A STUDY OF ECONOMIC FACTORS WHICH INFLUENCE CONSUMER SEARCH FOR PRICE INFORMATION

LINDA MARGARET KRAMER ZIMMERMANN, Purdue University

Abstract

This study was undertaken to identify consumers who seek price information prior to a purchase decision and those economic factors which influence search effort for lower prices. A conceptual framework was developed suggesting the following relationships: (1) wage rate has a negative influence on search effort because as wage rate increases search cost increases; (2) price dispersion for a good has a positive influence on search effort since as price dispersion increases the potential gains from search increase; (3) the portion of income represented by a purchase has a positive influence on search effort because as relative expenditure increases potential gains from search increase; and (4) level of skill in search has a positive influence on search effort since as level of skill increases search cost decreases. The data base was a sample of 311 couples married in 1968 in Peoria, and Decatur, Illinois, who were chosen for a panel by the Survey Research Laboratory of the University of Illinois. Panel members were asked whether they had purchased in 1968 an automobile or any of twelve durable goods. The price paid was recorded for each item purchased. For the first item purchased from the list, questions were asked about information-seeking prior to purchase. There were 95 new and 71 used purchases. Three of the information-seeking activities represented extent of search for price information: (1) discussion with others of potential brands, (2) discussion with others of potential dealers, and (3) number of dealers visited. Since no direct measure of wage was available, income and occupation were used to calculate two estimates of wage for both husbands and wives. Standard deviation of expenditures was used to represent price dispersion. The ratio of expenditure to income was used as the measure of relative portion of income represented by a purchase. Number of years of school completed represented level of skill in search. Regression analysis was the statistical technique used. Since preliminary results suggested that there were differences in price search behavior for new and used goods, the regression analyses were done both with constrained and unconstrained specifications. The constrained specification was based on the assumption that there were no differences in behavior between new and used. The unconstrained specification allowed for separate sets of regression parameters for new and used. Ratio of expenditure to income coefficients were consistently positive and frequently significant for all three dependent variables as hypothesized. Level of education coefficients were generally positive as hypothesized when discussion of potential dealers and number of dealer visits were dependent variables and frequently significant for all three dependent variables when wife's level of education was used. The parameter estimates of wage were inconsistent in sign and not often significant. The coefficients of expenditure standard deviation were almost always negative, not as hypothesized, generally not significant, and not meaningful because of their small size.

Degree

Ph.D.

Subject Area

Economics

COinS