Printer Friendly

Time as a direct source of utility: the case of price information search for groceries.

Time as a Direct Source of Utility: The Case of Price Information Search for Groceries

Dual earner households face increased demands on their time due to labor market activity. These households may choose to engage less in time intensive household activities such as price information search for grocery items. Evidence indicates, however, that working wives do use price information search strategies (Kaitz 1979; Hacklander 1978; Harris and Stevenson 1983). In order to explain the variation in time spent in price information search by dual earner households, a model is formulated that integrates household production theory, the economics of information, and the assertion that time may yield utility directly. In the model, engaging in search increases both income and satisfaction. The resulting "joint production" model of price information search is an extension of the traditional household production and search models that provides more information and insight into the search behavior of dual earners.

The possibility that enjoyment of time may affect time spent in an activity has been suggested in the literature but has not been formally modeled (Wilkie and Dickson 1980; Dow and Juster 1980). Economic studies that examined the demand for search time, but did not consider the utility value of time, found that increases in the price of a woman's time may increase or decrease time spent in search for grocery items. Search may also be a normal good (Carlson and Geiseke 1983). Although inconclusive, these studies have provided insight into taste and productivity shifters and family characteristics that may affect the amount of time spent in search. These include educational attainment, age of shopper, sex of shopper, and presence of children (Doti and Sharir 1981; Carlson and Geiseke 1983; Harris and Stevenson 1984). This research is summarized in Table 1. (1) In addition, relationships have been found between time use and ownership of durables (Strober and Weinberg 1980; Bryant 1988).


Two frameworks are useful when examining the relationships among increases in dual earner households, enjoyment of search time, and time spent in searching for lower prices. Household production theory outlines how goods and time can be combined to produce commodities that yield satisfaction (Becker 1965). Three categories of time use are leisure, household production, and market work (Gronau 1977). The economics of information asserts that consumers search for lower prices by utilizing time and purchased inputs, such as food advertisements in newspapers, to locate lower food prices (Stigler 1961). Search results in increased purchasing power that is equivalent to an increase in money income. Time is a major input into producing price information, and its effectiveness may vary across households with differential costs of time.

A "joint production" model that uses a household production framework and includes search time as a direct source of utility can be illustrated by imposing weak separability on a utility function and focusing on two home-produced goods: meals and price information. The following Lagrangian function is used to maximize a utility function where home-produced meals, search time, and a composite category of other home-produced goods yield satisfaction:

L = {Go(Xo,Ho;k),Gp(Xp,Hp),Hs;P} + [lambda]{w(H-Hp-Ho-Hs) + v - Po(Xs,Hs;PD)Xo - PpXp - PsXs} (1)


U = total utility, Go = meals produced with goods and time, Xo = direct inputs used producing Go, Ho = time inputs used producing Go, k = a vector of productivity shifters, Gp = other home-produced goods, Xp = direct inputs used producing Gp, Hp = time inputs used producing Gp, Hs = time spent in price information search, P = a vector of preference shifters, w = the market wage rate of the major shopper, H = total available time, v = total nonwage income, Po = price of direct inputs used in meal production, Xs = purchased inputs used in price information search, Pp = price of purchased inputs used in home production, Ps = price of purchased inputs used in price information search, and PD = price dispersion.

The joint production model of price information search differs from the traditional household production and price information search models in two ways. First, time spent in price information search is a source of utility for consumers. Thus, this time use directly enters the utility function. Second, if consumers search for price information, the prices they pay for direct inputs used in home production depend on how much they search. These prices are a function of the information produced. Because information is produced using time and direct inputs, given the market price distribution, it can be considered a home-produced good. The price function for direct inputs for which prices are searched is written

Po = Po(Xs,Hs;PD). (2)

The price paid for purchased inputs into meal production, Po, decreases at a decreasing rate as time or purchased inputs used in information production increase. (2) That is,

[Derivative]Po/[derivative]Xs and [Derivative]Po/[derivative]Hs [is less than] 0. (3)

It is also assumed that

[derivative.sup.2]Po/[derivative][Xs.sup.2] and [derivative.sup.2]Po/[derivative][Hs.sup.2] [is less than] 0. (4)

Although nothing can be said about the relative magnitudes of the second derivatives of the price function, [derivative.sup.2]Po/[derivative][Xs.sup.2] and [derivative.sup.2]Po/[derivative][Hs.sup.2] must be negative. If this were not the case, savings would increase infinitely, and a consumer would exhaust all resources to obtain price reductions on goods.

