An empirical investigation into the excessive-choice effect.

Over the past two decades, there has been an explosive growth in the field of behavioral economics, which focuses on the interface between economics and psychology. Behavioral economics has contributed insights related to loss aversion, the endowment effect, bounded rationality, and anchoring, to name a few. One psychological finding that is highly relevant to the practice of economics, but that has received little recent attention by economists, is the excessive-choice effect (ECE). This article describes the ECE and investigates the presence and magnitude of the effect in four experiments.

The standard economic model of the consumer assumes that utility is weakly increasing in choice-set size. If more choices are available, consumers can reconsider their purchasing behavior, either by discovering a consumption bundle yielding higher utility or remaining at the old bundle and utility level. This standard model stems from the assumption of perfect rationality, perfect knowledge, and costless cognitive processing. The standard model is periodically questioned by empirical observation, inducing researchers to seek bounded rationality models that explain the empirical puzzle.

This study concerns a new empirical puzzle. A number of psychology and marketing studies have demonstrated that more choice is not always better for consumers. Greater product variety can dampen overall product demand and even lower the utility experienced from consumption of the chosen good (Iyengar and Lepper 2000; Schwartz et al. 2002). This phenomenon, where increasing product variety lowers overall product demand, is termed here an excessive-choice effect. The ECE does not imply that the standard model is not useful or a good approximation to reality, but it does imply that real behavior must sometimes be portrayed using a bounded rationality framework.

The existence of an ECE has important implications for the practice of economics. It implies that firms that increase variety to attract consumers with differing tastes could actually lead to lower overall sales, a fact that is common knowledge among many business people but not necessarily economists. Consider Harvey Mackay, who is a CEO and best-selling author on business books. When reminiscing on the lessons learned from his first job, he remembers his boss telling him, "Never put more than three ties on the counter. It will only confuse the customer" (Mackay 2008). The presumption that more choice is better has led many agricultural economists to encourage product diversity through niche marketing strategies, certification programs, and labeling policies. While the ECE does not suggest such product diversity is necessarily bad, it does suggest that increased product variety can harm consumers, a possibility rarely considered by economists (except when variety is related to market power, which has been studied extensively by economists).

The possibility of an ECE also has implications for the methods used in studies that estimate the value of new products (e.g., Dahr and Foltz 2005) or the welfare gains/loss from adding/closing recreational sites (Phaneuf, Kling, and Herriges 2000). Standard models used in such applications impose the restriction that an individual cannot be made better off by eliminating a choice option. This assumption, however, contradicts many of the empirical studies (Dreze, Hoch, and Purk 1994; Broniarczyk, Hoyer, and McAlister 1998; Iyengar and Lepper 2000; Boatwright and Nunes 2001; Schwartz et al. 2002) that show that individuals can, in some cases, be made better off by reducing choice options.

Psychologists have even hypothesized that the consequences of the ECE go far beyond simple shopping decisions. Barry Schwartz, a leading psychologist in the field of choice, suggests that successful capitalistic societies produce too much consumer choice--with terrible outcomes. For example, Schwartz argues, "As the gross domestic product more than doubled in the past 30 years, the proportion of the population describing itself as 'very happy' declined by about 5 percent, or by some 14 million people ... Of course, no one believes that a single factor explains decreased well-being, but a number of findings indicate that the explosion of choice plays an important role" (2004, p. 71). Despite the clear relevance of the ECE for economic thought, the authors know of only two previous economic journal articles on this topic, both of which are conceptual in nature (Norwood 2006; Irons and Hepburn 2007).

The objective of this article is to provide an empirical investigation of the ECE in four experiments. The major questions of interest are: would consumers voluntarily reduce their choice-set size; can they replicate the findings of previous psychological studies in experiments; is the effect robust to changes in the design of previous studies; and does the presence of the effect depend on individuals' personalities as hypothesized by Schwartz et al. (2002) and Schwartz (2004). The next section describes the ECE more fully, which is followed by a section placing the effect in the context of bounded rationality. After presenting the research objectives, this article contains four sections, one for each experiment. The last section concludes.

