Special Session Summary Postpurchase, Intertemporal Consumption Behavior

Citation:

Adwait Khare and J. Jeffrey Inman (2003) ,"Special Session Summary Postpurchase, Intertemporal Consumption Behavior", in NA - Advances in Consumer Research Volume 30, eds. Punam Anand Keller and Dennis W. Rook, Valdosta, GA : Association for Consumer Research, Pages: 255-257.

Advances in Consumer Research Volume 30, 2003     Pages 255-257

SPECIAL SESSION SUMMARY

POSTPURCHASE, INTERTEMPORAL CONSUMPTION BEHAVIOR

Adwait Khare, University of Pittsburgh

J. Jeffrey Inman, University of Pittsburgh

SESSION OVERVIEW

The authors in this session report research that seeks to understand consumption behavior. In the past, such research has been constrained due to a lack of adequate data. The authors discussed how postpurchase consumption behavior is shaped by factors such as convenience and salience (Chandon and Wansink), habit (Khare and Inman), and socio-demographic trends and attitudes (Beres).

Chandon and Wansink examine when and how, exogenous product stockpiling increases the short-term rate of consumption. They address the following questions: 1) Are exogenously-stockpiled products really consumed at a faster rate? 2) How does stockpiling increase consumption rates for high vs. low convenience products? The questions are answered through multiple methods (scanner data analysis and field/laboratory experiments) and across product categories. Chandon and Wansink’s findings show that stockpiling does indeed cause people to consume products at a faster rate. Stockpiling increases the consumption rate more for high convenience products than for low convenience products. This is because, while stockpiling increases consumption quantity (given incidence) for both types of products, it only increases consumption incidence for high convenience products. Further, stockpiling only triggers consumption incidence when it increases the salience of the product at the point of consumption. Such increased salience is more likely to lead to consideration when the product is convenient to consume compared to when it requires preparation.

Khare and Inman argue and demonstrate that habitsBconsistency in behavior over timeBplay a dominant role in repetitive consumption behaviors such as food consumption behavior (Aarts and Dijksterhuis 2000; Ouellette and Wood 1998). They study habits in the consumption of various food-nutrients and discuss two types of habitsBcarryover and baselineBand show that meal-type (breakfast, lunch, dinner) and nutrient-type (positive, negative) moderate the nature of these habits. Carryover habit is observed when, within a meal-type, the level of a nutrient in a past meal significantly predicts the level of the same nutrient in the current meal. Baseline habit is observed when the consumption level of a nutrient is habitually different across meal types. Based on food consumption data from a diary-panel of individuals, Khare and Inman 1) show support for carryover and baseline habits 2) show that carryover habit is stronger at breakfast than at dinner 3) demonstrate that individuals’ baseline habits are such that they obtain positive nutrients mostly at breakfast and negative nutrients mostly at dinner.

Beres reports findings from The NPD Group’s Annual Report on Eating Patterns in America. These findings are based on the dietary consumption information collected from a panel of 2500 households over a two-week period staggered throughout the year. The data includes information on product usage (e.g., base dish, ingredient), preparation method (e.g., stove, microwave), meal occasion (e.g., breakfast, lunch, dinner), location (in-home versus away-from-home), seasonality (e.g., day of week, holiday), user (e.g., age, diet status, lifecycle), etc. These consumption data are supplemented by self-reports on nutritional attitudes and behaviors. Beres discussed findings regarding trends in Americans’ concern with maintaining a healthy diet (e.g., avoiding harmful substances such as fat and cholesterol), trends in tolerance for extra weight, and changes in meal preparation methods.

At the end of the three presentations, the session’s discussant, Valerie Folkes underscored the need to study actual consumption behavior and along with the audience discussed important issues related to food consumption behavior. Some of these issues were: increasing serving sizes of packaged as well as served foods, time-of-the-day effects in food consumption behavior, and possible changes in nutritional labeling.

 

EXTENDED ABSTRACT - S

DO WE CONSUME STOCKPILED PRODUCTS FASTER? A CONVENIENCE-SALIENCE FRAMEWORK OF STOCKPILING-INDUCED CONSUMPTION

Pierre Chandon, INSEAD

Brian Wansink, University of Illinois at Urbana-Champaign

In this research, we develop a framework of postpurchase consumption behavior, which we use to examine when, and how, exogenous product stockpiling increases the short-term rate of consumption. We ask two questions: 1) Are exogenously-stockpiled products really consumed at a faster rate? 2) How does stockpiling increase consumption rates for high convenience and low convenience products? We examine the first question with an analysis of household scanner data and by conducting a field experiment. We examine the second question with field and laboratory studies, with families and young adults, with multi-category or single-category choices, with retrospective and prospective measures of consumption, and with a total of 20 different products.