To use the above model to derive optimal demand for search time, the usual axioms of preferences and concavity of an information/price function that is monotonically increasing are assumed. If an equilibrium is to be reached, the marginal benefits must equal the marginal costs of obtaining the benefits for all first order equations simultaneously. Examination of the first order condition for the demand for search time indicates that

[derivative]U/[derivative]Hs / [lambda] - Xo [derivative]Po / [derivative]Hs = w.

The additional term to consider is the marginal utility obtained directly from time spent in price information search. (3) The first term on the left hand side of equation (5) can also be expressed

([Delta]utils/[Delta]hour of search) / ([Delta]utils/[Delta]dollar) = [Delta]$/[Delta]Hs where $ = dollars.

Equation (6) can be interpreted as the value of the marginal utility of search time when search time yields utility directly. It is literally the change in the value of utility obtained when a consumer engages in an additional unit of search time. Thus, the marginal utility obtained from search time depends on the marginal utility obtained from money.

Maximization of the Lagrangian function in equation (1) implies that the demand for time spent in price information search can be written as a function of endogenous marginal savings of purchased inputs (Xo[[derivative]Po/[derivative]Hs]), the wage rate (w), nonwage income (v), and productivity and preference shifters (k and P). It does not explicitly control for variations in quality found in the marketplace. However, Bellizzi et al. (1981) found that consumers perceive branded items to be of higher quality. These items are usually priced higher than their nonbranded counterparts (Handy 1985). Empirically, inclusion of a measure of perceived quality may control for differences in quality choice by a household.

The following hypotheses are formulated based on maximization of the joint production model of price information search and previous empirical findings.

(1) Enjoyment of time spent in price information search will be positively associated with time spent in the activity.

(2) An increase in the own price of search time will decrease the demand for search time and will depend on both the wage rate and marginal savings available from search.

(3) (A) Age of shopper will impact positively on search time if increases in age are associated with decreases in productivity. (B) Age will impact negatively on search time if increases in age are associated with experience.

(4) Educational level will be negatively related to time spent in price information search if educational level increases efficiency of time spent in an activity.

(5) Presence of young children will impact negatively on the time spent in price information search because young children compete for time spent in other activities.

(6) Presence of teenage children will impact positively on time spent in price information search because older children can cause food expenditures to rise, causing the benefits from search to increase.

(7) Presence of time-saving durables used in meal production will be positively associated with time spent in search because they can free time for use in other activities.

(8) (A) Male shoppers will spend less time in price information search if they are less informed of search strategies. (B) Male shoppers will spend more time in price information search if they are less productive in search.


The model is tested using grocery expenditure data for dual earner households, collected during June through September 1986, from a random sample drawn from Onondaga County, New York. Little doubt exists that substantial savings on groceries may be obtained through price information search. First, imperfections in consumer knowledge can support a distribution of prices in a market (Gastworth 1976; Maynes and Assum 1982; Salop and Stiglitz 1977; Stigler 1961; Telser 1973). Second, food suppliers have responded to continuing increases in female labor supply by increasing the number and price ranges of food products offered for sale. Price discounts used as promotional tools are commonplace (Bellizzi et al. 1981; Gallo 1982a, 1982b). Third, price dispersion can exist because the structure of the food industry cannot be described as perfectly competitive. Some firms may have control over prices (Collins and Preston 1969; Conners et al. 1985; Greig 1976).

