I’m going to go through the piece in order for thoroughness. First, the opening paragraph:

Earlier this month, Apple rejected an application for the iPhone called Google Voice. The uproar set off a chain of events—Google’s CEO Eric Schmidt resigning from Apple’s board, and the Federal Communications Commission (FCC) investigating wireless open access and handset exclusivity—that may finally end the 135-year-old Alexander Graham Bell era. It’s about time.

First, it’s not clear that Eric Schmidt resigned from the Apple board over the rejection of the Google Voice app – the small matter of an FTC investigation may have been a factor. In addition, there’s the fact that the two companies compete in the email, browser and now operating system markets, which meant that Schmidt was having to recuse himself from more and more discussions on the board anyway. Secondly, the FCC is very specifically investigating this particular case, not wireless open access or handset exclusivity in general, as a result of the Google Voice incident.

I have no real beef with Kessler’s next paragraph:

With Google Voice, you have one Google phone number that callers use to reach you, and you pick up whichever phone—office, home or cellular—rings. You can screen calls, listen in before answering, record calls, read transcripts of your voicemails, and do free conference calls. Domestic calls and texting are free, and international calls to Europe are two cents a minute. In other words, a unified voice system, something a real phone company should have offered years ago.

Except perhaps to say that a “unified voice system” is a bit strong – this “system” is entirely dependent on the existing phone systems offered by the “real phone companies”. Having said that, Google (or, more accurately, Grand Central) certainly provided some useful innovations and created a great service, and I’m a (small-volume) user of the service myself.

Apple has an exclusive deal with AT&T in the U.S., stirring up rumors that AT&T was the one behind Apple rejecting Google Voice. How could AT&T not object? AT&T clings to the old business of charging for voice calls in minutes. It takes not much more than 10 kilobits per second of data to handle voice. In a world of megabit per-second connections, that’s nothing—hence Google’s proposal to offer voice calls for no cost and heap on features galore.

The first sentence is important: contrary to the headline of the piece (“Why AT&T Killed Google Voice”), there are only rumors at this point, and AT&T has flatly denied that it blocked the app. The FCC investigation is seeking to answer in part the question of who blocked it, so a bit more nuance is probably in order there. But the bigger objection to this paragraph is the assertion that voice is simply a form of data, an idea Kessler returns to again and again throughout the piece. That needs a bit of examination, because it’s misleading.

While there’s a sense in which that’s true – all communication is ultimately “data” – it’s only true in the technical sense if it’s carried that way. Which it isn’t, on today’s cellular networks and most public telephone networks. Other than where voice over IP is used, voice is circuit-switched, which means it ties up an entire (virtual) circuit from end to end for the duration of a call, making it unavailable for other purposes. Data, on the other hand, is typically packet-switched, meaning that a data “connection” in fact only uses up network bandwidth when packets are actually being sent back and forth, otherwise freeing up that bandwidth for the use of others. As such, voice networks and voice calls use network capacity in a very different way from data, with different equipment required and different economics associated with them. It is therefore at best a gross oversimplification to say that “voice is data” in the current mobile market. In time, yes, we’ll evolve to voice carried over the data network, and at that point the statement will be true, but we’re a long way from that point yet.

What this episode really uncovers is that AT&T is dying. AT&T is dragging down the rest of us by overcharging us for voice calls and stifling innovation in a mobile data market critical to the U.S. economy.

For the latest quarter, AT&T reported local voice revenue down 12%, long distance down 15%. With customers unplugging home phones and using flat-rate Internet services for long-distance calls (again, voice is just data), AT&T’s wireline operating income is down 36%. Even in the wireless segment, which grew 10% overall, per-customer voice revenue is down 7%.

I’m not sure this episode uncovers anything about AT&T’s health as a company at all. While I think AT&T may be a little over-reliant on the iPhone (see more on that here), its financial results continue to be impressive. Focusing on the wireline results isn’t really relevant in the context of a piece about the wireless market, but yes, wireless voice revenues per customer are declining, because, erm, average prices are coming down, which rather goes against the grain of the piece. And it’s not clear that whatever action was taken in regard to the Google Voice application in any way stifled innovation in the mobile data market. (on the topic of wireline versus wireless revenues, this recent piece on MSN Money was a lot more insightful, but that’s not really the point here).

Wireless data service is AT&T’s only bright spot, up a whopping 26% per customer. How so? As any parent of teenagers knows, text messages are 20 cents each, or $5,000 per megabyte. After the first month and a $320 bill, we all pony up $10 a month for unlimited texting plans. Same for Internet access. With my iPhone, I pay $30 a month for unlimited data service (actually, one gigabyte per month). Is it worth that? The à la carte price for other not-so-smart phones is $5 per megabyte (one-thousandth of a gigabyte) per month. So we buy monthly plans. Margins in AT&T’s Wireless segment are an embarrassingly high 25%.

I’m not sure wireless data service is the only bright spot, but again the analysis here is odd. What is the point Kessler is trying to make here? That it’s embarrassing AT&T’s wireless margins are 25%? This from a former hedge fund manager? Is he objecting to the fact that AT&T allows you to text all you want for $10 per month when individual messages are 20 cents (in fact, it’s $20 per month)? I can’t tell what he’s objecting to here.

The trick in any communications and media business is to own a pipe between you and your customers so you can charge what you like. Cellphone companies don’t have wired pipes, but by owning spectrum they do have a pipe and pricing power.

