Your claims are unfounded though. The example given in the article, India, is an anomaly. Most countries are very happy with auctions and the results, both financially and coverage-wise. If they weren’t happy, why would they all keep running them? There are dozens of spectrum auctions planned over the next few years around the world, because every country realizes it’s the best way to raise money, the fairest way to distribute spectrum (ie they won’t get sued), and the best for consumers.

Now, is every country perfectly set up for a successful auction? No - but in those cases you’ll see special rules meant to promote competition in the market. New entrants may receive discounts, or certain blocks of bands may be set aside for new entrants only. Big companies may be capped in how much additional spectrum they can purchase is another common case. These auctions are almost never straight up “take what you can get”. You can imagine in the US if that was the case then AT&T and Verizon would own all of it.

My company writes the software for these spectrum auctions, and I’ve personally been a part of dozens of them.

He is correct. In Germany the auctions for umts licenses costed 50 billion euros. These insanely high auctions (because every telco believed they need these frequencies and outbid everyone) has negative price impacts today, something which happened 20 years ago! Other countries had better outcomes, but you cant foresee how such an auction will go and which impact it will have.

1. Many telecom companies in other countries that use auctions have pretty decent returns. See, e.g., ATT and VZ among others.

> other solutions exists - e.g. revenue sharing - which work very well

The terms of the revenue sharing need to be worked out somehow. That "somehow" is typically via some sort of competition for scarce resources (spectrum), e.g., an auction. Sure, the gov't could just set a price, but the inefficiency and lack of price discovery inherent to that approach is precisely the issue the auction seeks to address.

> Top down market design is the highest virtue of all even though hundreds of millions of very poor people might have suffered for academic theories of a few.

Can't stand this snarky, melodramatic rhetoric. I might as well talk about millions of very poor people suffering because they lack clean water and healthcare which the government could have provided with the revenue from spectrum auctions.

Cellphone service has markedly improved the lives of the very poor something most economists did not predict. In fact when India deregulated rates from some of the highest in the world (voice was around $.10 to $.20/min) most economists argued against it saying cellphones were a thing of the rich and the tax-losses to the govt from telecom revenue were not worth it. Deregulation was called a give-away to the telco companies and the rich etc. Instead, what came about was the biggest uptake in cellular and data usage and one of the cheapest rates in the world.

The poor benefit disproportionately in relative terms from access to cheap communications. Studies show cellphone and data usage to be highly correlated to improved markers in health, economic well being, education, etc. Poorly designed top-down auctions bankrupts companies and causes under-investment in infrastructure do directly cause suffering to hundreds of millions of very poor people. Maybe you find it snarky to point that out but I see it as a tragedy.

True, you can also argue that the auction revenue would have helped clean water and other developmental activities but that is an indirect effect.

This guy gets it... as an Indian that has lived in the USA / Canada for the last ten years, the mobile phone prices from before I left india and now are day and night .... used to pay around 12 -15 dollars for 2 gigs of data ... when I was in india this February, you could get 1.5 gigs of 4g data per DAY with almost unlimited calls for like 4 dollars a month ... it’s improved the lives of the people in so many ways ... so yeah maybe the government lost out on the auctions but it’s massively benefited the general populace ... and knowing how corrupt the Indian government is , I am happy with the way things turned out ...

when it costs almost nothing (remember, majority is loan here) to them why they need to charge. They take your data (totally unregulated) which is far more worth than the data ppl use. as an ISP, new apps, almost every industry under their control they got everyone's balls

Do you have a citation for your statement that economists opposed deregulation? That doesn’t sound like the default economics position — nor does not realizing that technology can transform poverty.

I’m not challenging you randomly here — usually the criticism of mainstream Western economists (especially those working with development) is that they overprescribe deregulation and underestimate the benefits of maintaining government revenue.

Economists != western economists! Western Economists did not care about the Indian telecom markets in the 90s and hardly carry any weight in Indian politics even now. India has enough economists unfortunately who can make a hash of things without help from Western economists ;)

It's appalling to hear that economists would think that improving the flow of goods and information would not benefit the economy.

But yeah, it's easy to miss. We've only had Roman roads, book printing, railroads, telephone, container shipping and the internet (to name a few) to point out this evident truth..

> Cellphone service has markedly improved the lives of the very poor something most economists did not predict. In fact when India deregulated rates from some of the highest in the world (voice was around $.10 to $.20/min) most economists argued against it saying cellphones were a thing of the rich and the tax-losses to the govt from telecom revenue were not worth it. Deregulation was called a give-away to the telco companies and the rich etc. Instead, what came about was the biggest uptake in cellular and data usage and one of the cheapest rates in the world.

Hard to believe and irrelevant to this discussion.

> Poorly designed top-down auctions bankrupts companies and causes under-investment in infrastructure do directly cause suffering to hundreds of millions of very poor people.

What is a top-down auction? Are there also bottom-up auctions?

That auctions bankrupt companies - why are they are participating if they can't afford to? - and that they cause under-investment - is a spectrum that's been bought in an auction less profitable? - sounds like propaganda from corporations that want spectrums for free. It neither makes economic sense, nor am I aware of any empirical evidence for it.

Give a man aid and he's fed for today. Give him a cellphone and he has a job.

Economists peddle wild ass guesses as science. Missing the boat on cellular is par for the course.