Two criteria were necessary for a respondent to be included in the study: (1) that the household contacted have two adults employed in the labor market and (2) that the respondent be the major grocery shopper. All households consisted of two adult part- or full-time workers, with or without children. Considered a population in itself, inferences can be made about this growing market segment. The combined telephone and mail questionnaires yielded 95 complete responses for a response rate of thirty percent.

Given the low response rate, a comparison of the latest statistics available on dual earner households in the United States and the Onondaga County sample was made to identify biases that may result if analyses are performed using an unrepresentative sample (U.S. Bureau of the Census 1986). The Onondaga County sample slightly underrepresents traditionally blue collar occupations and is slightly more educated than the national dual earner average. When broken down into individual earnings categories, the sample is slightly upscale with regard to experience and education. The distribution of children by occupation and age category is comparable to the national average. As a whole, no significant differences in mean earnings for dual earner households was observed. The limitations of using an upscale sample must be taken into consideration when generalizing the results to a larger population.


A summary of the theoretical constructs included in the model, the variables chosen to represent these constructs, and summary statistics appear in Table 2. Variables included in the empirical specification were chosen based on the theoretical framework and availability of data.

Time spent in search was obtained by asking the major shopper how many minutes per week were spent clipping coupons, reading food ads, and shopping for groceries in each store (a maximum of three was mentioned by respondents). The dependent variable, SRCHTIME, was calculated by adding minutes spent in stores other than where the major shopping is done to minutes spent reading weekly food ads and clipping coupons. (4)

The price of a shopper's time (WAGESHOP) is the hourly wage rate received in the labor market. The wage rate is both a component of the price of search time and a measure of the price of other time uses.

Because savings on groceries are a function of search time, it is not possible to include an explicit measure of marginal savings in the estimated equations. Therefore, marginal savings (MARSAVE) must be estimated. All exogenous variables in the system and an additional variable to measure price dispersion (SAVINGS) are regressed on a proxy measure of marginal savings. Marginal savings are approximated by savings obtained per hour of search time. Price dispersion is measured as the amount consumers perceive they can save on weekly groceries if they engage in search during the survey week. Given a one period model, price dispersion is assumed exogenous and equals the perceived distribution before the next search takes place. While consumers as a group may influence the price distribution, no one consumer maintains control over it.

The effect of changes in nonwage income (NWINC) on the demand for search time is measured as all income in a household not earned by the major shopper. This measure includes the wage rate of the nonshopping adult. Use of this measure may cause a pure income effect to be confounded by cross-price effects of the nonshopper's wage on time spent in search. However, no information was available on the other adults' salary. Thus, this effect could not be isolated.

Daily use of a microwave oven (MICRO1) is used as a proxy variable for the level of technology in a household. As a time-saving durable, microwaves can free time used in meal preparation to be used in other activities, including price information search.

It is difficult to identify variables that influence taste and productivity differences in the production of information search. Some theoretical background exists, but there is little empirical evidence that identifies characteristics of shoppers who obtain a high level of savings per time spent in search, or those associated with a "taste" for search (Douglas 1976; Doti and Sharir 1981; Dow and Juster 1980; Tauber 1972). Therefore, a descriptive analysis was performed on two categories of consumers identified from the data.

The first of these analyses was performed on consumers who obtained relatively high levels of savings. The variable MORESAVE was computed by dividing the total value of coupons used for groceries by time spent in search. The resulting savings per hour of search may be considered one measure of the productivity of time spent in price information search. Sixty-six consumers were above the mean score of $7.08, and twenty-nine fell below. The hypothesis that demographic differences may help identify more efficient shoppers was tested by computing the appropriate t-scores and performing tests for significant differences in the means or proportions of the data, depending on how variables were measured. (5) Productivity shifters that were found to be insignificant included educational level, age, and gender of the major shopper. Several significant differences in demographic characteristics were noted. These results are presented in Table 3.

