Čeština
Deutsch
Español
Français
Magyar
Italiano
Português
Русский
Serbian
Svenska
Free Market Failure: Telcos Charge More For Sending A Text Next Door Than Cost Of Sending Data From Mars
Infrastructure
Infrastructure
The telco industry charges more, kilobyte by kilobyte, for sending a text message from your phone to next door than what it costs to send the same message from Mars to Earth. This is the apex in this series of the dysfunctional telecom market, giving a background to why the telecom industry wants control of the Internet so badly, and is using every conceivable resource to stall, prevent, and delay its resulting economic development.
For the third installment in this series, we focus on text messaging. FoI reader Chris Monteiro suggested that we should describe how it is more expensive to send SMS text messages from your phone, kilobyte by kilobyte, than it is to send the same data from Mars to Earth. That couldn’t possibly be right, we thought, but nobody seemed to have done the math before.
So let’s do the math.
The cost of the Mars Global Surveyor probe was roughly 200 million USD for the satellite and launch, plus 20 million per year. So, 400 million USD. It operated for nine years, transmitting at an assumed average of 42,667 bps. Assuming it transmitted 24/7, that comes down to 42,667 × 3,600 × 24 × 365 × 9 / 8 / 1,024 / 1,024 / 1,024 = 1,410 gigabytes of data at a cost of 400 million dollars, or roughly 284,000 US dollars per gigabyte. That number includes the cost of the actual Mars probe and its launch, as well as the cost of the NASA crew handling its journey to Mars for almost a year before it started transmitting.
The charge for sending an SMS text message next door is about 5 USD cents (let’s use the same currency for simpliticy’s sake). Each text message is 140 bytes. This means that there are 1,024 × 1,024 × 1,024 / 140 = 7.67 million text messages per gigabyte. Multiplying this number by 0.05 gives us that the traffic charge when sending an SMS text message next door is 383,000 US dollars per gigabyte.
(UPDATE: As pointed out in the comments, an SMS text message is 160 characters but 140 bytes, so the cost per gigabyte comes even higher – it was originally and erroneously stated as 336,000 USD per gigabyte based on a 160-byte count. Downstream numbers adjusted accordingly.)
So the situation really is as insane in terms of telco industry overcharging as was suggested: the telcos charge more, kilobyte by kilobyte, to send a text message next door, than it costs to send data from Mars – even when including the cost of the actual Mars probe, its launch, and NASA personnel for ten years!
This is an abysmal failure of free market forces to converge the end price with the cost of production. (To add insult to injury, the practical cost of production is zero in this case, as SMS text messages are sent on unused slots in the wireless control channel.) Just for the sake of argument, let’s calculate what an SMS text message should cost if priced at net connectivity traffic rates.
We observed yesterday that wireline traffic costs at most 0.25 cents per gigabyte to the end-user. Fitting 7.67 million text messages into a quarter of a cent comes down to a production cost of at most 33 nanocents per text message, compared to the five typical cents charged. (Over-the-air traffic costs slightly more, but again, this travels in the control channel.)
Concluding, how do these numbers translate into actual telco profit markup on text service, to evaluate market efficiency? We recall that the profit markup on data roaming is an unbelievable 1,400,000%, and that a healthy, functioning market can sustain profit markups of 5% to 10% in the face of active competition. Comparing 33 nanocents to 5 cents gives us a mind-boggling telco industry profit markup exceeding 15,000,000,000% – fifteen billion per cent – on sending text messages next door. When SMS text messages are sent next state, which are charged at about 50 cents, the profit markup exceeds one hundred billion per cent.
This is not just an abysmal market failure, it is more than that. It is a personal insult to us who have publicly defended the belief of the workings of a free market, this being incontrovertible hard data that the telco industry needs to be politically restrained with a complete absence of humor.
For there are already next-generation companies who will have the side effect of killing the entire telco industry by providing the same services free of charge. From the Estonian Skype (voice services), via the American Google Fiber (wireline net connectivity), to the Spanish Fon (wireless connectivity), these companies are proven sustainable and can provide the entire telco offering free of charge as a baseline service in a completely viable next-generation business model.
No wonder the telco industry is trying its damndest to prevent this generational shift from happening. It’s going to wipe them out. There’s no room for a billion-percent profit markup if your competitor provides a better service for free, and especially so if that competitor will enable a whole suite of new products, services, and jobs that go way beyond the thinking of your own industry.
So it is bad policymaking – no, terrible policymaking – to allow the incumbent telco industries to prevent this economic growth (and their own replacement) by giving them any kind of control of the Internet, like with the upcoming coup attempt at the ITU’s meeting in Dubai, or trusting them politically with Internet rollout.
