The path to abundance: making data transfers available, affordable and sustainable

I was thinking, shouldn’t it be a fundamental human right to have the means of asking a question and get a quality answer in real time (based on first principles and evidence from credible sources) for free (think oxygen free)?

This guy has more global connectivity than John F Kennedy did when he was President of USA! Why can’t he have access to quality answers on a real time basis?

The Internet is kind of free right? Sufficient items of un equal value have been traded by very smart people to consistently deliver a whole that is exponentially bigger than the sum of its parts. The Internet is free (kind of) and it’s awesome (for some of us). But transferring data over wireless networks is not free (for now), and that is one reason why transferring data over wireless networks is not as awesome as the Internet itself.

Is there a financially feasible solution available per which a data transfer speed of 1 MB/second for a total quantum of 1 GB/month is available to anyone with access to handheld device for free? Maybe not today when the world is consuming 3.7 exabytes / month, and telecom operators have been forced to pay exorbitant sums of money to buy licenses, but could it become possible when the world starts consuming say a 300 exabytes / month? I think yes because at such high data consumption levels, the kind of free pool I’m speaking of will be a mere rounding error, but will provide the much needed basic necessity I speak of.

Cisco thinks the world will be consuming 30 exabytes per month by 2020, generally the world has been wrong about forecasts particularly related to data consumption (for e.g. a 2011 UMTS Forum report says, 10.5 exabytes by 2020).

What assumptions if changed would alter that forecast upwards? That becomes a difficult discussion but I heard a very compelling thought recently that may result in a more accurate forecast:

The best way to predict the future is to create it.

So why not figure out a way to rapidly and consciously accelerate global data consumption? 300 exabytes per month by 2020 sounds like a good target to me.

I believe, there has got to be a way of making data transfers over wireless networks free-ish, (at certain levels) while ensuring the people who are worried about this development are also satisfied, because some of them think that data transfers becoming free-ish over wireless networks is a bad outcome for the telecom industry as a whole. Bad means job losses for a few and maybe some angry investors as well because a lot of money has been spent on spectrum and telecom equipment.

I want to highlight though that bad does not equate fully with the impact on consumers or the world as a whole, which I think will be a massive positive. The relationship between GDP growth or GDP / capita and broadband penetration is well known. Increasing overall data consumption will make an increase in broadband penetration more viable in my opinion.

Ultimately the big elephant in the room which needs to be discussed is whether data ARPUs (Average Revenue per User) will suffer the same fate over time as voice ARPUs did? I think yes, because providing a pipe to transport data is a commoditised service, ultimately Mobile Network Operators (MNOs) will compete on price if their business models don’t change substantially from today.

Also, technological progress follows Moore’s law and considering the exponential nature of many technologies moves the world form scarcity to abundance (think data storage in 2001 versus today), data transfers will surely become free-ish in our lifetime, maybe within the next 10 years. I don’t think the MNOs are quite ready for this conversation because they have spent a whole lot of money buying equipment and licenses. I don’t blame the MNOs because they haven’t seen returns that justify cost of capital in quite some time either.

One way of making data transfers over wireless networks a LOT more cost effective while simultaneously reducing the cost of increasing broadband penetration is by compelling telecom operators to share infrastructure.

Here is a very rough analysis

1. On average a 60 meter telecom tower costs about $80k, active equipment is another $40k or so. This tower can accommodate up to 4 operators (if they shared the tower), but in quite a few markets, operators don’t share towers because of baseless competitive fears.
2. If 4 operators share the cost of putting up a telecom tower only, there’s a 60k / tower saving from the get go and the running costs are just another added bonus.
3. Take this a step further. If telecom operators shared ALL their assets, increasing coverage and improving quality may become possible while keeping the whole process sustainable for shareholders of telecom companies. (many telco shareholders have not seen dividend flow in a very long time in fairly mature markets even — Pakistan is a case in point).
4. This case becomes particularly compelling when one recognises that 10 pct of the sites (usually in urban areas) generate half the revenue. Therefore catering to the rest of the population doesn’t seem worth it financially. I believe this tail becomes a lot prettier if asset sharing (not just tower sharing) becomes a norm. Note that this tail is a lot uglier in typically low ARPU frontier markets which are generally big population centres as well (Pakistan, Bangladesh, Indonesia are all examples)

At a very macro level the challenge I see is that the telecom industry as a whole significantly overpaid for infrastructure and spectrum licenses. This over payment and over construction of infrastructure has now left behind very high legacy costs. The shareholders are upset, the management teams of MNOs are not quite sure what the future holds because the foundations aren’t sustainable due to high legacy costs. The rationale for sharing seems obvious, but causes understandable heart ache to the captains of telecom industry because they understand that the first operator who starts sharing will be the most better off by a mile, but they are also afraid of doing a bad deal or doing a deal with someone who can’t take care of their network.

The vendors and equipment suppliers like Ericcson and Huawei are struggling as well to direct R&D spend or develop a clear outlook because of the uncertainties faced by their customers. 5G won’t become a reality in my view unless the overall telecom system becomes a lot leaner and utilisation levels improve. Some vendors start justifying negative margin deals as a result just to stay alive.

Are the customers happy? The answer varies substantially between the haves and the have nots. Most recently companies like Facebook have made a noble effort to provide free services limited to their own properties. That concept did not fly in India. The opposing arguments are quite understandable as well. However the intent is awesome here. Make Internet access truly free for all.

I think eventually (in 10 years i.e.) all participants of the telecom value chain will realise that a whole lot of argument and negotiations was for nothing but rounding errors.

My plea is that we can make a bare minimum level of data transfer speed free provided we release some obvious choke points in networks by making infrastructure sharing viable for all stakeholders.

To achieve the desired end state, I think we need a frank and open conversation which goes something like this:

1. Dear shareholders of MNOs, you will not see dividend flows until your cost base changes, that won’t happen till you share infrastructure, your infrastructure is still valuable today so please start sharing it soon because delaying decisions to eek out the last few dollars from asset sales will make your infrastructure worth a lot less (last ones to share won’t be able sell their assets). Please bring an independent company to supervise the sharing because your competitive streak will keep you from achieving your goals. Don’t stop at sharing just your passive infrastructure, go beyond, all the way down to spectrum.
2. Dear executives of telecom companies, we understand your pain point and can imagine that your board meetings aren’t a lot of fun. A lot of you have been involved in a race to the bottom, you have continued to make investment cases based on survival than on returns. We feel your pain and want you to know that data ARPUs will likely suffer the same fate as voice ARPUs. That’s because data transfers are highly commoditised as a service and if you aren’t prepared to disrupt your services I can promise you someone else will. Imagine a world where a drone like this does more than take pictures and starts replicating networks or Skybender succeeds or…lots of other disruptive possibilities. Wouldn’t it make sense to just disrupt your own business? Especially if it means you increase the chances of saving our planet?
3. Dear telecom equipment providers, we understand your challenges associated with directing R&D spend when you can’t get much visibility from mobile network operators. It’s not their fault though. Ultimately your business can’t be about selling boxes either, so maybe it’s about time we had a chat about how we can collaborate so long as we agree that making data transfers free-ish is a worthwhile exercise?
4. Dear high ARPU customers of MNOs, you contribute to majority of the revenues. I’m sure many of you are concerned about the vast plethora of challenges ranging from climate change to food security to water shortages. Maybe you can become part of the long term solution discussion as well? I don’t see why not. Could you possibly subsidise this effort a bit? You already do a lot but I think it might be a good idea to involve you in this conversation. I wonder if the data plan you “pay for” is fully utilised? If not then would you mind sharing the excess bandwidth you already pay for provided your web surfing experience is not compromised?
5. Dear investors and financiers, 5–10 year financial forecasts predicting internal rates of return are guidelines at best. Please consider backing a vision and a commitment to execute on a plan as opposed to an easily manipulatable excel spreadsheet that makes a business case hold.

