Skip to main content

Logistics as a network function

4 min read

Logistics is the ultimate in network functions, if not the original network function. Jon Elledge at CityMetric makes a good case out of the enduring suckiness of home delivery firms:

In fact, there might be a structural reason why delivery firms are so often rubbish:  they’re accountable to the wrong people. When you order something online, you don’t pick who delivers it, the retailer does. As a result, you can’t boycott the delivery firm; neither are they the ones liable to compensate you if they screw up. There’s not enough payback for failure.

To make matters worse, many of these firms rely on self-employed drivers (this is particularly so at peak times such as Christmas, but seems to be true all year round). These guys are expected to do something like 100 drops a day, and are paid by the delivery. Leave aside the fact they’re even less accountable to you than their employer is, and consider how this’ll influence their behaviour. They have every incentive to prioritise easy deliveries, and no incentive whatever to care about you. If you’re slow to the door; if it’s difficult to park; if they forget to collect your parcel altogether, then that’s just too bad.

Would boycotting online retailers who use these firms change any of this? Eventually, perhaps. But even if the public were willing to give up its home shopping addiction, the lack of transparency regarding which delivery firms a retailer uses would rather blunt the attack.

The bottom line is that delivering parcels is an expensive game. You need a national network of depots and drivers and, ideally, a call centre (all of which might make one ask if we weren’t better off with a single national Post Office). The business is seasonal; the overheads are high. These are not obviously lucrative firms. It’s just possible that the service we get is the one we’re willing to pay for.

All logistical functions must be collectively owned and monopoly-managed. This is not (only) a political argument, but a network-theoretical argument.

In any network, the "last mile" is hideously resource-intensive, while only the trunk routes ever hit efficiencies of scale related to volume. In any process of privatisation, the trunk route franchises are bought up first, because those economies of scale make for easy profits... but those profits are only possible if you're not also losing money on keeping up the last mile, where economies of scale rarely (if ever) apply. Why should an operator pay for the last mile if they're not operating it, you might ask? Well, because without the last mile feeding traffic to and from the trunk routes, there would be less trunk route traffic to carry. The profits to be made on the trunk routes are actually rent extraction from the last mile; the separation of the two is a financial fiction which can only ever work against optimal function and efficiency. Because once the network is fragmented and the best routes have been bought up, the operator companies can then run the numbers on the last mile operations remaining, point out to councils that they're inherently unprofitable as constituted, and then demand subsidies in exchange for providing last mile connectivity, which the state has no option but to provide (because if people and stuff can't move around, your economy is fucked).

Which is how you end up with the delightful irony of Stagecoach raking in subsidy money from dozens of UK councils in exchange for sparse and sketchy local bus routes, while also running extremely profitable (because separately contracted) franchises on intercity coach routes and train lines. It'd be interesting to run the numbers and see whether the profits of trunk route operators and the subsidies paid out to last mile operators would be anywhere close to cancelling out if you looked at them nationally; my suspicion is that it'd be closer to that than not.