Consumers with relatively high levels of savings per time spent in search come from larger families and have more teenage children. This finding strengthens the use of YNGKIDS and OLDKIDS as productivity shifters. Inclusion of these variables instead of total family size may capture effects due to differences in family composition better. Level of nonwage income may also be an indicator of productivity in the search equation. Weeks worked are not included explicitly in the estimation as inclusion would cause simultaneous equation bias, since in the household production model time and goods are simultaneously allocated.

Because of a lack of evidence as to the appropriate measurement of enjoyment of an activity, a descriptive analysis was also performed on consumers who like to spend time in price information search versus those who do not. Data on enjoyment was obtained by asking respondents whether they regard time spent clipping coupons, time spent reading the food ads, or time spent grocery shopping as work, enjoyment, or a combination of both. Those stating that at least one activity was enjoyed were separated from the rest of the sample, setting up the dichotomous variable LIKESHOP. The hypothesis that no significant differences exist between consumers who like to spend time in price information search and those who do not was tested. Significant results are presented in Table 4.

Results indicate that differences between consumers who like to search versus those who do not are associated with the number of stores shopped at (STORES), reason for choosing a store (CHOOSEST), a proxy for shopping when stores are crowded (QUE), the demographic variable profession of major shopper (PROFS), and a dummy variable indicating whether consumers were raised when convenience foods and money saving coupons were commonplace rather than an oddity (AGEDUM). (6) Consumers who like to search also spend significantly more time clipping coupons (TIMECLIP) and reading food ads (TIMEREAD).

Lack of many significant demographic differences between shoppers who enjoy and do not enjoy search time is not surprising. Enjoyment of an activity is a psychological construct and is not expected to be associated with any specific demographic characteristics. These results also indicate that enjoyment of an activity measures something other than productivity or tastes differences. Thus, a rationale for including a measure of enjoyment in the analysis is formed.

Often consumers' statements about their feelings are not accurate. A better measure of enjoyment may be made by calculating an index using the above results and principal components analysis. Shopping when lines are long (QUE), number of stores shopped at (STORES), reason for choosing a store (CHOOSEST), age of major shopper (AGEDUM), Working in a professional occupation (PROFS), and the direct measure of whether or not a consumer likes searching for information (LIKESHOP) were included in the analysis. TIMECLIP and TIMEREAD were not included, as they make up a portion of the dependent variable, search time, used in subsequent regression analysis. Six factors resulted. The factor loadings presented in Table 5 are the correlations between the factor and the original variable. Factor one appears to be associated with enjoyment. The Eigen value associated with the first factor was 1.55 and accounted for 26 percent of the total variation in the exogenous variables included. These results are fairly consistent with the hypothesis tests in Table 4, which compared shoppers who like price information search and those who do not. Only profession of the major shopper did not load heavily on the factor interpreted as enjoyment of search time. Factor one was titled ENJOY and was the index used in the subsequent analysis.


All respondents in the sample engaged in some price information search. Given this and the endogeneity of marginal savings, two stage least squares is appropriate in estimating the demand for search time. First, an estimate of marginal savings (ESTMARSAVE) was obtained using the equation

MARSAVE = [[gamma].sub.0] + [[gamma].sub.1.SAVINGS] + [[gamma].sub.2.WAGESHOP] + [[gamma].sub.3.NWINC] + [[gamma].sub.4.COLL] + [[gamma].sub.5.AGE] + [[gamma].sub.6.ENJOY] + [[gamma].sub.7.MICRO1] + [[gamma].sub.8.OLDKIDS] + [[gamma].sub.9.YNGKIDS] + [[gamma].sub.10.BRANDED] + [[gamma].sub.11.SHOPPER] + E.

Next, the demand for search time is estimated using the equation

SRCHTIME = [[beta].sub.0] + [[beta].sub.1.WAGESHOP] + [[beta].sub.2.ESTMARSAVE] + [[beta].sub.3.NWINC] + [[beta].sub.4.Coll] + [[beta].sub.5.Age] + [[beta].sub.6.ENJOY] + [[beta].sub.7.MICRO1] + [[beta].sub.8.OLDKIDS] + [[beta].sub.9.YNGKIDS] + [[beta].sub.10.BRANDED] + [[beta].sub.11.SHOPPER] + e.