Findings from the scanner data and the field experiment show that stockpiling does indeed cause people to consume products at a faster rate. The association between stockpiling and consumption that is shown in previous scanner data analyses (e.g., Ailawadi ad Neslin 1998) is therefore not simply caused by consumers endogenously stockpiling in anticipation of a higher demand for the product. Studies 1 and 2 also show that stockpiling increases consumption rate more for high convenience products than for low convenience products. The reason is that, although stockpiling increases consumption quantity (given incidence) for both types of products, it only increases consumption incidence for high convenience products. Study 3 shows that this impact of stockpiling on consumption incidence is mediated by salience. Stockpiling only triggers consumption incidence when it increases the salience of the product at the point of consumption. Finally, Study 4 shows that convenience moderates the effects of salienceCand hence of stockpilingCon consumption incidence. This is because the higher awareness created by stockpiling and salience is more likely to lead to consideration when the product is convenient to consume compared to when it requires preparation.

 

HABIT REGIMES IN CONSUMPTION

Adwait Khare, University of Pittsburgh

J. Jeffrey Inman, University of Pittsburgh

We argue and demonstrate that habits play a dominant role in repetitive consumption behaviors such as food consumption behavior (Aarts and Dijksterhuis 2000; Ouellette and Wood 1998). Habits are reflected through consistency in behavior over time. We study habits in the consumption of various food-nutrients (e.g., saturated fat, calcium, etc.). In this context, we discuss two types of habitsBcarryover and baselineBand show that meal-type (breakfast, lunch, dinner) and nutrient-type (positive, negative) moderate the nature of these habits.

Carryover habit is observed when the level of a nutrient in a past meal is a significant predictor of the level of the same nutrient in the current meal. Carryover habits reflect within meal-type behaviors. For example, if the level of fat in yesterday’s lunch predicts the level of fat in today’s lunch well then carryover habit with respect to fat is observed. We argue that carryover habit can take one of two forms: fixed vs. cycling. When carryover habit is significant and has a positive valence (i.e., the habit coefficient is significant and positive), we categorize this type of carryover habit as fixed carryover habit (e.g., eating the same breakfast cereal everyday). And, when carryover habit is significant and has a negative valence (i.e., the habit coefficient is significant and negative), we categorize this type of carryover habit as cycling carryover habit (e.g., alternating between three breakfast cereals in a fixed order) (c.f., Brickman and D’Amato 1975).

Baseline habit is observed when the consumption level of a nutrient is consistently different across meal types. For example, if a person consistently consumes lower levels of protein at breakfast but higher levels of protein at dinner then baseline habit with respect to protein will be observed i.e., the person’s protein consumption baselines differ across meal-type.

Given the repetitiousness and situational stability of food consumption behavior, habits are very likely to develop in food consumption behavior. Habits can develop for a variety of reasons. For example, with respect to food consumption, observed eating behavior could be so because consumers deliberately planned it that way (Connors et al. 2001) or because consumers have been conditioned to do so (Bargh and Chartrand 1999). It should be noted that irrespective of what the bases of habits are, once developed, all habits operate automatically i.e., at a lower level of cognitive processing. We argue that habits are beneficial because they help consumers manage their utility over time (Kahn, Ratner, and Kahneman 1997) and because they help consumers minimize effort in decision-making.

We hypothesize and based on a diary-panel of individuals (n=769) show that:

$ A majority (over 70%) of individuals exhibit carryover habit in food consumption behavior.

$ The magnitude of fixed carryover habit is stronger at breakfast than at dinner.

$ A majority of individuals (over 75%) exhibit baseline habit in food consumption behavior.

$ For positive nutrients (e.g., calcium) the nature of baseline habit is such that for a majority of the individuals the level consumed at breakfast is the highest.

$ For negative nutrients (e.g., saturated fat) the nature of baseline habit is such that for a majority of the individuals the level consumed at dinner is the highest.