That’s an odd statement if ever there was one. The very definition of providing a communication service is giving customers a way to communicate, which has traditionally meant a phone line (whether wireline or mobile). There’s no trick here, and the idea that the only reason for having a pipe to your customers is to gouge them seems an odd summary of the last 135 years of telecoms history. Spectrum isn’t the pipe either – the cellular networks carriers build out to make use of the spectrum are the pipes (or perhaps we should say “series of tubes“?).

Aren’t there phone competitors to knock down the price? Hardly. Verizon Wireless, T-Mobile and others all joined AT&T in bidding huge amounts for wireless spectrum in FCC auctions, some $70-plus billion since the mid-1990s. That all gets passed along to you and me in the form of higher fees and friendly oligopolies that don’t much compete on price. Google Voice is the new competition.

So, the point is there is no competition because the government forced the wireless carriers to spend billions on spectrum? Isn’t this supposed to be a piece about how the government can solve problems in the mobile industry? It’s also not clear how the spectrum fees have created a “friendly oligopoly” (note singular noun – I’m pretty sure once there are several oligopolies it stops being an oligopoly) – scarce spectrum has created the present oligopoly in the wireless market in the US, as it has everywhere else. And no industry wants to engage in price competition, especially if it has large sunk and/or fixed costs, and most would prefer to compete on features (e.g. exclusive phones). If price competition is your goal, you’re fighting a losing battle, for everyone concerned.

But now we come to one of the biggest problems with all the stories about Google Voice – the idea that it is somehow competition for the mobile phone companies when it’s running on their devices. As it currently works, Google Voice works on the basis of hooking up your existing connections (wired, wireless, whatever) to its service, not creating new endpoints. While there are workarounds, the vast majority of users see this as a way of integrating their existing endpoints into a more unified service. Since the US uses an airtime system (i.e. charging for inbound and outbound calls) for billing for mobile voice usage, every domestic Google Voice call that connects to an AT&T iPhone generates exactly the same usage and therefore the same charging as a call originated through the standard iPhone interface. AT&T is still very much in the picture here, and still making money off those calls. All that’s changed is the interface. AT&T (and even Apple) might not like the fact that the system bypasses their interface, but AT&T still gets the money, so this isn’t competition for revenue, just the UI. Others such as Skype and Truphone, which actually provide an alternative client for making calls over VoIP from the iPhone, thereby bypassing the carrier, are the true competition for AT&T in this space, but their apps are still sitting in the iTunes app store.

By the way, Apple also has a pipe—call it a virtual pipe—to customers. Its iTunes music service (now up to one-quarter of all music sales, according to NPD Market Research) works exclusively with iPods and iPhones. The new Palm Pre, another exclusive deal, this time by Verizon Wireless, tricked iTunes into thinking it was an iPod. Apple quickly changed its software to lock the Pre out, and one would expect Apple locking out any Google phone from using iTunes.

I guess this is why the stretched analogy about the pipe was introduced earlier – to somehow equate AT&T’s and Apple’s business models. There’s another factual error here – a pretty glaring one – in that Sprint, and not Verizon, has the exclusive on the Palm Pre. There’s a strange reference to “any Google phone” here too – as if it’s a theoretical possibility rather than a reality in the form of a growing number of Android devices, which as far as I know haven’t attempted to connect to Apple’s proprietary store.

It wouldn’t be so bad if we were just overpaying for our mobile plans. Americans are used to that—see mail, milk and medicine. But it’s inexcusable that new, feature-rich and productive applications like Google Voice are being held back, just to prop up AT&T while we wait for it to transition away from its legacy of voice communications. How many productive apps beyond Google Voice are waiting in the wings?

I’m not sure Kessler has really introduced any evidence on the topic of overpaying for mobile plans, but we’ll let that slide – the examples used seem to have been chosen more for their alliterative potential than for any sensible reason, but we’ll let that slide too. On to the real point here – that blocking Google Voice is somehow propping up AT&T. How so? Are we really “waiting for AT&T to transition away from its legacy of voice communications”? In what way? Didn’t Kessler say earlier that wireless data was the fastest growing area of revenue for AT&T? Isn’t the iPhone primarily about data, rather than voice? Yes, Apple desperately needs to make the approval process for apps quicker and more transparent. But I’m not sure there are lots of apps “waiting in the wings” – what would they be waiting for? There are lists of rejected apps, but many were rejected for good reasons and most later found their way onto the iTunes store in a modified form. If you’re waiting, go ahead and submit your app and see what happens.

So now the FCC and its new Chairman Julius Genachowski are getting involved. Usually this means a set of convoluted rules to make up for past errors in allocating scarce resources that—in the name of “fairness”—end up creating a new mess.

Some might say it is time to rethink our national communications policy. But even that’s obsolete. I’d start with a simple idea. There is no such thing as voice or text or music or TV shows or video. They are all just data.

The FCC is getting involved, in a very limited and specific way – by investigating the case of the Google Voice application in particular. It’s not clear under what authority they’re doing this, but since it’s just a fact-finding mission at this point that doesn’t matter all that much. Since it is just fact-finding, talk of convoluted rules seems a little premature too.

I’m also not sure how “rethinking our national communications policy” can be obsolete – doesn’t that by definition mean replacing whatever may be obsolete about the whole thing? And for all the reasons I explained above, it’s utterly over-simplifying and hugely erroneous to simply treat all voice as data today – the vast majority of voice is not data – it’s voice, with all the quality, cost, infrastructure and other implications that has. But more to the point: data, broadly speaking, isn’t regulated. Voice is. Unless you want to move to an unregulated voice market (which the carriers would love, by the way) or a heavily regulated data market, that’s an unworkable proposition.

Now, on to the specific proposals.

We need a national data policy, and here are four suggestions:

• End phone exclusivity. Any device should work on any network. Data flows freely.

Any device can’t literally work on any network, especially since we have both GSM and CDMA networks (and now also WiMAX networks) in the US. But it is already possible to take any unlocked GSM phone and attach it to the T-Mobile or AT&T networks. Verizon has also built a model for allowing devices to be attached to its network in response to the open access provisions attached to the 700MHz spectrum it acquired (thank you, Google). But that’s a separate issue from exclusivity. Exclusivity certainly has its pros and cons – it allows devices to reach market at lower prices because the carriers are willing to make concessions in return. But it means that, as a consumer, you have to pair networks and devices you otherwise wouldn’t. But this is one of the prices we pay for a relatively unregulated market – we don’t always get what we want, but we do get a lot of good stuff. We have to take the rough with the smooth.

• Transition away from “owning” airwaves. As we’ve seen with license-free bandwidth via Wi-Fi networking, we can share the airwaves without interfering with each other. Let new carriers emerge based on quality of service rather than spectrum owned. Cellphone coverage from huge cell towers will naturally migrate seamlessly into offices and even homes via Wi-Fi networking. No more dropped calls in the bathroom.

I love that Kessler uses WiFi as an example of interference-free networking. Anyone who’s tried to use WiFi in the presence of lots of other WiFi networks (e.g. in any relatively densely populated neighborhood) knows how untrue that is. Carriers have spectrum so they can provide predictable service and be able to make reasonable business plans. Would anyone roll out expensive 3G or 4G networks if they weren’t very sure they’d still have the spectrum to light them up when they were finished? WiFi is terribly suited for replacing cellular networks, which seems to be what Kessler is suggesting, even if it may be useful for extending coverage indoors (although femtocells seem to be the preferred solution for that problem in the US, without any regulatory intervention). And unless you have WiFi in your bathroom, I’m afraid there’s still a chance of dropped calls there.

• End municipal exclusivity deals for cable companies. TV channels are like voice pipes, part of an era that is about to pass. A little competition for cable will help the transition to paying for shows instead of overpaying for little-watched networks. Competition brings de facto network neutrality and open access (if you don’t like one service blocking apps, use another), thus one less set of artificial rules to be gamed.

A rather dramatic change of speed from Google Voice on the iPhone, but I’ll go with it. This has already happened, both at the municipal level and nationally through legislation. Competition for cable is coming both from the telcos (ironically) which are rolling out fiber-based networks, and through online-based offerings such as Hulu, iTunes (ironically) and Amazon. Interestingly, Kessler then employs the argument that competition provides de facto net neutrality, when he doesn’t seem to think the same argument applies with mobile networks (where there is significantly more competition and more choice than there is ever likely to be in local TV provision).

• Encourage faster and faster data connections to our homes and phones. It should more than double every two years. To homes, five megabits today should be 10 megabits in 2011, 25 megabits in 2013 and 100 megabits in 2017. These data-connection speeds are technically doable today, with obsolete voice and video policy holding it back.

I’m not sure how government “encourages” things – it can really only mandate them or ban them. The doubling Kessler talks about has actually happened over the last 15 years without any help from government (speeds available to many consumers have gone from 30kbit/s in 1995 to almost 30Mbit/s today). The highest speeds aren’t uniformly available yet, but they are spreading rapidly and available to more and more households – again, without regulation. Voice policy has nothing to do with the deployment of broadband. And video policy is one of the major drivers behind the rollout of these networks, since the telcos are aggressively rolling out TV services over their new faster networks.

Technology doesn’t wait around, so it’s all going to happen anyway, but it will take longer under today’s rules. A weak economy is not the time to stifle change.

Data is toxic to old communications and media pipes. Instead, data gains value as it hops around in the packets that make up the Internet structure. New services like Twitter don’t need to file with the FCC.

And new features for apps like Google Voice are only limited by the imagination. Mother-in-law location alerts? Video messaging? Whatever. The FCC better not treat AT&T and Verizon like Citigroup, GM and the Post Office. Cellphone operators aren’t too big to fail. Rather, the telecom sector is too important to be allowed to hold back the rest of us.

“It’s all going to happen anyway” is the best possible summary of why a heavy-handed government intervention is the wrong approach. We’ve never justified government intervention on the basis of simply speeding up something that is happening already and there are good reasons for that.

Data isn’t toxic to old pipes – it’s what’s keeping them relevant – old copper lines had new life breathed into them by DSL, and are being supplemented or replaced with fiber connections which are designed to deliver lots and lots of data (yes, including video and even voice). Data doesn’t have any inherent value: it all depends what the content is. And today’s model has consumers or advertisers paying for content on the one hand and ever bigger pipes to stream it through on the other.  Those pipes are provided by telcos, who don’t find them toxic in the slightest. And those new apps? Mother-in-law location alerts? Check. Video messages? Check. The telcos don’t need bailing out – you were arguing, Mr Kessler, just a few paragraphs ago about how high their margins are.