Because the variable ENJOY is formulated using variables not included in the above estimating equations but that may be correlated with them, equations (10) and (11) were also estimated excluding the variable ENJOY (Belsley, Kuh, and Welsch 1980, ch. 3). Table 6 presents results of the reduced form estimates of marginal savings and structural equation estimates of the demand for search time without ENJOY. Theoretically, excluding ENJOY from the estimated equation does not make sense without also excluding ESTMARSAVE. However, this test helps to determine whether a statistical problem exists. Table 7 presents results with the inclusion of ENJOY. Comparisons of Table 6 and Table 7 indicate that the estimated coefficients are relatively similar, as are their standard errors. All exogenous variables found significant in the estimated equations that do not account for enjoyment remain significant in the equations that do. In addition, when added to the equation, ENJOY is also significant. Therefore, the measure of enjoyment adds some additional information and collinearity does not appear to be a problem. Thus, results presented in Table 7 are discussed.

OLS results in Table 7 indicate that SAVINGS, YNGKIDS, and OLDKIDS are significant in the equation used to predict marginal savings. Higher perceived savings are associated with decreases in marginal savings. Since theoretically marginal savings decrease at a decreasing rate as search increases, consumers who perceive greater available savings may obtain lower marginal savings. Presence of children is negatively associated with increases in savings. Perhaps, families with children have less time to search for lower prices, thus obtaining fewer savings.

In the structural equation, ENJOY is significant and positive. Increases in enjoyment of search time increase time spent in weekly price information search. Hypothesis one cannot be rejected. Unfortunately, the use of principal components analysis to derive an index for enjoyment of search time allows only the direction of the effect and not the magnitude to be used in interpretation.

The coefficient of ESTMARSAVE is significant and negative, while the coefficient of WAGESHOP is insignificant. (7) To analyze the effect of the price of search time on demand for time spent in search, the coefficient of ESTMARSAVE is the appropriate parameter to examine. A simple calculation illustrates this. This calculation is necessary because the wage rate includes the price of all nonsearch time uses. The wage rate minus marginal savings equals the price of search time. The estimated equation is written

[Mathematical Expression Omitted] where Hs is the dependent variable, search time, [X.sub.k] is a vector of the exogenous variables (and estimated marginal savings), and e is the error term. To examine just the coefficients ([alpha]) on the price of search time and the price of other time uses, the relevant part of the above equation should read

[[alpha.sub.1]]Ps + [[alpha.sub.2]]W = [[alpha.sub.1]](W + Xo([derivative]Po/[derivative]Xs)) + [[alpha.sub.2]]W = ([[alpha.sub.1]] + [[alpha.sub.2]])W + [alpha.sub.1]]Xo([derivative]Po/[derivative]Xs), (8)

where Ps is the price of search time, w is the wage rate, and Xo([derivative]Po/[derivative]Xs) are total marginal savings on goods for which search occurs. However, the coefficients actually estimated are [[beta.sub.1]] and [[beta.sub.2]], as in

[[beta.sub.1]]Xo([derivative]Po/[derivative]Xs) + [[beta.sub.2]]W. (9)

The coefficient of marginal savings really measures the effect of a change in the price of search time on search time demand. The own-price effect of a change in the price of search time on the demand for search time is negative, as predicted by theory. As the price of search time increases, the amount of time spent in price information search decreases. However, ESTMARSAVE may overstate true marginal savings due to the way the variable was measured. Marginal savings were approximated by average savings per hour of search (see section on variable measurement). Consequently, only the direction and not the size of the estimated coefficient may be interpreted. Hypothesis two cannot be rejected.

The coefficient on nonwage income is positive and significant. Indications are that search time is a normal good. As nonwage income increases, the amount of time spent in price information search by dual earners for groceries increases.

The effect of age of the major shopper is significant and negative. Results indicate that for a one year increase in age, search time is decreased by about seven minutes per week. Age appears to measure experience of an individual. As experience in search increases, time spent in search decreases. Hypothesis three B cannot be rejected.