These findings have important implications for public-policy making and managerial decision-making. For example: 1) Based on the observed asymmetry between positive and negative nutrients, perhaps dietary guidelines for positive nutrients could focus on breakfast foods and those for negative nutrients could focus on dinner foods. Such targeting could increase the effectiveness of dietary guidelines 2) The asymmetry between positive and negative nutrients has implications for designing nutritional descriptors of foods. Nutritional descriptors for breakfast foods could focus on accentuating the positives (e.g., high calcium) and nutritional descriptors for dinner foods could focus on downplaying the negatives (e.g., low in saturated fat).

 

EATING TRENDS IN AMERICA

Cindy Beres, The NPD Group, Inc.

The NPD Group in 1980, as part of its National Eating Trends service, began to collect complete intake information on the eating patterns of America. This ongoing research effort involves thousands of individuals maintaining a daily diary of all the foods and beverages they consume over a two-week period. Our sample of households is a rolling sample. That is, on the Monday of every week in a year, about fifty households start recording their dietary intake information for a period of two-weeks. At any given point of time, a group of fifty households is in the first week of their participation period and another group of fifty households is in their second week of participation. The 2500 households we sample every year are representative of the Census Bureau’s demographic distribution of the household population. Participating households are compensated for their cooperation. The daily diaries completed by the participating households contain sections for each main meal occasion, carried meals, snack occasions, as well as meals eaten away-from-home. The collected data include information on product usage (e.g., base dish, ingredient, side dish), preparation method (e.g., stove, microwave), meal occasion (e.g., breakfast, lunch, snack), location (in-home versus away), seasonality (e.g., day of week, season of year, holiday), user (e.g., age, diet status, lifecycle, ethnicity), etc. We then calculate the nutritional ingredients of each food item eaten. Our sampled households’ head individuals also report their nutritional attitudes and behaviors. These data enable us to address issues involving demographic dynamics, health related behavior and attitudes, meal preparationBwho does it, how it is done, etc., and food related trends such as which foods are eaten when, consumption of which foods is increasing (decreasing), etc.

A sampling of some of the findings in our most recent analysis include: (Health) Ours being an "aging" nation, health will remain an important issue. However, with regards to food, health is not the front runner that it once was. Americans are less watchful when it comes to avoiding harmful substances such as fat, cholesterol, and sodium. Health issues have reached a point where Americans are finally able to confess that the most important things about food are its taste, looks, and smell. (Weight) Americans appear to be more tolerant of extra weight than ever before. This tolerance is not odd given that more people are overweight than before. However, there remains, as always, a core group of dieters consisting of approximately a third of adult females and a fifth of adult males. (Meal preparation) Changes in what Americans eat seem less to do with food and more to do with preparation. Nearly half of the food dollar now goes to foodservice outlets rather than supermarkets. Restaurants have become "prepared supermarkets" for today’s households as take-out continues to be a driving force in that industry. Still, most meals are prepared in-home. The in-home meal is undergoing changes too. In-home meals now have more frozen main dishes, fewer homemade dishes, and fewer side dishes.

These trends are among those that we will discuss, along with concomitant implications. However, despite the changes in food consumption, on average, our eating habits evolve quite slowly. For example, the foods consumed ten years ago are, for the most part, the same as foods consumed today.

REFERENCES

Aarts, Henk and Ap Dijksterhuis (2000), "Habits as Knowledge Structures: Automaticity in Goal-Directed Behavior," Journal of Personality and Social Psychology, 78 (1), 53-63.

Ailawadi, Kusum L. and Scott A. Neslin (1998), "The Effect of Promotion on Consumption: Buying More and Consuming it Faster," Journal of Marketing Research, 35 (3), 390-398.

Bargh, John A. and Tanya Chartrand (1999), "The Unbearable Automaticity of Being," American Psychologist, 54 (7), 462-479.

Connors, M., C. A. Bisogni, J. Sobal, and C. M. Devine (2001), "Managing values in personal food systems," Appetite, 36 (3), 189-200.

Kahn, Barbara E., Rebecca K. Ratner, Daniel Kahneman (1997), "Patterns of Hedonic Consumption Over Time," Marketing Letters, 8 (1), 85-96.

Ouellette, Judith A. and Wendy Wood (1998), "Habit and Intention in Everyday Life: The Multiple Processes by Which Past Behavior Predicts Future Behavior," Psychological Bulletin, 124 (1), 54-74.

----------------------------------------