Special Session Summary Do We Use More When We Buy More? the Effects of Stockpiling on Product Consumption


Brian Wansink (1998) ,"Special Session Summary Do We Use More When We Buy More? the Effects of Stockpiling on Product Consumption", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 21-22.

Advances in Consumer Research Volume 25, 1998      Pages 21-22




Brian Wansink, University of Illinois at Urbana-Champaign

We know surprisingly little about what happens to packaged goods after they are carried from the store. This is because much of the research in our field has focused on why consumers choose products and not on why they use them. This session investigated the important interrelation between purchase (choosing) and consumption (using) by focusing on what happens to products after they are brought home from the store. This topic is of interest to a growing number of scholars in marketing, economics, nutrition, and consumer science. Moreover, it has become an issue of recent interest among researchers and managers of packaged goods companies such as Nabisco and Kellogg.

Because of the emerging nature of this research, there are two important objectives: (1) build an integrated framework of how inventory levels influence consumption, and (2) determine the most fruitful areas for future research. The session began with research that addressed the question of "When and Why Promotional Stockpiling Accelerates Consumption" (Chandon and Wansink). This work challenges the long-standing assumption that promotion-driven consumer stockpiling only borrows from future sales and cannot create primary demand. This paper proposes an integrative framework of consumption decision making which suggests that promotional stockpiling can influence the consumption of substitutable and impulse products by raising their salience at the point of consumption, by decreasing their replacement costs, and by increasing their holding costs.

Study 1 reports a field study in which both diary panels and recall data show that household stockpiling accelerated the consumption of six product categories that were distributed to 56 households. Consistent with the framework, this consumption acceleration was mediated by product inventory salience. These results are further supported by a scanner data study (study 2) which shows that the stockpiling of promotion packs accelerates the purchase of usage variant products (cookies and fruit juice) but not that of usage invariant products (laundry detergent). In study 3, these psychological mechanisms behind consumption are then further explored with a laboratory study. These three studies provide three different methods that support the hypothesized framework.

This general model emphasizes that high household inventory levels of a product can increase consumption rates. The paper, "Can You Eat Just One? Stockpiling and Consumption Self-control ," (Wansink and Wertenbroch) investigates circumstances where the opposite may be true. Building on the self-control literature, a model is developed which shows that high inventory levels only accelerate product usage among individuals with low toaverage levels of willpower (self-control). When individuals have high levels of self-control, these same high levels of inventory actually decelerate usage. The intuition behind this is the "flood gate" effect. With small amounts of inventory, the inventory itself serves as a self-governing restricter of consumption. With large inventory levels, however, an individual with high willpower knows that he or she cannot begin consuming this stockpile of food (or a large bag of chips) without risking a break in the flood gates of self control. A lab study and a field study are conducted to investigate this relationship and determine the categories where the effect is greatest.

Stockpiling can lead to accelerated consumption among a large segment of the population. But why do consumers stockpile to begin with? Joseph Nunes believes that consumers may purchase too much because they cannot accurately predict their consumption. In "Predicting Product Usage: Buying Too Much, Using Too Little!" he argues that when comparing competing package sizes, consumers must gauge how frequently they will use a product in order to choose the appropriate package size. The paper explores how consumers forecast product usage and consumption frequency and how they incorporate this into their purchase decision.

The session concluded with observations and insights from Brian Zaff (Ph.D. Ohio State) who is Director for Sensory Research at Nabisco. Recent in-house research conducted by Dr. Zaff and his team will be discussed and related to these findings on stockpiling.



Pierre Chandon and Brian Wansink

Does household stockpiling influence consumption? Though deceptively simple, this question has lead to disagreements among retailers, conflicting assumptions among researchers, and inconsistent promotion strategies among manufacturers. At the center of these differences is the issue of whether sales promotions only influence purchase decisions or whether they can influence also consumption habits. Although some researchers have challenged the assumption of constant consumption rates after a promotional stockpiling (Blattberg and Neslin 1990) there has been little empirical evidence to support their view. One reason is because scanner data does not include any measure of household inventory or consumption. Another more general explanation is that we still have a very limited understanding of how consumers make consumption decisions.

This research extends normative models of consumption behavior (Assuncao and Meyer 1993) by proposing a framework of how consumers decide what and how much to consume. This framework explicitly recognizes the two stages of consumption decision-making: (1) a consideration stage in which consumers decide which products to consider consuming, and (2) a choice stage in which consumers choose which product to consume among the alternatives in their consideration set. Importantly, the inputs to each stage are different. Product inventory salience can trigger need recognition and product consideration, while price and holding costs become relevant only at the stage when a product is being considered for consumption.