The presence of both young and teenage children significantly decreases the amount of time spent in price information search for groceries. Hypothesis five cannot be rejected. Since young children compete for time spent in other activities, it is plausible that time spent in search will decrease with their presence. However, the presence of teenage children is also associated with decreases in time spent in price information search. Hypothesis six is rejected.

Daily use of a microwave oven is significant and negatively related to time spent in price information search. Hypothesis seven is rejected; search time and presence of a time-saving durable do not appear to be complements.

The proportion of a shopper's grocery basket comprised of branded items is significant and negatively related to time spent in price information search. As a control for quality variation, these results indicate that shoppers appear willing to pay premium prices for increases in perceived quality.

Neither sex of the shopper nor attainment of a college education have an impact on time spent in price information search. Hypotheses four and eight are rejected.


Given the significant and positive effect that enjoyment of search time has on demand for time spent in price information search, it appears that including this psychological construct is appropriate. Though this result is intuitive, researchers have tended to neglect the inclusion of enjoyment when modeling the demand for time.

The model of price information search also indicates that the price of search time can be measured by a modified wage rate. Findings indicate that this measure appears appropriate, given the negative own-price effect of search time on the demand for search time.

The finding that search is a normal good is consistent with previous research that included both working and nonworking wives (Carlson and Geiseke 1983; Levedahl 1988). Both of these studies included total family income. If search is properly modeled using a household production model, the effect of total family income may include price effects, placing suspicion on the interpretation that search is a normal good. Positive effects of income are counterintuitive to expectations because higher income persons spend a smaller share of total income on food than lower income persons.

Levedahl's (1988) study of coupon redemption as a form of price information search hypothesizes that a positive association between income and coupon usage may be due to either increased ability to search or increased purchases of frequently promoted, more expensive, branded goods. Significance of the income coefficient partially supports the first reason. However, one would expect significant effects of education as well. This was not the case in this study. It may also be true that higher income persons purchase more of the higher priced brands for which coupons are offered. (8) However, the negative finding on the BRANDED coefficient does not support the second reason higher income persons may tend to search more. In the present study, BRANDED was used as an indicator of quality. Perhaps, consumers who purchase branded goods because they signify quality do not spend time searching for lower prices because they are brand loyal. The present study focused on more than just coupon redemption, making direct comparisons difficult.

These findings point to the need for future research in at least three areas. (1) Why do higher income persons tend to spend more time searching for lower prices? (2) Is the number of name brands purchased an indicator of consumers who are loyal to perceived quality and therefore search less? (3) Are true income effects confounded by the presence of price effects imbedded in the income measure?

The negative effect of age is at odds with one previous finding (Carlson and Geiseke 1983). It appears that in dual earner households, age affects search time differently than in a sample that includes both earners and nonearners. However, Levedahl (1988) found age to be positively associated with coupon clipping, one aspect of price information search.

The negative effects of both young and teenage children on search time may be explained. Regardless of age, children take time and, thus, compete for time spent in other activities. This finding is supported by that of Doti and Sharir (1981), who found negative effects of children on time spent shopping.

The negative effect of use of a microwave oven (MICRO1) is unexpected but explainable. The use of such an appliance can decrease time needed for cooking. Time may be freed for use in other activities, indicating the possibility that noncooking time and presence of a time-saving durable are substitutes. Bryant (1988) found the opposite result. Persons who own "time saving" appliances may spend more time engaging in the activities that use them. This may be the case in this study.


This study developed an economic model of demand for price information search in which the psychological variable enjoyment was explicitly included in the specification of a consumer's utility function. Understanding the effect of variables that influence time spent in price information search may be beneficial for both the demand and supply sides of the marketplace. Dual earner households appear to be concerned with the price of groceries, and they do take advantage of savings available on them, especially if the major shopper enjoys search time. By taking advantage of available savings, these households can free income for other uses. This finding also has implications for marketers who target high priced grocery items at dual earner households who face increased demands on their time. It appears that dual earners are sometimes, but not always, willing to pay the entire premium charged on these products. Marketers may be able to utilize promotional techniques such as sales and coupons to gain market shares among these households.