(End note: With a free market, we are referring to a non-discriminatory market that is optimized for low barriers to entry, and where end prices converge towards cost because of functioning competition. We are not referring to an unregulated market where no rules are enforced by governments, for the simple reason that no such market exists on planet Earth. The comments derailed somewhat around this subject on the last article. As these articles cover important subjects for future policymaking, and they are read in Brussels, please stay on the topic of information policy.)
About The Author: Rick Falkvinge
Rick is the founder of the first Pirate Party and is a political evangelist, traveling around Europe and the world to talk and write about ideas of a sensible information policy. He has a tech entrepreneur background and loves whisky.
I remember reading about the insane per megabyte cost of SMS’s a year or two back in Slashdot.
The next question after pointing out that the free market failed in this case would be to examine how this is possible. According to classic market theory, someone would capitalize on this ridiculously high profit margins by setting up a competing communication infrastructure and then grab market shares by charging less than his competitors for using it. That should be simple, considering how low the actual costs are. He would also be able to pay an extreme interest rate on the loans he had been granted to set up his business which means that the aquisition of money for such an investion should be possible. Still, he would make a fortune if you consider how much money is spent on over-the-air-communication.
So why doesn’t this happen? I would really like to read something on why this is such a closed market and what can be done to resolve this madness (and maybe what pirate party affiliates are doing against it).
Laying down fiberoptic trunks and putting up towers is expensive — not locally, perhaps, but people expect their carriers to also get a signal in the next state over, so you’d need a huge investment just to get started. (Contrast with Google Fiber, which is essentially a bunch of local wired connections.)
Except this is quite specifically about SMS text messaging, which appeared as part of a minor upgrade to the system and was never even factored into infrastructure investments as they were planned and made.
In general, your argument is valid. However, not in this specific case.
It still is, actually. The point was that to offer for example, a USA wide network over which to send text messages, to say nothing of voice or internet connectivity would require a massive initial expenditure. Besides that, government enforced monopolies make it effectively impossible to offer such a service even if a company wanted to compete in that market.
I would appreciate a piece on the way the telecoms and cable industries promised to have most of America running on fiber optic lines by the year 2000 in exchange for the FCC allowing mergers and loose regulations. Promises that they never fulfilled, as most of the US’s home consumers are still connected by copper wire.
As part of the deal, they were paid money by the government to finance these fiber optic lines that never appeared. They were also given government mandated monopolies as repayment for agreeing to lay said fiber. Today, they still enjoy monopolies or duopolies in most markets, pocketed the money for infrastructure improvements, and complain that Netflix and Bittorrent are killing them by actually using the bandwidth that people pay for.
Falkvinge, you area hack and a buffoon.
And your solution would be government run telcos?
There are 3 main factors:
1) The usable part of electromagnetic spectrum is limited in size and the range reserved for public phone networks is already sold out to big telcos. If you can’t get your own frequency range, you can’t provide any service to customers either.
2) Metcalfe’s law. The usefulness of a telecommunication network is proportional to the square of the number of connected users. You have no choice but to connect to big telcos’ networks because early on the vast majority of your users will make calls out of your own network. And big telcos will of course charge you for these calls. Unless you have tons of cash to subsidize outgoing connections, your most important rates early on won’t be that much different from big telcos’.
3) Market inertia. The biggest enemy of “perfect” is “good enough”. Everybody has a mobile phone now and while the rates are ridiculously high compared to actual costs, they’re also low enough so that most people won’t care about saving a hundred bucks a month or so. Saving so little is simply not worth the extra effort. Roaming rates on the other hand can stay so ridiculously high (compared to normal rates) because relatively few people travel abroad. Roaming is a niche market that has little to no effect on the main telco market.
This does seem strange! Until you realize that the telecoms market is NOT free at all. I am guessing that “Government intervention failure” as an article title would not get as much debate!
The free market actually works because it does offer alternatives – relatively unregulated (and free) services from Google, Skype, Whatsapp, etc.
The problem here is not business but rather government.
There are a few alternatives that route SMS through the much cheaper standard data channel on your phone if possible; Apple’s iMessage system is probably the most popular.