Look, I’m sympathetic to many of the points which are buried in the piece – above all, I don’t like the fact that the Google Voice app was blocked on the iPhone – I’d have used it on mine, and I think it was shortsighted of the AT&T/Apple combo to block it. But I don’t believe that justifies government intervention. It should spur me to either: use the web version (www.google.com/voice), which is what I have done, or find a carrier / device combo that will support it. I don’t believe the US mobile market is perfect as it stands: texting prices for individual messages are probably too high, mobile broadband is patchy, AT&T’s network coverage sometimes stinks, etc. etc. But I don’t believe any of this justifies the kind of intervention Andy Kessler talks about in his piece. Rather, I think “It’s all going to happen anyway” and players like Google can only help it on its way.

Bookmark with:

(Warning – this is going to be a long post – skip to the bottom if you just want to read the conclusions!)

Note: I wrote this post on the evening of 22 July, whereas AT&T’s Q2 numbers were released the morning of the 23rd. A quick look through the Q2 numbers reinforces several of the points made here and doesn’t appear to contradict anything significantly. It appears the iPhone dragged down AT&T’s OIBDA margin by a couple of points – less than last time around, although that reflects the timing of the two launches in the quarters they impacted most strongly.

First, the good stuff

Let’s start with the positive aspects of the iPhone for AT&T. The obvious thing to look at is the impact on customer additions, and since very few (if any) mobile subscribers will choose an iPhone as their first device, the key metric to look at is customer wins from other carriers. Here’s a Morgan Stanley chart from late 2007 – based on their extensive surveys (all the usual caveats apply, but this is a really decent sample size and illustrates what I believe to be real trends):

(click to see a larger version)

They key message here is that after the launch of the iPhone AT&T gained positive ‘flow share’ against Verizon for the first time in the history of their survey – that is, more customers switched from Verizon to AT&T than the other way around. AT&T also gained subscribers from Sprint, but lost them to T-Mobile, which had a very good quarter too.

Morgan Stanley also asked survey respondents what they had been before they were AT&T iPhone customers:

Given that this was the end of Q3 and the iPhone launched just before the end of Q2 2007, one third were already iPhone customers. Another third were existing AT&T customers, and the last third were new to AT&T – so these represented switchers from other carriers, and 12% of AT&T gross adds in the quarter. That’s not an enormous proportion of the total, but it’s certainly a measurable impact.

Of course, we all know that subsequent years / versions have seen much larger numbers of iPhone purchases, and so the effect has been greater. The same Morgan Stanley survey repeated at the end of Q1 had different results – in part because of the much larger installed base (70% of the total), but also seeing a much higher share of gross adds coming from the iPhone:

In both Q4 of 2008 and Q1 of 2009, around a third of AT&T’s total gross adds came from the iPhone, which has clearly become a very important source of wins from other carriers to AT&T. It probably has a little over 8 million iPhone customers on its books at this point.

So that’s gross adds and wins from other customers. Secondly, there’s service revenues. iPhone customers have to sign up for not only a voice plan but also a hefty data plan. Some may sign up for text messaging or additional voice minutes on top of the minimum, but the very least a standalone iPhone customer will pay per month is $70. That’s already quite a bit higher than AT&T’s average ARPU at around $50, and AT&T’s management has stated that it sees ARPU of around 1.6 times higher on average for iPhone customers compared with its other postpaid customers. That’s a major boost to AT&T as well.

Thirdly, there’s the issue of a ‘halo effect’ from the iPhone onto the rest of AT&T’s business, the most prominent symptom of which is foot traffic to AT&T stores. It has stated that it saw significant increases in foot traffic from prior levels once the iPhone launched, and there are undoubtedly customers and potential customers who came in to look at the iPhone and ended up buying something else (something cheaper, in all likelihood!). That’s really hard to quantify, but I don’t doubt that there’s a positive effect there. In addition, AT&T’s exclusive on the iPhone creates a perception of AT&T as a purveyor of cool devices.

To sum up on the good stuff: the iPhone has been a significant source of gross adds and has had a positive impact on flow share, it brings subscribers to AT&T who have a significantly higher ARPU than its other customers, and the device has a halo effect that’s difficult to quantify but no doubt useful as a third source of benefit to AT&T from the arrangement.

Now, the bad stuff

So what about the bad stuff? This falls into several major categories:

• Subsidies and hefty cost of acquisition
• Impact on the network
• Ceding control to Apple
• Being cast as the evil gatekeeper
• Over-reliance on one device.

First, subsidies. This is a murky area but a simple comparison of the subsidised price and the non-subsidised price suggests a $200 subsidy for the 16GB version and a $300 subsidy for the 32GB version. That gels with information various people seem to have pieced together from Apple’s earnings data which suggests Apple gets around $300 from AT&T for every device sold on top of what the customer pays. That’s a hefty subsidy indeed, compared with the $150 subsidy that appears to be AT&T’s standard for other ‘PDAs and smartphones’ and $50 to $100 for cheaper phones. And it basically means AT&T is $300 in the hole from the get-go, and has to put a lot of that high ARPU into paying off that deficit, rather more so than for other devices. Divide that $300 by 24 months (the standard contract length) and you get $12.50 per month of that high ARPU that’s just paying off the device. If AT&T keeps up its current policy of letting higher-ARPU customers get a new subsidised phone every year, that rises to $25 per month. Compare that with around $6 for one of the other high end phones, which basically have the same voice plans and similarly-priced data plans (though not all will take them). That’s the most quantifiable negative impact of the iPhone on AT&T’s business – it amounted to 10 cents of EPS dilution in Q3 2008, for example. The other negatives are less easily quantifiable, but I’d argue they’re at least as important.