(1) For a summary of determinants of search efforts for a variety of durable and nondurable goods, see Beatty and Smith 1987.

(2) Looked at as a benefit, decreases in the price paid for Xo imply marginal savings increase at a decreasing rate, or ~[derivative]Po/[derivative]Hs~ and ~[derivative]Xs~ are positive. This is the form seen in some of the literature.

(3) Alternatively, (5) can be expressed

w + Xo [derivative]Po / [derivative]Hs = [derivative]U/[derivative]Hs / [lambda]

(4) This measure will capture time spent in prepurchase search and time spent to compare prices across stores. It is assumed that consumers who shop at more than one store do so to obtain lower prices on certain items. It does not capture the effect of comparison shopping within one store.

(5) Standard errors for the two test groups were calculated separately.

(6) Consumers were segmented based on whether or not they grew up with convenience foods and price saving coupons, which were more frequently seen after World War II.

(7) Insignificance of the coefficient on the wage rate may indicate that some nonsearch time uses are complements to search time while others are substitutes. The net effect cannot be determined conclusively.

(8) Coupons are available more often for manufactured foods and branded items (Bellizzi et al. 1981; Gallo 1982a, 1982b).


Alcaly, Roger (1976), "Information and Food Prices," Bell Journal of Economics, 7: 658-671.

Beatty, Sharon and Scott Smith (1987), "External Search Effort: An Investigation Across Several Product Categories," Journal of Consumer Research, 14(1) (June): 83-95.

Becker, Gary (1965), "A Theory of the Allocation of Time," The Economic Journal, 75 (September): 493-517.

Bellizzi, J. A., H. F. Kruechebery, J. R. Hamilton, and W. S. Martin (1981), "Consumer Perceptions of National, Private, and Generic Brands," Journal of Retailing, 57(4) (Winter): 56-70.

Belsley, D., E. Kuh, and R. Welsch (1980), Regression Diagnostics, New York, NY: John Wiley and Sons.

Bryant, W. Keith (1988), "Durables and Wives' Employment Yet Again," Journal of Consumer Research 15(1) (June): 37-47.

Carlson, J. and R. Geiseke (1983), "Price Search in a Product Market," Journal of Consumer Research, 9: 357-365.

Collins, N. R. and L. E. Preston (1969), "Price-Cost Margins and Industry Structure," Review of Economics and Statistics, 51(3) (August): 271-280.

Collins, N. R. (1968), Market Structure and Cost-Price Margins in the Food Manufacturing Industries, testimony before the Senate Subcommittee on Antitrust and Monopoly, Hearings on Economic Concentration, part 2.

Conners, J. M., R. T. Rogers, B. W. Marion, and W. F. Mueller (1985), The Food Manufacturing Industries: Structure, Strategies, and Performance, Lexington, MA: Lexington Books.

Doti, James L. and Samuel Sharir (1981), "Households' Grocery Shopping Behavior in the Short Run: Theory and Evidence," Economic Inquiry, 19(2) (April): 196-208.

Douglas, Susan (1976), "Cross National Comparisons and Consumer Stereotypes: A Case Study of Working and Nonworking Wives in the U.S. and France," Journal of Consumer Research, 3: 12-20.

Dow, Greg and E. Thomas Juster (1980), "Goods, Satisfaction with Activities, and Well-Being," Working Paper, Institute for Social Research, Survey Research Center, University of Michigan.

Gallo, Anthony E. (1982a), "Coupons: Part I," National Food Review, NFR 18 (Spring): 11-15.

Gallo, Anthony E. (1982b), "Coupons: Part II," National Food Review, NFR 19 (Summer): 12-26.

Gastworth, Joseph (1976), "On Probabilistic Models of Consumer Search for Information," Quarterly Journal of Economics, 90 (February): 38-50.

