Assessing Chain-Restaurant Impact—

A regression model provides a direct statistical test of the impact of new units on existing units in a competitive area while allowing for additional factors that may moderate the causal link. The model is easily accessible to management practitioners. In its simplest form, it is as follows: If Yt is the customer count in month tfor an existing unit, then Yt= a + PjXlt + 8t, where random shocks in period tare indicated by atand the entrance of the new unit by X4t. Under ideal conditions, the estimate of X would indicate the average monthly customer count of the unit exclusive of the effect of the new unit's entry. The estimate P3 ideally indicates the net average number of customers per month that the new unit has absorbed (the existing unit's lost business).