That said, I think such solutions haven’t been all that attractive to the market because at a fair price, the entire world SMS market is pocket change.
the only people to blame are politicians! they allow industries to produce their own monopolies. take away the monies paid out in lobbying and back-handers (prime example is what has just happened in Brazil, where a new law that received world-wide praise from multiple sources has now been scrapped because of the lobbying done by the monopolies already in place that refused to join the digital era!), the politicians may be less inclined to back certain industries in being able to keep progress locked down so as to maintain their firm grip on what is profitable to them. it is exactly the same principle as the entertainment industries. get the politicians to continuously ramp up old laws or introduce new laws that keep things done the way they are done now rather than encouraging new ways, new business models new inventions and progress itself not only grinds to a halt but actually starts to regress. these industries that refuse to progress have very conveniently forgotten, ignored or both, that they are only in the position they are now in because of the progress that occurred prior to them getting what they now have. talk about selfish little kids. ‘if you dont play my game, i’m taking the ball home’. shame they dont take it to hell and stay there!
[...] in series: Telcos charge more for sending a text message next door than the cost of sending the same message fr… You've read the whole article. Why not subscribe to the RSS flow using your favorite reader, or [...]
Just a side note, a SMS message is not 160 bytes but only 140 bytes (up from the previous limit of 128B). It can fit 160 characters only because of a special lightweight encoding.
That makes SMS even more expensive :)
See: http://en.wikipedia.org/wiki/Short_Message_Service#Message_size
Thanks – will update.
(Edit:) Good thing I updated, I also noticed that I had calculated the production cost of an SMS message wrong – I had calculated the per-gigabyte traffic cost as 0.25 euros rather than 0.25 cents, getting the profit markup two magnitudes too low. These ridiculous numbers make you doubt your math twice and thrice…
Try Australian SMS charges.. We’re normally paying about $0.30 -> $0.55. Majority of the time it’s the latter.
Well, the practical cost of production of course isn’t zero. They have to build this cell phone network to begin with. And they have to operate it, paying electrical bills and salaries to tech personell and customer service reps and the sales force and the billing department.
And you really can’t compare prices that way, mixing wholesale with retail. I mean, the cost of x units of something, loaded on a pallet in the factory’s basement isn’t the same as the total cost of x units individually shrink wrapped and sold piece by piece in kiosks all over the country.
How relevant is cost of production anyway? In a free market, the price is what the customers are willing to pay, right?
By the way, I pay zero extra for SMS. I have, like, a thousand or so “free SMS” per month included in my price plan. A fixed mothly fee gives me “free” traffic (voice/data/SMS) up to a limit that I never reach. So this 5¢ figure doesn’t apply to me.
In the general case, you would be right, but not in the case of SMS text messaging.
This particular functionality was enabled by a minor upgrade as an afterthought into the system, and revenues from this service were never considered in return-on-investment calculations when building the infrastructure.
Normally, you would be able to point at the cost of building the supporting infrastructure (“because otherwise you wouldn’t have the infrastructure in the first place”), but in this special case, that does not apply.
Second, your statement…
…is only true from a narrow corporate perspective, but not from a policymaking perspective. For policymakers, a functioning market also requires that there is functioning competition, which can be observed by seeing end price converge with production cost to within 5% to 10%. When you have profit markups in excess of one billion percent, there is not a free or functioning market.
This is also why we raid cartels at dawn and haul their asses of to jail, even if they price services at a point where somebody is willing to buy – for the cartel would be holding back the enabling of next-generation products and services through an inflated price point, hurting the overall competitiveness of the region.
Cheers,
Rick
I don’t see the “SMS is an afterthought” argument as particulary valid today, a couple of decades later. The companies have kept on investing in the networks during all this time, taking SMS revenues into account. Revenue streams have grown and shrunk according to usage trends, and if voice can’t support the costs anymore, data and SMS step in. I don’t see how the technical implementation, using “free space” in the protocol, is relevant for the pricing.
It is a service, built on top of a network, that costs vast amounts of money to build and to run day by day, and as Andreas Fink points out in comment #7, there is also the cost of frequency spectrum in most countries. You really can’t separate one service from the cost of running the network, and say it has zero cost.
The points that next_ghost make in comment #2.2 are also very important.
“free market failure”? .. i think the Telcos would disagree.
While I agree on the heavily overpriced SMS, the price calculation is not really comparable. If you talk about mars to earth, you have one single point to point connection. So make a big dish point it to earth and you’re more or less done. In terms of SMS dealing with millions and millions of subscribers, you need to have a big network with many hops and nodes. And every of those nodes needs to process protocols which were developed in the 1980ies which are complex and CPU intensive to process. If you compare it with IP which is packet switched where you have millions of routers in the field and there are thousands of competitors putting the market down, the GSM infrastructure was dominated by giants who charged a fortune and there where only a handful of GSM infrastructure providers, even today. If SMS would simply be an IP packet service, the thing would change but for historical reasons they are not.