Grieg, W. S. (1976), The Changing Structure of the Food Processing Industry, Washington State University, Bulletin #827, (September).

Gronau, Reuben (1977), "Leisure, Home Production, and Work--The Theory of the Allocation of Time Revisited," Journal of Political Economy, 85(6) (December): 1099-1123.

Hacklander, Effie (1978), "Do Working Wives Shop Differently for Food?" National Food Review, Economic Research Service (April): 20-23.

Handy, Charles (1985), "Surveying Private and National Brand Prices," National Food Review, NFR-29 (Winter-Spring): 14-17.

Harris, Brian and James Stevenson (1983), Working Women as Shoppers, University of Southern California, Food Industry Management Program.

Harris, Brian and James Stevenson (1984), Men as Shoppers, University of Southern California: Food Industry Management Program.

Kaitz, Evelyn (1979), "Getting the Most From Your Food Dollar," National Food Review, Economic Research Service (Winter): 26-29.

Levedahl, William (1988), "Coupon Redeemers: Are They Better Shoppers?" Journal of Consumer Affairs 22(2) (Winter): 264-283.

Maynes, E. Scott and Terje Assum (1982), "Informationally Imperfect Consumer Markets: Empirical Findings and Policy Implications," Journal of Consumer Affairs, 16(1) (Summer): 62-87.

Redman, Barbara (1980), "The Impact of Women's Time Allocation on Expenditure for Meals Away from Home and Prepared Foods," American Journal of Agricultural Economics, 62(2) (May): 234-237.

Reilly, Michael (1982), "Working Wives and Convenience Consumption," Journal of Consumer Research, 8: 407-418.

Salop, Steven and Joseph Stiglitz (1977), "Bargains and Ripoffs: A Model of Monopolistically Competitive Price Dispersion," Review of Economic Studies, (October): 493-511.

Stigler, George (1961), "The Economics of Information," Journal of Political Economy, 69(3) (June): 213-225.

Strober, Myra H. and Charles B. Weinberg (1980), "Strategies Used by Working and Non-working Wives to Decrease Time Pressures," Journal of Consumer Research, 6 (March): 338-348.

Tauber, Edward (1972), "Why Do People Shop?" Journal of Marketing, 36 (October): 46-49.

Telser, Lester (1973), "Searching for the Lowest Price," American Economic Review, 63(2) (May): 40-50.

U.S. Bureau of the Census, Current Population Reports, Consumer Income (1986), Earnings in 1983 of Married-Couple Families, by Characteristics of Husbands and Wives, Washington, DC: Department of Commerce.

U.S. Bureau of the Census (1986), Selected Characteristics of Families, Washington, DC: Department of Commerce.

Wilkie, William and Peter Dickson (1980), "Consumer Information Search and Shopping Behavior," unpublished research paper.

Jane Kolodinsky is an Assistant Professor at the University of Vermont, Burlington, VT. This research was funded in part by the New York State Board of Agriculture and Markets, members of the Board of the NYS Milk Promotion Order, and by a Grant-In-Aid of Research from Sigma Xi, the Scientific Research Society.
COPYRIGHT 1990 American Council on Consumer Interests
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kolodinsky, Jane
Publication:Journal of Consumer Affairs
Date:Jun 22, 1990
Previous Article:Rehabilitation or liquidation: consumers' choices in bankruptcy.
Next Article:Family composition, parental time, and market goods: life cycle trade-offs.

Related Articles
The effects of consumer education on consumer search.
Quantity surcharges on groceries.
Infoseek and Starfish Software Partner to Provide Easy Access to Infoseek Information Services
BLAZE Web Performance Pack
Clarify Ships Account Manager With New Release of ClearSupport
UMI Unveils ProQuest Direct(R) 1.5
An online shopping solution.
Relationships among Information Search Activities When Shopping for a Credit Card.
Small Businesses Stand to Lose Big from Deregulation Troubles. Introduces Free Grocery Coupon Search Tool.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters