A Principal-Agent Problem

A Principal-Agent Problem

When a consumer orders from an online retailer, the retailer typically chooses the carrier, negotiates the rate and pays for the service, sometimes passing on part of the cost to the consumer as a delivery charge and sometimes absorbing it. The consumer, who meets the carrier's driver and lives with the consequences of its performance, usually has no say in which carrier is used and often does not know until the tracking email arrives. The carrier's commercial customer is the shipper. Its operational customer is the recipient. The two want different things. The shipper wants low rates, reliable collection, a smooth integration with its systems and as little trouble as possible. The recipient wants the parcel to arrive when promised, intact, where they can get it, and wants someone to answer the phone when it does not. An economist would see something like a principal-agent problem here, though not the textbook kind, since the carrier is a contractor rather than an agent in any legal sense: it serves a shipper who is paying for the satisfaction of a third party, the recipient, who has no contract with the carrier at all.

Several consequences follow. The first is that competition among carriers for shippers' business tends to be fought heavily on price, because price is what the shipper sees on the invoice, while quality is diffused across millions of recipients who cannot switch. The second is that a carrier's reputation with the public can be poor without its commercial position suffering in proportion, until the poor reputation begins to damage the retailers who use it, at which point it matters very suddenly. British readers will recognise the pattern from the long public battering of the carrier formerly known as Hermes, which renamed itself Evri in 2022 amid persistent complaints about lost and mishandled parcels, while continuing to win and keep the business of large retailers. The third consequence is that the recipient's voice reaches the carrier mainly through two channels, both of them indirect: complaints routed back through the retailer, and the slow, aggregated pressure of retailers choosing carriers whose delivery experience reflects well on them.

Amazon's decision to build its own delivery network is, among other things, an escape from this structure. When the retailer and the carrier are the same company, the recipient's experience becomes a direct commercial concern of the party running the vans, and the company can trade off delivery cost against customer satisfaction inside one profit-and-loss account rather than across a negotiation. That integration has costs as well as advantages, and the integrated model is not obviously reproducible by companies without Amazon's peculiar combination of volume, capital and control of its marketplace. But it explains why so many of the improvements in consumer delivery speed of the past twenty years came from a retailer rather than a carrier. The carriers' customers had been asking them for lower prices. Amazon's customers had been asking it for faster deliveries, and Amazon was in a position to hear them.

from Delivering the Future: The Companies, Technologies and Decisions Reshaping Delivery (2026)

Next
Next

Moral Aggregation