And one big cost has been totally forgotton in the comparison. The frequency spectrum license fees. In Germany those where auctioned off by the state for over a billion euros at the time. So the Telco’s have to invest into a expensive infrastructure and then pay billions for the license fees. No wonder they do want to make a return of investment in some point of time. So its unfair to compare those numbers with highest speed fiber infrastructure which can transmit billions of bits per seconds. Transmitting over radio in open space is way more expensive. Don’t forget you can not take your fiber connection mobile in your pocket.
Never the less, with the right infrastructure, SMS can be transmitted internationally at maybe 2/100′s of a cent. The rest of the cost is what the mobile operator puts into his pocket as termination fee and this can be as high as 0.06 euro’s (approx 0.08 USD). if you take those figures you are in a realistic area. The problem is that because of the limited spectrum, every mobile operator set his own termination price and there is no incentive to bring it down as this is not the cost his customers are paying, this is the cost others are paying when trying to send a message to the operator’s customer. So there is no competition there as the operator can dictate the prices.
The solution is to use other alternative messaging services such as Global.AQ, What’s App, Skype, MSN, Yahoo, AIM, ICQ etc. The disadvantage is that those are not as universal as SMS as the other side needs to have the same system. But after all, its up to the end user if he is willing to pay for a SMS he sends or not. And as long as everbody does, it will stay expensive as it is today.
I did a similar calculation in 2011, but used Cassini-Huygens which is operating around Saturn. At the time, we in Canada were paying 15 cents per message.
http://www.stargard.ca/SMS%20by%20the%20numbers.html
One thing I should point out is although the maximum amount of data that an SMS message can send is 140 bytes, the average message size is really only about 30 bytes. You know, “Where are you?” “Pick up milk.” “Wanna hang?” and the like. But it costs the same.
I calculated that the data from Saturn costs $274,815 per GiB, so the numbers are certainly similar.
Luckily, roaming texts now only cost 10 cents in Canada. See here, for an example http://mobilicity.ca/coverage-map/. Mobilicity uses Rogers as their roaming partner.
Have you started talking in majestic plural, Rick? :-þ
Well, we’re a team running this site now (a score of guest writers and some 60 volunteer translators in addition to myself). I thought it about time to start acting the part.
You could argue it’s a success of the free market. Telcos took free space in the signalling protocol, which had no real value, and turned it into a cash cow by creating a service that millions of people use everyday.
http://everythingsysadmin.com/2008/12/how-sms-works.html
It’s certainly a success for the individual business to discover additional revenue streams after all the investments have been made – but to talk of a free market success, competition must work and price must converge with cost.
With 15,000,000,000% profit markups, that’s obviously not happening.
Be careful what you wish for. The day the price of an SMS drops to zero, every scumbag and scammer on the planet will spam the fuck out if it and it becomes worthless as a way to communicate for everyone else. Or at least until we figure out some anti-spam mechanism that works.
Smartphones already have enough capability to run simple spam filters (which only need to be run against messages from numbers which you have not contacted and which aren’t in your corporate phonebook if it is a work phone).
Also, with SMS you have a paper trail for billing purposes, so it is possible to prosecute spammers (or their customers) if they are in a country with “do not call” lists, or to black-hole SMS messages from spam-freindly carriers (as was once common on UUCP-net with email and Usenet)
Sounds liek a plan to me dude.
http://www.IP-Fake.tk
[...] blogueur Falkvinge a écrit des dossiers sur les Telcos très intéressants. Il met en avant les différentes [...]
The greatest weakness of the free market is the will of some companies to evade its restrictions by means other than competition (such means usually involving the government). It’s not so much as a failure of free markets as it is a failure of the industry to actually BE a free market.
[...] Daten senden vom Mars billiger als SMS (englischer Text) [...]
Except that the price you are charged for SMSs isn’t really just allocated to the cost of sending SMSs. Segmenting out a highly desirable feature, charging extra for it, and using that revenue to subsidize other portions of a product is a very common pricing mechanism. In fact the fact that others pay more for SMS will allow a network operator to charge less for those who don’t want SMS.
This is not a market failure. Operators have a specific revenue target per user based on all of their costs. Were there to be some spate of competitive SMS price wars, or a national decree that all SMS should be free, the costs would merely be allocated elsewhere on the bill, by either increasing the average base rate or finding some new creative way to tack on extra fees.
You really just have to view your bill as a whole and stop obsessing about the break down. The mechanism by which a bill is broken out into ala carte options has a lot more to do with human psychology and marketing 101 than it has to do with the cost of providing those options.