Secondly, the impact on the network. This one has been bubbling to the surface recently and I think it’s really significant. A good illustration is a recent PC World 3G network test. In its big analyst event in 2007 and again at subsequent events, AT&T has boasted about its spectrum depth – see this excerpt from a March 2008 presentation, for example:

It specifically states that its advanced spectrum position is a major advantage over competitors because of the superior performance with high network load and better coverage and quality it allows. But that PC World test put AT&T bottom by a long way out of the big three US carriers in reliability and throughput. Its reliability in several major markets was just 60% and in one – Orlando – it was just 52%, where reliability is defined as “the percentage of tests for a given city in which we could detect a signal, connect at a reasonable speed (faster than dial-up), and sustain an uninterrupted connection for the duration of a 1-minute streaming test”. That’s terrible! And it’s significantly worse than either of its two major competitors. Sprint, of course, has a network with plenty of capacity since it upgraded and lost a bunch of customers, while Verizon seems to have managed to keep its network investment running ahead of customer needs. (Sprint marginally edged Verizon in the tests on reliability but it was a close run thing.)

Why is AT&T’s network performing so poorly when its spectrum position was supposed to be such an advantage? Well, one reason could be the iPhone itself. There are various data points here, but among them are the following:

• Google stated in February 2008 that it was seeing 50 times more searches from iPhones than any other device
• AdMob figures from earlier this year suggest that in April the iPhone generated 43 percent of mobile Web requests and 65 percent of HTML usage.
• Net Applications reported a 0.09% browser marketshare for the iPhone Safari browser in December 2007 – already two times greater than all Windows Mobile devices put together.

It’s clear that the iPhone has a tendency to generate significantly more data usage than other devices, and this seems an entirely plausible explanation for the poor performance of AT&T’s 3G network, both in these tests and in real-world experience (as anyone who’s visited the new Yankee Stadium with an AT&T device knows, it’s virtually impossible to get a reliable connection in there despite strong signal strength – this year’s SXSW is another prominent example of the impact iPhones can have on a network.) If the iPhone really is having that serious a negative impact on AT&T’s network, it will obviously affect the iPhone customers themselves but also anyone else trying to use an AT&T device (including laptop modems) to access high-bandwidth applications, including video, file downloads and so on.

I’m going to wrap the next two bullets into one paragraph here because they’re two sides of the same coin. The perfect illustration here is the Apple WWDC keynote in May: when MMS and tethering were announced to applause, in each case when the presenter indicated that AT&T support would not come until later in the year there was a smattering of laughter (see the two relevant clips here). That was a fairly benign response, but on Twitter (just try this search – even weeks after the announcement) and elsewhere it’s been much more vituperative. iPhone users are ticked at AT&T for not supporting these features and they’re not afraid to say so. It’s a significant negative brand association for AT&T, and one it hasn’t really dealt with well. But at the root of the problem is the fact that AT&T has ceded control to the device vendor (Apple) in the way it never has for any other device. With any other device, AT&T would carefully control the release and announcement of features, but here Apple is making announcements to the whole world simultaneously and AT&T is singled out as one of the few carriers not supporting these new features. AT&T cannot prevent Apple from announcing these features and it cannot stop the egg from landing on its face when it’s not ready to support them. The degree of control ceded to Apple has put AT&T at a significant disadvantage and it has ended up becoming the ‘evil gatekeeper‘ in iPhone customers’ minds.

Lastly, AT&T’s over-reliance on a single device. Earlier, I quoted Morgan Stanley survey data to suggest that the iPhone drives around a third of AT&T’s gross adds, and that number is probably higher in quarters when new iPhone versions launch. That’s great as a driver of switching from other carriers and gross adds, which is fine for now. But it also means AT&T is heavily reliant on the device, and the risk is that it will eventually lose exclusivity over the device. It’s probably fair to say that the iPhone itself is the most significant driver for customers switching to AT&T and the iPhone, rather than anything about AT&T. Therefore, if the iPhone were available on their existing network, there would be no reason to switch. And there goes a third of AT&T’s gross adds, at least in theory. Meanwhile, that short-term impact on flow share seems to have been a bit of a flash in the pan – here’s a more recent version of that Morgan Stanley chart:

The negative Verizon flow share trend has reversed, and the AT&T trend really isn’t one at all – its numbers are bumping around around zero. Verizon seems to have recovered very quickly from the original iPhone launch (and whatever broader issues were plaguing in previous quarters) and has seen only minor blips from the subsequent launch of the iPhone 3G. So, yes, AT&T continues to gain subscribers from Sprint and T-Mobile, but against its strongest competitor it is doing worse than it was when the iPhone first launched. What if Verizon gets the iPhone? Almost all the negatives still apply but one of the largest positives goes away.

In summary

So, to summarize:

There are three major positives associated with the iPhone for AT&T:

• Subscriber additions
• Higher ARPU
• The ‘halo effect’

Conversely, there are several negatives:

• High device subsidies
• Negative network impacts
• Ceding control and brand management to Apple
• Over-reliance on a single device.

