Go to Top

Amazon, Uber And The Postal Service

USPS from behind, open to reveal packages for delivery
photo by Lisa Brewster

President Trump thinks Amazon is getting a free ride courtesy of the U.S. Postal Service and other USPS customers, and he could be right. But he probably isn’t.

I don’t blame the president for being misinformed in this case. There is a lot of misunderstanding about how rational managers set prices in competitive businesses. There is also a considerable amount of off-base commentary – most but not all of it innocently misguided – about whether USPS is providing services to Amazon below cost, and whether the market for package delivery services would permit postal managers to raise prices.

For a reality check on the second point, kindly pull out your smartphone and open Uber, assuming you have the app. If you happen to be in a city or suburb, chances are you’ll see several ride-sharing vehicles in your immediate vicinity. In a crowded downtown, there could be more than a dozen. Now put your phone away and hold that thought. We will come back to it later.

First, we must consider what the term “below cost” means in the case of USPS and Amazon. There are at least four types of “cost” that are relevant in pricing negotiations between the Postal Service and a very big customer. These are fixed costs, variable costs, opportunity costs and avoided costs. Most of the mischief in this debate arises when self-interested parties (I’m looking at you, FedEx and UPS) define the costs of a competitor, namely the Postal Service, to suit their own purposes.

Like any business, the Postal Service has fixed costs and variable costs. Payroll, including employee benefits, is mainly a fixed cost. It isn’t “fixed” in the sense of being an immutable law of nature; if you hire or fire workers, or raise or lower their base pay, you change the level of your “fixed” costs. But as you look at your existing workforce, you are going to pay pretty much the same base salary and provide benefits at the same cost whether those workers are productive or idle. So you want to keep them productive.

At a certain point, you have reached maximum output that the workers can generate with existing facilities in a standard work period. That’s when you start paying overtime or making capital investments. Overtime is a variable cost. The fuel it takes for a postal vehicle to cover its assigned daily route is a fixed cost (with trivial variation according to the weight of the payload), but the cost of fuel to make a second run to carry overflow packages is a variable cost. The cost of the vehicle itself and its regular maintenance is fixed; the additional wear and tear of that second run is variable.

Capital investment to make existing workers more productive is a fixed cost once it is incurred. If you can afford it, and if you are confident that future demand will cover that cost, you’ll spend money on equipment now to avoid paying greater sums in overtime and other variable costs later. Otherwise you will rely on variable costs, because they give you more flexibility.

To avoid losing money on its Amazon deliveries, USPS must charge at least enough to cover its variable costs for delivering those packages. Any excess above those variable costs leaves the Postal Service better off than it would have been if Amazon had taken its business elsewhere. This excess is “margin.” Every successful business must ultimately generate enough margin to cover both fixed and variable costs, but you don’t generate more margin if you include fixed costs in your pricing to such an extent that you send potential customers to your competition.

USPS competitors like FedEx and UPS have an incentive to drive postal rates higher, because doing so lets them price their own package delivery services higher without losing business to the mail carrier. One way to do that is to steer the debate toward including more fixed costs in USPS pricing formulas, even when the Postal Service would be better off leaving those costs out. After all, those costs are fixed; USPS will pay them whether it gets Amazon’s business or not.

So should USPS pursue all the Amazon business it can possibly get, at any price above its variable costs? Not necessarily. There is also “opportunity cost” to consider. If a cut-rate Amazon package crowds out another package that could be delivered at the same marginal cost but at a higher price, the Postal Service could be better off sending Amazon’s business elsewhere. I have not seen any indication this is the case, however. In fact, FedEx and UPS themselves use USPS for expensive, labor-intensive “last mile” delivery in many cases, especially in less-populated areas. Here is another reason they would want the Postal Service to lose business from Amazon: It would make the Postal Service more desperate to take FedEx and UPS packages at prices closer to variable cost, squeezing the Postal Service’s margins. This would further weaken FedEx and UPS’ hamstrung competitor, even as it relieves them of some of their delivery burdens.

Now look at the situation from Amazon’s point of view. Despite what Trump may think, Amazon has ample alternatives to using the Postal Service for delivery. It can, of course, simply take its business to FedEx and UPS, piggybacking off their increased leverage with USPS.

Amazon could also build out its own last-mile delivery network and then compete with the three package carriers. Amazon is the Jedi master of turning its infrastructure expenses into profit-making businesses. Thousands of enterprises sell through Amazon in order to use its payment and fulfillment systems, and thousands more use Amazon’s web services to run their own online operations.

Why doesn’t Amazon do this? Partly because it has opportunity costs, too. Amazon has a lot of money, but its resources are not unlimited. If using existing package carriers is actually cheaper than creating its own delivery network, the “avoided cost” of creating that network provides incentive enough to stay out of the delivery business. Even if Amazon could save a few bucks delivering its own packages, it might choose not to do so in order to deploy its resources more profitably elsewhere. In other words, opportunity costs could be too high, even if avoided costs were not.

Finally, there are those Uber drivers, who are almost entirely overlooked in the discussion of parcel delivery services.

Every one of those drivers could deliver packages just as easily as they already deliver passengers – or restaurant meals, via the Uber Eats platform. Collectively they are a vast, underutilized network of potential last-mile couriers, who could drop off packages en route to or returning from one of their regular passenger or food delivery runs. Besides Uber, there are competing services like Lyft, and even traditional licensed taxis – not to mention every local delivery truck that sits idle for many hours of the day awaiting runs for the local produce supplier or furniture store.

All they need is to be mobilized, and probably nobody could do that better than Amazon. The company is already experimenting with its own in-house version of side-gig package delivery, which it calls Amazon Flex. Uber drivers, and all those other potential delivery sources, serve to lower the potential avoided cost of creating a new, competing parcel delivery service from scratch. The Postal Service and its existing competitors have to keep their prices down to avoid incentivizing Amazon or some other major shipper to create a whole new business model for the service of moving boxes locally from point A to point B.

The USPS has huge ongoing financial challenges. In my opinion, it is unlikely to make it through the 21st century in its current form. But too-cheap package delivery is not the source of its problems, and raising prices significantly is not going to be a solution.

An outmoded mandate to deliver first-class mail at a uniform price to every address in the country no matter how remote, which in turn leads to a bloated and hopelessly uneconomic labor force made less competitive by antiquated union rules and contracts, is what ails the Postal Service. An analysis by Citigroup last year found that delivery prices would need to rise by $1.41 per package in 2018 to reflect what its authors consider the true cost of delivery, an increase of about 40 percent. Even if the Postal Service wanted to hike its prices so dramatically, existing law has capped price increases at the rate of inflation.

Ultimately, all of the proffered solutions, from expanding into financial services to raising prices for monopoly and competitive products, amount to the same thing: looking for a magic source of money to cover the shortfall created by an irrational business model.

A workable answer would be to free the Postal Service of its government strictures and labor rules to be fully competitive; have the taxpayers (whose political demands created the Postal Service’s existing unfunded liabilities) assume its current debts; and let state and federal governments subsidize services to remote areas to the extent they choose, just as we do for aviation with the Essential Air Service program. I don’t expect this to happen any time soon.

In the meantime, don’t overlook those Uber cars in your neighborhood. We can be sure that Amazon won’t.

, , , , , , , ,