Why do I care about all this?

I am an early stage investor in a company called Towershare. A company that is trying to make data transfers over wireless networks more available and affordable by addressing an urgent and obvious need — the need to share passive infrastructure by telecom operators because they simply overbuilt networks in the first place (because they never thought sharing was ever an option when networks were first being built) and can now realise substantial savings by sharing infrastructure ASAP.

At the same time Towershare is trying to ensure that its own investors earn a fair return for the risks undertaken already + risks associated with buying more towers in multiple geographies. Like all companies, Towershare raised capital because it promised a return superior to the likely cost of capital to its investors.

Unfortunately, capital still has a cost because capital isn’t abundant as yet, capital is not free-ish. Eventually it is capital that we must set free. Making data and electricity free are probably the first of many steps of making capital free. Making data free will be easier than making electricity free. But I digress, so back to data…

I think data is a force of nature. It has gained a ton of momentum. I don’t see a world where data transfers over wireless networks won’t become free-ish in the next 5 years.

I want to know how can I use the Internet to help me figure out a way to fulfil the financial return expectations of Towershare’s investors + efficiency improvement expectations of mobile network operators (Towershare’s customers) while ensuring sufficient contribution are made towards making data transfers over wireless networks affordable (free-ish) and widely available? Is 300 exabytes of monthly consumption by 2020 a meaningful target?

It’s either that or somebody help me make Bitcoin mainstream enough so that data consumers can just make a deal with me directly and I won’t even have to bother satisfying this construct known as cost of using fiat currency or continue to remain subservient to the arcane concept of “value chain” or go through bundles of regulations because governments are afraid of “foreign currency” leaving their economy. Maybe in 2019 this would have been more feasible, too soon for that today. I digress yet again.

Do keep in mind that the marginal cost of making data transfers free using existing infrastructure is quite low, almost zero in many cases. We have sort of already paid for the infrastructure and just need to re-adjust it a bit, result will be de-bottlenecking. Pakistan for example has 33,000 or so towers, all 5 operators can provide spectacular coverage using only 18,000 towers if they just shared infrastructure. I have a hypothesis:

Making data transfers over wireless networks free-ish may be a lot easier than you think and may simply require telecom operators to share all of their physical assets (including spectrum). This is called Active Sharing.

I believe a fully shared network (i.e. active + passive) won’t be bad for the telecom industry executives or investors, because such actions may make it easier to increase global broad band penetration while protecting the financial return expectations of MNO shareholders. Such actions in my view will help us get closer to the 300 exabyte consumption per month goal. A goal which provides sufficient economies of scale to make a certain quantum of data transfer at a certain speed free!

Keep in mind that about 4 billion people are still off-line! Sharing an excerpt here from an ITU report that came out in September:

“Released today just ahead of the forthcoming SDG Summit in New York and the parallel meeting of the Broadband Commission for Sustainable Development on September 26, the report reveals that 57% of the world’s people remain offline and unable to take advantage of the enormous economic and social benefits the Internet can offer.”

We are underestimating the benefits that come with making data transfers free and abundant (like sunlight) to those who can put such data to good use (possibly save the world even)

I think making data transfers over wireless networks free will enable the telecom industry to sell more meaningful products and services than commodities disguised as differentiated brands.

We are underestimating the benefits we could gain by connecting with those people who we have never spoken to. Also, we are running out of time…

I mean as a species we only have 1 degree Celsius of temperature increase to go, after which the earth is in some serious shit. That 1 degree increase may come within the next 40 years under a very BROAD and likely set of assumptions.

The best brains of the human race have some ideas on how to reverse the climate change trend but we will need many more people pulling in the same direction. We still haven’t figured out how to use the atmosphere’s CO2 to generate power or for that matter what to do with the atmosphere’s CO2?. One of the things which stand in the way is the price paid by consumers to transfer data over wireless networks.