The challenge in coming to an overall conclusion here is that not all these benefits and disadvantages are quantifiable – there’s no single yardstick by which we can measure them all. But it’s clear that, along with the positive impacts of the iPhone on AT&T there are several negatives, and they’re getting worse. At this point, even those subscriber additions are merely keeping AT&T on par with its competitors rather than allowing it to gain significant ground. Meanwhile, its network is being trashed, it has heavy device subsidies to deal with, and it has gained a reputation as the ball and chain holding back the iPhone from reaching its full potential. At present, the pros and cons are at best 50/50, but should AT&T lose exclusivity the balance would definitely shift towards the whole iPhone experiment having been a negative one for AT&T. After all, exclusivity was the “quo” in the quid pro quo that the AT&T/Apple deal represented. When that ends, the benefits to AT&T largely do as well.

Some have leapt to AT&T’s defence in recent days (see this example) but the approach here seems misguided. It’s not as if anyone forced the iPhone on AT&T. In fact, we’ve known from the beginning that it was offered to Verizon first but it turned Apple down. The reason? “We just couldn’t reach a deal that was mutually beneficial” according to Verizon’s spokesman. And that sums up the problem perfectly: the iPhone had the potential to be mutually beneficial for both parties if the right deal was struck. But whereas Verizon didn’t see what Apple proposed as mutually beneficial, AT&T apparently did, or was keen enough on the idea to strike a deal that wasn’t as balanced as it might have liked. But although Verizon was roundly pilloried at the time for passing on the hottest device of the decade, it’s starting to look like a smart move at this point.

What are your views?

Bookmark with:

But one of the things I was most encouraged about and interested in was some of AT&T’s activities that I would classify as fitting within my social telco definition. As AT&T expands into the cloud computing space, it will be launching storage-as-a-service and compute-as-a-service offerings in the coming months. But it will also be developing something it calls “platform-as-a-service”. It’s not the best name, because it’s not all that descriptive – as a colleague said, it reminds you of Salesforce’s efforts to expand from its core CRM-as-a-service proposition to something broader. But the concept is good, even if the nomenclature isn’t.

What platform-as-a-service would do is expose both computing and network functionality through APIs to third parties. This would allow those third parties – whether ISVs, enterprises developing their own apps or even web players – to create apps that would be able to issue commands to AT&T’s compute and network infrastructure. The focus in AT&T’s presentation on the topic was on the compute-type commands, but the plan is very much to roll out more or less the full set of Parlay X commands over time in handfuls at intervals over the next year or two, starting later this year. 

One of the things I found interesting about this is how AT&T is approaching some of the greatest challenges associated with exposing this kind of functionality to third parties: namely, verifying identity and providing security in order to protect privacy. AT&T is piggybacking off its efforts on the mobile application side here, benefiting from its DevCentral developer community and the processes and interfaces developed to facilitate the development of mobile apps by partners. This is probably a good model for other telcos with well developed mobile developer ecosystems since a lot of the legwork has already been done. 

I’m going to be requesting a more detailed briefing on these activities, and especially on the linkages between consumer and business efforts in this area, and will likely put something together for publication subsequently. But this certainly looks like an interesting and promising initiative from AT&T in this area, and one that other telcos can learn from.

Bookmark with:

I spoke to a group of salespeople from a software vendor who sell to the telecoms industry this week. The topic of my presentation was how telcos are transforming themselves and in particular the evolution of customer care (or customer experience as they’ve apparently been taught to call it by Accenture). For all the complex and important sounding changes I discussed with them in terms of processes and software applications, the fact is that there is still something very fundamentally wrong with customer care when people are still having experiences like the one detailed by Thomas Hawk on his blog. It goes way beyond software and back to the approach taken by the telco to the whole question of care.

Telcos need to:

• see customers as customers, not as a series of transactions, and as such give each of them one number to call which will allow them to deal with any issue associated with their account
• equip the person on the other end with a holistic view of the customer so that they know everything about them, and capture whatever information they enter in the IVR stage permanently so they don’t have to keep repeating the same information over and over again to each new person/machine they talk to
• empower individuals in call centers to really solve problems for customers, and not just pass them over to someone else. Give them more decision making authority and allow them to chase other employees as necessary to get things done instead of making the customer do the running around
• remember that if the customer is calling care it is because you have not done something that they want you to have done, and they expect you to be able to fix it. This is your problem, not theirs, and they have every right to expect you to deal with any complexity that exists that may make this more difficult to achieve than it should be. Don’t tell them why you can’t do it – instead, figure out when and how you will do it, and let them know.

All of this is so basic, and yet telcos still struggle with it at the most fundamental level. A good chunk of the transformation of the customer experience that’s happening at the moment should be going into changing attitudes rather than applications, or all the investment in the world in new software and processes isn’t going to make a difference.

Bookmark with:

The problem with many of these reports is that they focus on price and availability of high speeds without investigating the negative effects associated with heavy government intervention in the market. As an example, Japan is often cited as the beacon of international broadband, often closely followed by Korea, but in both of these countries the government has intervened in a heavy-handed fashion to achieve the results seen today. European countries often also score highly, often because of the competition introduced via local loop unbundling regulations – a less intrusive form of intervention than in Japan and Korea but nonetheless a much more aggressive form of regulation than that which applies in the US.

