It has been diffcult to pick up the business section of a newspaper this past year without reading about some development affecting electronic home shopping: broadband networks that can accommodate increased two-way data transmission, experiments with existing media like CD-ROM and on-line services, and new alliances. The most enthusiastic observers predict industry sales of more than $50 billion during the next ten years. Delayed trials of certain services have made others more cautious. In neither camp, however, is there much of a consensus about the likely structure, profitability, and impact on traditional retailing of electronic home shopping. Very little about the opportunity it represents is clear.

Our own bottom-up forecast of how the industry will develop is conservative, yet suggests attractive potential. As shown in Exhibit 1, our estimate is based on the product categories that can be addressed by remote channels. Since we assume that the new electronic services will not increase overall consumer spending, our estimate includes the share of current channel volume likely to be cannibalized by interactive services. The result: we think there will be a $4 to $5 billion market in the US by 2003,...