Smart data pricing: To share or not to share?

This paper studies a monopoly telecom operator's decision on the adoption of shared data plans. A shared data plan allows sharing data quota among multiple devices or users, while conventional single device data plans only allow the use of a single device. We devise analytical models and compare a simple shared data plan (also called bundling pricing) to single device data plans (also called partitioned pricing). When consumer valuations (utilities) of different devices are independent, we find a threshold on the unit usage cost below which the shared data plan yields more profits than single device data plans. The optimal price for the shared data plan is less than the sum of the single device data plans. If consumers' valuations for different devices have different distributions, this disparity reduces the relative value of the shared data plan against the single device data plans. We also show that shared data plans increases the social welfare and consumer surplus when it yields a higher profit. We further extend the analysis to complementary valuations on different devices: a consumer's valuation for using both devices may be higher than the sum of his utilities on the devices when only one device is used. We identify a threshold on the unit usage cost below which the shared data plan yields more profits. The price of the shared data plan is larger than the sum of the single device data plans for very strong complementariness, while it's always less with independent valuations. We also show numerically that a strong complementariness shrinks the range of the unit usage cost where the shared data plan has a higher profit.