Another flaw is that it is assumed that faster speeds are always a good thing, and that diminishing returns never set in. The fact is that, beyond a certain point (currently around 10Mbit/s or so) extra bandwidth is just that – extra. There is almost no application in existence today which requires more than 20Mbit/s when in peak use, and so 100Mbit/s is a senseless benchmark. Getting more of the population to 5, then 10 and ultimately 20Mbit/s is a reasonable goal, but beating up on the US for the paucity of 100Mbit/s connections is an exercise in futility that could lead to bad investment decisions. Verizon’s FiOS infrastructure is certainly capable of delivering that kind of bandwidth, although AT&T’s U-Verse probably isn’t under the present architecture. But the point is that it doesn’t matter.

Another thing that’s rarely examined is the pricing side of the equation. Providers in Korea in particular have a very spotty financial history due to the suicidal price wars they’ve engaged in. But that competition has been spurred by the fact that they – like the US – have inter-modal competition between various infrastructures, not just regulation-based service-level competition on a common infrastructure. The latter gets quick results in terms of number of providers and price competition but it rarely foments real innovation because the underlying wholesale services everyone is using are the same. With a shift from bitstream to local loop unbundling products that changes somewhat, but competing infrastructures – especially ones built on fiber – are much more likely to provide real differentiation.

Hence the massive speed and price competition that’s been triggered in the US in areas where fiber has been rolled out by either AT&T or Verizon. In time this will reach more and more of the population and provide a further boost to both the speed and price sides of the equation.

The biggest issue for me is that these reports rely a lot on the question of timing. Where the US is today, other countries either were yesterday or will be tomorrow. We’re all heading down pretty much the same path, just at different speeds. The impatience that often accompanies the criticisms of US broadband deployments is misguided too. It usually leans on an argument about competitiveness and the ways in which broadband can transform the way we work by providing more opportunities for home working / teleworking and so on. But guess what? The measly 5Mbit/s so derided in these studies is just fine for most homeworkers and is available to almost everyone. The biggest barrier to adoption of home working is the cultural change involved, not the technology. Many companies and individual managers are still uncomfortable with the idea and suspect that a home worker is a less productive worker. That attitude needs to change more than we need more government intervention or sackcloth and ashes about the parlous state of the US broadband market, and it’s great that we finally have a study that seems to get that.

Bookmark with:

Firstly, it’s been interesting to see the shadow the iPhone casts over everything even though Apple isn’t visibly present at the show. Sprint’s big announcement was around the Samsung Instinct, which is a clear iPhone competitor. But the devices on display were running beta software which was glitchy and slow, and it was clear that – though they have some nifty features – these devices are not a match for the iPhone. AT&T itself had another device which mimics certain aspects of the iPhone – the LG Vu – but it is another poor match for the device on everyone’s minds. Of all the things that people love about the iPhone – the design, the UI, the browser, the ease of use – none of them are matched by most of the devices on display here, even though the manufacturers of those devices have been making phones for far longer than Apple. The Sony Ericsson Xperia X1 showed the most promise of any device I saw at CTIA, but won’t be launched for several months.

And AT&T appears to be keen to cement the thought leadership the iPhone deal has given it. Its announcement that it will deploy Microsoft Surface tabletop computers in some of its stores will further up the cool factor for AT&T and put more pressure on its competitors to find ways to compete. I haven’t seen much from AT&T’s competitors that can match it in terms of providing differentiated experiences on devices or in stores. (I have to admit that throughout the Surface presentation I was thinking about this YouTube video which I first saw a few months back – “take that, Apple”).

I discussed managed mobility services with several players at CTIA, and found broad consensus in several areas. It seems clear that the next several months will see launches from major players including both AT&T and Verizon around managed mobility services, and that a range of factors are coming together to create a fertile environment for uptake of these services. The complexity I have referred to previously in the enterprise mobile arena is creating demand for these services. And technology is now available to enable the supply side, both from specialists like Mformation, Sybase and Nokia/Intellisync and from RIM and Microsoft. Launches in the next few months from those two big carriers and increasing uptake over the next year or two should follow.

“Openness” appears to be becoming the new “convergence” in that it is a term everyone seems to feel compelled to insert into every pitch and keynote despite the fact that it means different things to different people. AT&T still appears frustrated that Verizon has got so much attention for playing catch-up with the GSM world: as Ralph De La Vega (head of AT&T Mobility) put it today, “we were open before open was cool”. But he also suggested AT&T now views Android much more favorably than it did at first, ironically because Android will be “open” to AT&T’s branding and applications in the device UI, rather than being restricted to just Google and open source software. I’m hoping the open thing will soon blow over at least in the form of hype, and that we’ll start to see some significant real moves towards openness. Android will be important to watch when it launches – Texas Instruments is demoing two Android devices here – but it can’t be the only game in town.

Carriers need to get better at explaining that they already offer openness on the RIM, Windows Mobile and Palm platforms, where users get unfettered access to the Internet and the ability to install their own applications. But they also need to find ways to extend that openness all the way down the portfolio for those customers who want that. And they need to stop pretending that “choices” and openness are synonyms. Just because you give your customers a choice between two hand-picked applications does not mean your approach is open. Allowing them to pick the application they want regardless of whether you have endorsed it is. And carriers still have some learning to do in this department.

Overall, the show is as always a nice snapshot of a point in time for the wireless industry. But I hope that by the time the Fall show rolls around we’ll have moved forward in all these areas – compelling devices, managed mobility and openness in particular.