Maybe one thing we can do is engage in a dialogue on how the telecom tower sharing industry can play a role in making data transfers over wireless networks free-ish?

Why telecom tower sharing as one of the many possible starting points? It’s a simpler obvious roadblock to overcome and it’s mutually beneficial to a lot of telecom industry folk, plus investors like it too. But of course the captains of telecom+investment industry worry about over payment/ under payment / looking bad / being unsure if the decision makes sense because the world is becoming harder and harder to explain.

For me, I don’t see a world where tower sharing won’t happen in the Middle East South Asia plus Central Asia region. Pakistan for example offers the lowest speed in 3G service. Quickly doing a passive network sharing deal will improve the average network speed without increasing consumer costs (all else being equal). It may also give the telecom equipment suppliers a lot more clarity on where to direct R&D. I’m not dogmatic about saying that tower sharing is the only essential first step, I’m saying its one of the many and is definitely essential because it’s benefits are just so glaringly obvious.

There’s a well known chart that plots HDI with electricity consumption pasted below, I have tried looking for a similar one with data consumption on the horizontal axis, but haven’t found one yet, so I’m hypothesising that 1 MB/second for free is a good enough speed and 1 GB/month is a good enough volume for humanity to get most basic questions answered from credible sources while curiosity is at peak levels.

This last element is super important because learning happens only when curiosity is at peak levels. Just think about all the facts you remember today after hearing about them only once (what’s so special about those facts? You learned them when you were SUPER curious and got the answer in that moment)

I need the same graph but Data consumption in MBPS on the horizontal please!

I couldn’t have imagined a world where a small little device like the smart phone could have educated me perpetually. I couldn’t have imagined a world where geographic barriers were so easily breached. The Internet has made my life a LOT better. I’m grateful to the people who made this happen. I think it would be a travesty if we don’t make their mission forward immediately. As We May Think is an article everyone must read. The second last paragraph in particular is so relevant:

Presumably man’s spirit should be elevated if he can better review his shady past and analyze more completely and objectively his present problems. He has built a civilization so complex that he needs to mechanize his records more fully if he is to push his experiment to its logical conclusion and not merely become bogged down part way there by overtaxing his limited memory. His excursions may be more enjoyable if he can reacquire the privilege of forgetting the manifold things he does not need to have immediately at hand, with some assurance that he can find them again if they prove important.

Mechanisation of records has brought us very far, it’s time we made accessing those records a whole lot easier.

Money is raised when IRRs seem likely or so we are told. Because we still have decision makers in powerful places who mostly care about projected cash flows and returns and nearly not enough about the long game. The game that involves feeding 9 billion people sustainably. Or reducing CO2 from the atmosphere before climate change wipes us out. Stuff that’s likely to come due in the next 4–5 decades.

In my view though, IRR promises are more about fear and less about greed. I’ve seen most decision makers are afraid of doing bad deals. Mostly afraid of perception. It’s a matter of time though, until the power shifts from houses of capital to houses of talent. I think houses of talent driven by purpose and meaning will be less affected by fears of perception than houses of capital. Because houses of capital are insecure, as they ought to be, the trends aren’t favourable for businesses sitting on capital, unsure what to spend on. And no, looking cool or traditional self interest is not a worthy cause, it won’t get enough people galvanised. No Massive Transformational Purpose = low probability of success.

Attributes of an Exponential Organisation

Talent will trump fiat currency or the traditional notion of what we call financial capital. It’s just a matter of time. But first we must level the playing field. Free data transfers over wireless networks at legit speeds can level the playing field. And if all participants are willing to share the upfront cost, the benefits will more than compensate.

With so much demand for data, isn’t it time the telecom tower sharing industry goes beyond tenancy and power cost savings and focuses instead on providing the best shared infrastructure solutions that makes data transfers over wireless networks so unbelievably cheap that they are virtually free without harming the financial and strategic interests of the telecom industry as a whole?

I think it’s doable. I think at a minimum it is worth having a serious conversation.

– — — -