Hacker News Reader

Comment by TazeTSchnitzel | Hacker News Reader

It's good for Amazon, but is it good for consumers? Isn't a postal service something of a natural monopoly — the more competing services there are, the less efficient each is, because instead of one vehicle with one load of deliveries, you have two or more vehicles with fractions of that amount?

It's certainly not good for the environment at least.

View on HN

Natural monopoly means that it's not possible for more than one supplier to survive. This is clearly not the case for packages delivery.

For consumers the more suppliers the better.

Consumers benefit assuming all suppliers are equivalent, which is not at all the case with Amazon. They're gaining a price advantage over using real carriers mainly through incredibly exploitative labor practices in the warehouse and abusing uber-like contractors to cover the last mile for pennies on the dollar.

Not only is the human/societal cost high, consumers directly suffer as well as untrained contractors working irregular routes rush to meet unrealistic deadlines. Every day a different person in a rented van pulls up to throw armloads of packages at my building, may or may not be the right entrance, they don't know the place and have no time to check, before racing off to the next location.

Consumers benefit because of competition and redundancy several suppliers create.

Competition forces suppliers whose service is deemed inadequate by the market to change or disappear.

Clearly the market wants packages delivered fast and cheap.

If the model is failing as you describe, then more competition is a good thing. Consumers will flock to whoever that provides better service for what it is they’re willing to pay.

I’m not sure how consumers suffer - if trained or untrained contractors fail at their jobs, consumers take their business elsewhere.

It sounds like you’re against Amazons delivery practices. Fine. But the more competition there is, the better. UPS or Fedex absolutely shouldn’t own the private delivery market.

I think what he means is sometimes there is a benefit to having one service provider. For instance, it is cheaper if a city offers trash pickup in a district to one service provider. The logic is that it’s much more efficient to have one truck stop at each house versus trucks zig zagging around each other.

hm, being 'more efficient' would be true if the ideal way was known.

Waterfall would also be 'more efficient' if we knew the right product to make and way to make it. If agile has taught us anything it's that we don't know and we can't know. We learn through trial and error and continually innovate. If there is only one company there is no incentive to innovate and discover better efficiencies.

So while the 'logic' is right, the supposition[0] is incorrect.

This is why 'the market' is the best way in practice. We observe an ebb and flow of centralization (Amazon is a good example) and decentralization in various industries. This is caused by innovation and disruption. When one company innovates/disrupts so well and continues to execute it can become a de facto "monopoly," gaining near 100% market share.

When the industry and market are allowed to continue unabated[1] this 'monopoly' can eventually be broken if the company fails to innovate and leaves the industry vulnerable to upstart disruptors. The very possibility means even with 100% market share the company is not a true monopoly, and in practice the consumer benefits from the company's 'existential fear' which motivates it to continue to serve its customers in better and better ways.

[0] knowing the best way [1] assuming the major player(s) doesn't lobby/enact regulations to protect its market/industry position

edited footnote formatting

The data doesn’t support this though in the example I gave. This is why many cities like LA are switching to a single provider per area. Note, there still is competition. The contracts go through a bidding process and a competitor could try and develop more efficient processes in a different district of LA. It just doesn’t make sense to have five providers serving the same block in a residential area as it drives up the costs for everyone. For more rural areas, the issue becomes that the capex to support the landfill and truck buildout doesn’t typically support multiple competitors.

I suppose this isn’t much different than a development team trying to work in both Postgres and MySQL. Sure, maybe you hedged your bets and your team uses both to experiment which is better, but at some point, it likely is better to develop expertise and experience in one of them especially if it’s a small team.

I think there is enough packages to be delivered that you would more likely have two trucks with full loads anyways. These companies aren't making any shipments with anything less than a full truck, I'm sure of it

But surely they would rather have one truck service a shorter route than have two trucks service the same longer route if it leads to the same overall utilization rate.

I think Amazon delivering stuff I want is better for the environment than USPS delivering junk mail daily.

I feel delivery companies should bid for providing service for the last couple of kilometers, then the winner should get an exclusive contract for some time frame. That's how it's done for other services like regional rail public transport.

Did I hit a nerve by picking public transport as an example? If it does not work out, regulations are bad. Competing non cooperating infrastructure service providers are a bad idea if they hinder each other.

>That's how it's done for other services like regional rail public transport.

Ask any Brit how privatized rail's working out for them.

Regional rail works okayish in Germany. A part of the regions and lines are carried by private business, others by the former monopolist that is still fully owned by the state.

Why exclusive contract? What is gained? Rail right of way is exclusive just from geometry and there are various historical "accidents" involved in ownership from no longer relevant reasons but there are myraid ways to deliver mail which may or may not be excluding.

A forced bidding system seems to give gameability and artifical barriers to entry for little gain. To give an anchhacic example of the sort of problems it would be akin to requiring a business that can serve all of New York City competively before bidding. Having a mail room in the lobby of a large apartment would violate that due to serving only residents.

I was thinking of parcel deliveries. Narrowish European inner city streets are regularly blocked by delivery trucks and fundamentally it's unnecessarily many of them because there are so many different companies servicing the same street. Where I live double parking on bicycle lanes is extremely common (the individual delivery people often see no alternatives), which poses massive dangers to cyclists. Also these trucks typically are quite large, worsening all issues, because a full load is meant for a long route.

If there was just a single company responsible for some "last mile" area, smaller vehicles like this electric model [1] would suffice. They would distribute parcels from a depot that is filled by larger long distance trucks that do not need to belong to the same company.

The last point is important: If DHL, FedEx, UPS etc. each have their own final storage buildings and delivery fleet, those depots could not all be situated close enough to the target addresses.

[1] https://en.m.wikipedia.org/wiki/StreetScooter

Reply on news.ycombinator.com