A preliminary survey of product characteristics enables us to make predictions as to which categories are more likely to exhibit consumption acceleration. Study 1 reports the research results of a field experience in which 56 households were provided with different levels of stockpiled products. After monitoring consumption for 10 days, it was shown that those who had stockpiled amounts had higher daily consumption rates than those who did not. A mediation analysis shows that stockpiled products were consumed at a faster pace primarily because they were highly salient.

Study 2 uses scanner data to infer the average household consumption period for three products. Consistent with our model, the results show that households increase their consumption frequency of fruit juice and cokies when they buy them in a promotional pack (i.e., larger-than-usual bonus pack) but that buying laundry detergent (a usage invariant product) in big promotional packs has no impact on its weekly usage. Study 3 uses a laboratory study to more closely examine the mechanisms behind the consideration and choice stages.

These studies provide insight into some unresolved issues. First, they reconcile economic models (cost-related models) of consumption behavior with psychological models (salience-related models). Second, they show that under some conditions sales promotions can create primary demand and expand product category sales. Finally, the insights they provide about product category differences can explain some of the inconclusive prior findings (Moore and Winer 1978) as well as the absence of promotion "dips" in some product categories (Neslin and Schneider 1996).



Brian Wansink and Klaus Wertenbroch

High household inventory levels of a product have been shown to increase consumption rates under typical circumstances. We investigate circumstances where the opposite may be true. Building on the self-control literature, we develop a model which shows that high levels of inventory accelerate product usage only among people with low or moderate levels of willpower (self-control). In contrast, these same high inventory levels decelerate usage among individuals with high willpower. The intuition behind this is the "flood gate" effect. With small amounts of inventory, the inventory itself serves as a self-governing restricter of consumption. With large inventory levels, however, an individual with high willpower knows that he or she cannot begin to eat any without risking a break in the flood gates of self control.

A lab study and a field study are described which investigate how one’s level of self-control can lead to decreased consumption as inventory levels rise. The field study examines the usage patterns of consumers with low and high self-control to determine how they are influenced by low and high levels of stockpiling. The lab study further investigates the mediating mechanisms behind this phenomenon.

This paper contributes to our understanding of the individual differences related to household consumption behavior while also contributing to the self-control literature. On the managerial side, these studies suggest that promotions that encourage stockpiling may end up actually deterring consumption among some consumers. This knowledge has implications for targeting and segmentation as well as for package size and bundling decisions.



Joseph C. Nunes

For many goods and services, consumers are increasingly finding themselves able-or even required-to choose between paying one fee for unlimited access or paying for each unit they consume. In this situation, the consumer must evaluate the likelihood that he or she will use enough of the product to make paying the flat fee worthwhile.

Little systematic research has been conducted that explores how people make predictions about their own behavior and no work has been done to date which examines how people make purchase choices requiring predictions of their future usage frequency. This research studies how consumers devise and use their estimation of future usage when choosing between two specific pricing schedules-the flat fee for unlimited access and paying per use. This choice is explicit with numerous on-line and telephone services (e.g., AOL, automatic callback), but buyers have been making this type of decision for years. It occurs when someone decides between buying and renting an infrequently used product, such as ski or scuba diving equipment. It arises again with many membership decisions, such as semi-private health or golf clubs "Do I join, or pay each time I play?").

In order to optimize their allocation of resources the buyer does not want to buy a product they will not use, or to pay for more than he or she will use. A series of lab studies shows how these estimates can be made. The resulting model of prediction is valuable to firms in helping them invest in the correct capacity, market the appropriate sizes, and influence demand when prudent.



Brian Wansink, University of Illinois at Urbana-Champaign


NA - Advances in Consumer Research Volume 25 | 1998

Share Proceeding

Featured papers

See More


Trading Crypto Currency: The Ideological Shaping of Consumer Financial Decision Making

Burcak Ertimur, Fairleigh Dickinson University
Ela Veresiu, York University, Canada
Markus Giesler, York University, Canada

Read More


J8. Exchange with The Rich, Concern with The Poor: The Effects of Social Class on Consumer Response to Brand Relationship

Bing Han, Shanghai Jiao Tong University
Liangyan Wang, Shanghai Jiao Tong University

Read More


Cultural Values and Consumers’ Brand Preference

Jessie J. Wang, Miami University, Ohio
Ashok K Lalwani, Indiana University, USA
Devon DelVecchio, Miami University, Ohio

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.