Why was this downvoted? Seems like a fairly well-written argument - would love to hear counter-arguments instead of blanket downvoting.

I'll give it a go. It's a nice argument, but it's needlessly negative, almost for show.

Consider the claim that's it's a loss if the company shows poor returns. If it had good returns, the anti-auction folks would claim some small group is getting rich off what used to be a public good. Poor returns for the company means that the auction is limiting the plunder and that sounds like a win to me.

And why is this a loss for the government treasury? If the returns are poor, there wasn't much more juice to squeeze from the orange, right? So it sounds like the government (and often the taxpayers, sort of) are getting as much as they can for very little work.

I'm much more sympathetic to the deeper arguments about structure of auctions and how they might be improved. Why not auction the spectrum off for a shorter time? Maybe use smaller chunks of spectrum? The auctions are just a competitive algorithm and they can become more accurate with finer resolution.

Governments gets 2x revenue from revenue sharing than from auctions so govts do lose. Your argument uses self-justification but there is a counter factual for which empirical data exists.

Companies also lose as they have fixed liabilities on their balance sheet. If companies are hamstrung by debt then they do not have resources to invest in their network, subscribers suffer.

Who exactly wins for this grand experiment in top-down auction design?

Revenue sharing should be implemented via auction. Otherwise how do we discover the efficient percentage to be shared? All of the problems you pointed out have little to do with auctions, which are just a way for markets to discover prices when supply is very limited, and everything to do with poorly implemented auctions.

>Governments gets 2x revenue from revenue sharing than from auctions so govts do lose.

Citation needed.

>there is a counter factual for which empirical data exists

And yet you provide none.

Meta-ironic because it is in the article :)

It's rev-share vs pay upfront with financing right?

A telco could never be overburdened by the rev-share in the same way as a upfront payment. If the rev share is too high so using the spectrum isn't economical, the telco could simply not build out (and probably lose their license, which is OK for them, it wasn't economical). If the prepaid price was too high, they still have to pay it from spectrum activities and other activities; but they probably can't sell the spectrum to retire the debt, because the price was too high.

Isn't that incentivizing companies without much existing market share in the area to just bid high revenue shares? At best they get some money and at worst they close up shop thus coming back close to neutral. Maybe they even get to screw a competitor for a few years as they stall build out. Of course the population gets hurt if the latter happens since they get no infrastructure.

You can adjust the auction terms to try to avoid bad outcomes. A) bid out smaller alocations to allow more networks B) require a deployment plan to bid, and progress on the plan to keep the allocation C) some equitable way to periodically rebid in absence of abandoned spectrum D) some sort of open access provisions (maybe spring this on the market after it's largely built out)

Of course, some of that requires a competetent regulator with clear authority, as well as not too much corruption and reasonably swift court systems to handle disputes. But a point in time, single payment auction works better with those things too.

You will find in any rev-share agreement, a requirement that the Co. actually make "best efforts" or the like towards actually generating revenue or they lose the license rights. Best efforts is not some fluffy term, but basically requires the company to move heaven and earth if necessary otherwise they are in violation of the agreement.

I can totally understand how rev-share can be better in comparison but I don’t think your argument here addresses the constraint of greater government revenue derived from the rev-share model.

I suspect that part of that may be market growth beyond expectations. If telcos were bidding about the same net present value for the one time auction and the rev-share auction, with their revenue projections and revenue came in higher, that's going to be better for the government.

It's also easier to bid higher in rev-share, because you don't need to finance that net present value; it's just going to be built into the price of service, and your competitors will likely have similar rev-shares (depending on the auction process), so it's going to be built into everyone's prices.

I think there's also some value in a stream of payments rather than a one time payment. Although you can exchange one for the other with financial tools, a government may find a stream of payments to it matches well with the stream of payments it's making.

If the government auctions off spectrum at a flat price then it gets certainty, which normally comes at a premium; doing revenue share the government retains more of the risk, so orthodox economics would say that it will get a better price on average.

There's also an argument that the government doesn't actually realise any advantage from selling off the risk: if the telecom companies fail, it's still the government's problem. So the government ends up paying extra for something it can't actually use.

I think you may have misread what was written when you claim

> It's a loss because a big, capitalized winner takes all auction benefits a small number of entities with big pockets.

This is the exact opposite of what the parent poster was arguing against which was the claim that the TelCos had poor returns (ie they were NOT particularly beneficial to those entities).

Spectrum is generally sold on 15Mhz blocks. 10 for download, 5 for upload.

Its a statement, not an argument, as I read it. Theyre ignoring other confounding factors, the agency of both telecoms and policy makers, and basing the lose-lose-lose on a limited subset of auction markets. Additionally theres no other suggested mechanism besides a vague wave at “revenue sharing” where you have the same problem of determining what the appropriate rate is.

OP suggested markets work. HN folks downvoted without presenting arguments.

Without disagreeing with the lose-lose-lose premise how do you draw the blame to the auction mechanism? The telecoms are in control of their bid values. The government policy makers are in control of minimum bids, related contingencies, and how the proceeds are spent. And yes, consumers are saddled with the confluence of the two.

With those controlling factors or problems Im not clear how a different mechanism, say fixed rates and allocations, is above the same failings?

ironically, an industry still using outdated modes of revenue measurement can't be said to be thinking "bigger" or better.