Working Paper: CEPR ID: DP8862
Authors: Gregory S. Crawford
Abstract: Empirical models of differentiated product demand are widely used by both academics and practitioners. While these methods treat carefully the potential endogeneity of price, until recently they have assumed the number and characteristics of the products offered by firms are exogenous. This paper presents a progress report on an ongoing research agenda to address this issue. First, it summarizes how the appropriate choice of 'orthogonal' instruments can yield consistent estimates of own and cross-price elasticities in the presence of endogenous product characteristics. Second, it summarizes how to measure 'quality markups' and the welfare consequences of endogenous product quality in U.S. cable television markets. Related papers and extensions to consider multiple product characteristics and dynamics are also discussed.
Keywords: Cable Television; Endogenous Product Characteristics; Endogenous Product Choice; Endogenous Quality; Orthogonal Instruments; Welfare
JEL Codes: C31; C52; L15; L40; L82
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| orthogonal instruments (C36) | consistent estimates of own and cross-price elasticities (C51) |
| endogenous product characteristics (L15) | consistent estimation of price parameters (C51) |
| price and quality uncorrelated (L15) | consistent estimation achievable (C51) |
| appropriate instruments correlated with price but uncorrelated with error term (C36) | consistent estimation of price parameters (C51) |
| market power over quality (L15) | welfare effects from quality choices (L15) |
| welfare effects from quality choices (L15) | comparable to welfare effects from price choices (D69) |