The Excessive-Choice Effect

Iyengar and Lepper (2000) were the first to popularize the ECE. In three separate experiments, they documented cases where individuals were less likely to choose from a large than a small choice set (e.g., twenty-four vs. six choices). For example, they found that a larger percentage of consumers purchased specialty jams and chocolates when offered a small number of varieties (i.e., six) as opposed to a larger number (i.e., twenty-four or thirty). They also found that students were more likely to complete an extra credit essay assignment when given six essay topics as opposed to thirty. The ECE has also been found in studies of supermarket sales (Dreze, Hoch, and Purk 1994; Broniarczyk, Hoyer, and McAlister 1998; Boatwright and Nunes 2001), where increasing product variety of a general good on supermarket shelves either does not change or reduces total sales volume of that general good.

Two explanations for the ECE have been offered: search costs and regret avoidance. Norwood (2006) demonstrated how search costs (where utility is penalized for each additional item in a choice set considered) in conjunction with heterogeneous preferences can lead to the ECE. The idea behind the model is that when many options are presented to consumers, they may randomly sample from among all available options when making a choice. Due to search costs, consumers may only peruse a subset of the total number of varieties. As is often the case in markets, if each additional variety that is introduced appeals to a smaller and smaller share of the population, increasing variety can be harmful to the majority of consumers, as it reduces the chance that their most preferred variety will be present in the subset of varieties that they peruse.

Schwartz (2004) argues that the ECE is the result of a particular personality type referred to as maximizers. Maximizers peruse all varieties within a choice set carefully, insisting on choosing the one best item. Satisficers, the opposite of maximizers, are content with a good choice even if it may not be the best. When presented with a large number of varieties, a satisficer is more likely to purchase the first item whose utility crosses a certain threshold, while maximizers are more likely to consider each option and deliberate carefully on which one item is best. Satisficers are content with a good option, while maximizers insist on the very best.

Individuals' propensity toward having a maximizer or satisficer personality is measured using a self-administered questionnaire developed by Schwartz et al. (2002) and Schwartz (2004). Responses are used to construct a scale indicating one's personality on the satisficer--maximizer spectrum. Maximizers can experience profound regret with their purchases, always wondering if an alternative item would produce greater utility. As individuals are faced with numerous choices on a daily basis, the accumulation of this regret is such that maximizers are less happy in life and more likely to experience depression (Schwartz et al. 2002). When maximizers are presented with a large choice set, they may anticipate the regret associated with such choices. To avoid this regret, they simply abstain from making the choice. While the direct utility from consuming a good may be positive, the negative utility from regret is greater. The net utility of the choice is negative, and no purchase is made.

Irons and Hepburn (2007) formalize this argument by invoking regret theory in the construction of a consumer decision-making model. Their model provides a mathematical structure to Schwartz's (2004) arguments. Even when a choice appears to maximize expected utility ex ante, consumers may avoid the choice if it may appear ex post to not maximize utility. By incorporating regret into a search model, they show that greater variety can reduce the probability of a purchase. According to Schwartz (2004), whether regret enters into utility, as in the Irons and Hepburn article, depends upon whether the consumer is a maximizer. Whether the ECE is caused by search costs, regret avoidance by maximizers, or both, is currently unknown. This article tests whether maximizers are more likely to exhibit the ECE than satisficers, thereby testing the hypotheses offered by Schwartz (2004) and Irons and Hepburn (2007).

While the ECE is a relatively new phenomenon, the idea of bounded rationality behavior is not. Moreover, while the ECE may be inconsistent with the "standard" economic theory of consumer behavior, economists are historically quick to formulate new theories to accommodate such behavior. To better articulate the relationship between the ECE and economic theory, the following section is provided.

Print Get the Mag Weekly Updates