Bookmark with:

The Verizon and AT&T deals are pretty much identical – $99 per month for unlimited calling. T-Mobile threw in unlimited texting, which makes sense since its user base tends to skew young and therefore is more prone to communication via thumb than mouth. US Cellular’s is a national offering too but its user base is more regional.

Financial analysts and investors have largely seen all of this as a bad thing, either because it will start a price war, or because it will take everyone spending over $100 on voice currently and bring their spending down to $100, by definition decreasing their spending (Om Malik would appear to be a case in point).

There is already speculation that Sprint will attempt to undercut all of the above, which it could do by simply charging less than $99 for its unlimited voice plan, or presumably by charging $99 or slightly more for a plan that would include unlimited voice and messaging and/or data usage. Certainly, Dan Hesse has suggested that he has what have variously been described as “nukes” or simply “missiles” he can fire off to kick-start the turnaround at Sprint, and one of these is presumed to be a dramatic move on prices. With the other carriers having now stolen a march on that particular idea, he may need something new.

However, it’s not clear that it would have made a huge difference even if Sprint wasn’t playing “me too” at this point. Think about it. Wireless churn stands at somewhere between 1 and 3% for the larger US carriers. That means that in any given month, only 3% (or fewer) subscribers switch carriers, or put another way the average lifetime of a subscriber is between 3 and 8 years. Even the most dramatic move on pricing would be unlikely to loosen up more than a small number of additional subscribers in any given period. Look at the iPhone – growth appears to have slowed, and there are doubtless several reasons. But one is the simple fact that many people are locked into 2-year contracts (which by themselves would limit churn to just over 4% if everyone stayed in them) and over three quarters of US wireless subscribers are currently with a carrier other than AT&T.

Given that Sprint currently has negative “flow share” towards the other three big carriers, just turning that trend around would be something. But simply reducing prices will not likely do the trick on its own, especially when competitors are making similar moves. Forrester has a survey which has been used by Morgan Stanley to look at brand loyalty, and it illustrates where Sprint’s problem really lies:

Verizon scored an average response of 7.7 out of a maximum score of 10, AT&T and T-Mobile scored 7.2 each, with Sprint Nextel averaging 6.1 among their customers. Factors such as reliability, trust and prior experience
were rated as key factors in making a carrier choice.

Sprint has by far the lowest rating of any of the main carriers (Nextel’s independent rating is even lower), and this ties in directly with its churn. It needs to be doing a better job of keeping existing customers happy at least as much as it needs to win new ones.

As to the question of whether the impact of unlimited pricing plans will be good or bad, it’s hard to argue they’ll be good. The answer really depends on which of four resulting trends is strongest:

• existing customers of a carrier switching to a higher-priced plan (i.e. going from limited to unlimited), which would have a positive ARPU and revenue impact
• existing customers of a carrier switching to a lower priced plan (i.e. because they currently spend more than $99 either because their plan costs more or because of overages), which would have negative ARPU impact
• customers switching from other carriers, which would have positive subscriber and ARPU impact, but which seems relatively unlikely on the whole because the model has swept all but one of the major carriers in the space of a week
• new customers signing up with the carrier because of the new plan (which seems least likely of all, since current wireless non-subscribers tend to be poorer, with poor credit scores, and are therefore much more likely to adopt pre-paid or at least low-priced postpaid offerings.

Given that the fourth trend is likely to be negligible, and third also small, that leaves the first two. There is an argument for switching from a lower-priced to a higher priced plan if it allows you to make another simultaneous change – i.e. to switch your calling from another network to your wireless carrier. If people cut the cord either at home or in business as they make this change, they may save money overall while increasing spend with their wireless carrier. The premium on top of more modest allowances of minutes is likely to be at most $50 and probably considerably less, so it would be competitive with unlimited wireline calling plans. However, it seems likely that the percentage of subscribers currently paying more than $100 for their voice services who will switch to the $99 plan will be close to 100% within the first few months. While providing some goodwill benefits similar to those enjoyed by Sprint with its Fair & Flexible plans and AT&T/Cingular with its Rollover minutes, it’s not clear those will translate to sufficient churn reductions to offset the loss in ARPU / revenue.

Had Verizon been alone in making this move, the picture would look very different, even if it only had a few months of lead time over the other carriers. But because the others have responded – or are likely to respond – very quickly the overall impact seems likely to be at least slightly negative.

On the other hand, it’s also worth asking what would have happened to voice ARPU over the next year anyway. It has been stable for some time, and with most growth coming from prepaid and family plans at present it is likely to drop considerably in the coming years. Per-minute pricing has been dropping for some time, since that ARPU has been buying ever larger numbers of minutes over time. The current model for consumer communications has flat-rate pricing as its endgame every time (see broadband, wireline voice, TV), and although it has taken a very long time to get there with wireless, we’ve arguably had several baby steps already – the “bucket of minutes concept” and the elimination of long-distance and roaming charges being among the most obvious. This will doubtless accelerate the decline in ARPU somewhat, but overall it may simply cap voice ARPU at a nice high rate (about twice current ARPU), freeing consumers to increase spending on data, which is where all the growth is today regardless. It may not be as bad as some people think.

Note: the image used in this post is a picture I took myself a while back on a walk through NYC. Unaccountably, a huge inflatable rat was sitting on the back of an unattended trailer outside a Verizon Wireless store. Seems somehow strangely apt for this story. Original can be viewed here.

Bookmark with: