimulating our thinking in this direction.
While it is reasonable and intuitive to expect that past earnings and credit usage might be a good predictor of the future, this expectation is neither consistent with the life-cycle hypothesis nor has been empirically validated by prior research. A second antecedent of the credibility of the limit is based on the idea that the credit limit is a signal of future earnings potential and that differences in the perceived meaningfulness of this signal result in differences in the credibility of the limit.

Note that this second antecedent of credibility is much broader in scope than the issue of whether the process of setting the limit is credible. In many cases, credibility can be low due to both a low credibility associated with the process used to determine the credit limit as well as a low meaningfulness associated with the signal. For instance, a consumer who is told that his bank considers ‘‘current wages and other credit indices in a simple formula that provides the credit limit’’ and that ‘‘this method is fast and convenient but probably not very scientific’’ could assign a low credibility for two reasons. First, he may cast doubts on the process used to determine credit limits and question whether it uses the appropriate variables and the right formulas.

Second, he may simply believe that since the limit is calculated in a nonscientific manner, it is not a diagnostic signal of future earnings potential. In the present research, we do not aim to distinguish between these two antecedents of credibility. Our claim is simply that the credibility associated with the credit limit moderates the effect of the credit limit on the propensity to spend. As a routine practice, many credit card issuers freely use credit limit as a tactical marketing tool—credit limits on most accounts have been steadily ratcheted up (Lunt 1992), and when credit card users with good payment records reach their limit, their credit line is typically increased (Punch 1992).

As a result, the presumed rationality of the credit-limit-setting process is called into question. What implications does this have for consumer beliefs about the credibility? As consumers become more experienced with credit cards and start observing seeming irrationalities like frequent and rapid changes in their credit limits and large discrepancies between the credit limits offered by different issuers, their beliefs about the credibility will reduce. Prior research in two streams is relevant here.
First, research in the area of expertise shows that experienced individuals examine information at a deeper level (and not superficially), seek explanations for presented data and are more likely to question data that is presented to them (Chi et al. 1988). In this context, this suggests that experienced consumers are more likely to question the validity of the credit limit. Second, research in the area of consumer skepticism suggest that consumers are skeptical of claims made by marketers (e.g., advertising claims, Ford et al. 1990; or pricing, Licata et al. 1998) and, furthermore, that experience and age seem to increase the degree of skepticism (cf. Licata et al. 1998, Mangleburg and Bristol 1998).

Again, experienced consumers seem to be more likely to discount the presented information. Both streams of literature therefore suggest that for experienced consumers, the credibility will be low as compared to that of novice consumers. In a related vein, experienced consumers are likely to be exposed to a greater number of credit card offers with greater variance and, hence, do not believe in the diagnosticity of credit limits any more. Consequently, as experienced consumers rely less on the available credit as a heuristic to determine their future earning potential and are less likely to integrate it with their ‘‘spending money,’’ the effects of credit limit on the willingness to spend will be attenuated. Our expectations are captured more formally in the following hypotheses: HYPOTHESIS 1 (H1). The propensity to spend when credit is available will depend on the size of the credit limit available to the consumer, but only when the credibility of the limit is high. When the credibility is low, the effect of credit limit will be weaker. HYPOTHESIS 2 (H2). The credibility of limit will be lower for experienced consumers as compared to relatively novice users of credit. We next describe a series of five studies designed to test these hypotheses.