Sorry Rick, I disagree with your calculations! How would you calculate based upon a carrier providing unlimited text, voice and talk, email, etc. for $50 plans, which are available?
One could make a example of a $7 bottle of water to argue that the utility industry has failed.
This entire article is based on a false premise – that it costs everyone 5 cents to send a text message. Sure for a few people who refuse to subscribe to any sort of monthly plan it costs 5 cents a message. Every single carrier offers unlimited texting as part of any post paid plan or for $5 extra for any prepaid plan.
So based on the fact that unlimited texting is included as part of my Sprint Simply Everything plan, or as part of my daughter’s $30 a month t-mobile prepaid plan, I too can quote numbers to prove that the free marketing has succeeded because texting costs as little as I want (by inflating the number of texts I send under the unlimited text plan).
This article is a prime example of people stirring outrage over absolutely nothing based on a completely false premise.
[...] This is the apex in this series of the dysfunctional telecom market, giving a background to why the telecom industry wants control of the Internet so badly, and is using every conceivable resource to stall, prevent, and delay its resulting economic… [...]
Classic mistake. Prices have very little to do with cost of a good or service and everything to do with what people are willing to pay for the good or service. There are undoubtedly some market distortions here, but it hardly constitutes a market failure just because people are willing to pay a good deal for something that is cheap to produce. That happens all the time.
Rick,
Being an evangelist does not excuse you from getting your facts and your analysis wrong.
First, lets’s start with the basics of SMS. It was designed into GSM, the precursor of most of the world’s current digital cellular systems from the outset.
The expectation, however, was that usage would be very low; the parallel was the then current paging networks. Projections envisaged an upper limit of a few messages a day at most.
Those earliest 2G networks did not have data communications capabilities. There was *voice* communications, and there was *control* communications.
The primary purpose of the *control* communications channels was to set up, handover and tear down voice calls.
It was recognized that these *control* channels were inherently *data* channels that could also be used to transport messages.
It was NOT an upgrade as an afterthought.
Moreover, it was an extraordinarily expensive way to transport *data* because it all went through the *control* system; it was transported over SS7, which ran over X25. Enormously inefficient, ridiculously expensive, and *NOT* designed to transport data. Yes, it used unused control channel slots over the air interface, but once it reached a BTS, it transited through the BSC on the control side, not the transport side.
So, you’re wrong on the facts about its origin, and you’re wrong on the facts about its relative cost. It was designed in at the outset, but on the presumption of very low levels of usage, and it was relatively speaking, extraordinarily expensive to carry.
Which brings me to your major fallacy, and your seeming complete failure to grasp Economics 101 when it comes to fixed costs and marginal costs.
I’ll make this simple.
Some things (such as digital cellular networks) have relatively high fixed costs (what it costs to operate them, independent of the level of usage) and relatively low marginal costs (what each additional unit of usage costs). In fact, most of the time, most cellular usage incurs zero marginal costs. What happens instead, is that when the network gets busy, the quality of service degrades, so over the *medium* term, the marginal cost of capacity takes into account the need to expand capacity to support that service at an adequate level of quality.
The *price* for the service, has to cover the *marginal* costs AND over the long term, otherwise the network doesn’t get built, because it doesn’t cover the cost of providing coverage (really expensive) and the requisite capacity to deliver a reasonable quality of service (in general a much lower cost of service).
Now, for a good (such as cellular service) which has more than one component, for each of which the marginal cost is low, the optimal economic approach is to price them depending on the willingness of people to pay. Strictly speaking, the marginal *price* should be in inverse proportion to the own-price elasticity of the good for each relevant segment.
In simple terms, the price for SMS reflects (1) the need to cover (with other services) the total costs of the network, for *coverage* (expensive) and capacity, and (2) people’s willingness to pay for SMS relative to other goods (such as voice).
SMS costs a few cents, because people have valued it that highly, relative to other goods, such as voice communications and data communications.
Moreover, FWIW, the relative cost of SMS is much higher than for any other form of digital cellular communication, precisely because it uses the (relatively *much* more) expensive *control* (or switching) infrastructure. Still low, relative to the price, but *much* higher than for voice, which in turn is *much* higher than for data.
Your analysis is completely, totally and utterly wrong.
Happy to talk more….
Respect your many contributions, and look forward to your publishing a retraction which reflects the reality…
First, you are confused if you think the telecom duopoly here in the US represents the free market. It’s not been an unregulated thing for well over a century. Second, the market is working – the volume of text messaging actually dropped during the last year, as alternative services (Facetime, Skype, FB messaging, etc) started to take market share away.