Categories

Think local, act global

One of the curiosities of the mobile phone market is how vast it is but also how heterogeneous it has always been. I wrote about this in 2010:  Smartphone parochialism: How operator policies prevent or promote platform adoption. This observation was influenced by my time at Nokia where I became amazed at how differently users behaved in different countries.

There were many causes. Some cultural, some historical, some economic and some policy-driven. The result was that it presented a challenge to any company with global ambitions and indeed it was rare to see the same company do well in every market. Japanese companies did well in Japan, European companies did well in Europe and US (and Korean) companies did well in the US. Nokia was most successful because it was able to apply an European model more broadly (but not in the US). Samsung succeeded by simply adapting to each and every market with hundreds of products. But it was a particularly “provincial” market.

My assumption was that when smartphones would become the majority of phones in use there would be a normalization of behavior and thus a homogeneity of preferences. This is, after all, what happened in other platform games. PC form factors are globally consistent, PC operating systems are equally preferred around the world, FaceBook, Google and Twitter are also, unless censored, uniformly popular. Apple’s iPod eventually also became a global phenomenon with no material difference in preference by market. Likewise for game consoles.[1]

However, a decade after the broad adoption of smartphones, the relative popularity of various platforms is still unevenly distributed. Prior to the iPhone, Symbian was strong everywhere but the US and BlackBerry was very strong in the US but weak elsewhere. Windows Mobile had footholds in some markets but little traction in others. Today the picture has changed but it’s still a patchwork of preferences. Consider the following graphs.

Screen Shot 2013-09-18 at 9-18-12.57.15 PM

They are the most detailed sets of data on platform phone purchases but they not the only data: We also know that Android is vastly more popular in India from web usage statistics. We know that BlackBerry is sensationally popular in Latin America and the Middle East. We know that within Europe there is a gap between North and South in iOS vs. Android and even more pronounced between East and West. We know that whereas iPhone was a huge hit in Korea upon its launch there, Samsung and LG swamped it in the last two years.

But the data shown above is the story of iOS being very popular in Japan and increasing its share in the US while being flat in EU5. Again these three markets are different but each contradicts the global picture.

Note the differences in trend not just absolute performance. This is what is puzzling. There seems to be a divergence between markets: Markets are become more islands onto themselves rather than one whole. This contradicts all the theories of how computing markets should behave (network effects, ecosystems, monopoly power, etc.) It implies that there is still a great deal of friction in the market; friction which I suspect to be due to the peculiar nature of telecom economics.

I wonder if this observation is, as usual, late on my part and that Apple’s iPhone product marketing has seen it years ago.

Notes:
  1. Although Japanese do seem to prefer Japanese platforms more. []
  • willo

    I think there are a variety of reasons for this:
    * Marketing budget / Exposure
    * Channel distribution / models
    * General opinion of brand / journalistic favors

    I think the general “mood” in a market is greatly influenced by media. If say a country has two large media outlets, and both has tech journalists that greatly favor one over the other, the continuing media coverage/bias will greatly shift general consensus towards one of the other. Small countries like those in Scandinavia have few tech journalists, and if those align you will see a general consensus in the market towards those bias opinions.

    I would say that countries that appear to have Apple Retail stores widely available tend to have a more positive opinion towards IOS/Apple than those countries that do not.

    Similarly, those countries with fairly market proven subsidized models tends to be more in favor of IOS/Apple.

    I think Apple knows best how to market and segment their product line-up, although I think the regional managers of those countries matters performance wise. Only time will tell if Apple is able to transform the handset business into a more subscription based model. Clearly they are trying their best.

  • mat

    The reason for Android in Europe is simple: Price

    I thought about updating my iPhone 3GS to an iPhone 5 (back in Dezember) but got a Nexus 4 instead because the price was 350€ compared to 640€. Its just not woth the price difference for me. (And i would expect many other to think the same)

    • http://www.asymco.com Horace Dediu

      If price is the primary factor why didn’t you buy a cheaper phone. Seems you could have done a lot better than €350.

      • Will

        Because it’s not just about the price, it’s value for money.

      • mat

        Exactly. The cheaper Android phones had to many compromises (Slow, no Updates,…)

  • vbonline

    Phone subsidization is not really popular in Europe (because of price, TCO of a 2 year contract WITH a current iPhone is close to 2.000 Euro, Buying the phone from Apple and get a cheap card is regularly under 1.000 Euro for two years)…

    With a cheap(er) Android phone you can get TCO under 500 Euro for two years.

    So it is a question of available money. Northern Europe (think Switzerland, Germany, Scandinavia) has a bigger iPhone share, because it is hit less by recession than Southern Europe (think Greece, Italy, Spain, Portugal)

    Overall Android share is bigger, because it is cheaper… Just to repeat to the fellow Americans: The iPhone 5s is NOT a 199 USD phone by ANY means… ;-)

    • Will

      Where in Europe? You can get the iPhone 5 for £30 a month in the UK with unlimited data and £30 up-front http://www.three.co.uk/Discover/Devices/Apple/iPhone_5?tab=Features.&memory=16&colour=White

      Subsidies are quite common in Europe.

      • vbonline

        http://www.swisscom.ch/en/residential/mobile/subscription-tariffs.html

        Unlimited contract with LTE speed approx. 136 Euros per month…

        I pay 8 Euros (10 CHF) per GB in Switzerland and buy my iPhone from Apple…

      • Will

        136 euros for the BEST DEAL in the world! Unlimited data, roaming data AND unlimited calls to EU/US/Canada?! It’s also half the price for under 26… I want that!!!

        There are also cheaper alternatives if you want …

      • vbonline

        http://www.swisscom.ch/en/ghq/portrait/market-competitors/market-shares.html

        Swisscom market share : 67%, so a lot of people aren’t looking for alternatives…

      • Will

        No, I mean you can get the 50 euro plan with unlimited data if you want. That’s quite reasonable. Plus, Switzerland is quite an expensive and it’s not representative of Europe.

      • vbonline

        50 Euro / month gives you 256 kilobit/s (Yes, Kilobit!). This is not reasonable, this a “no offer”. If you want to use HSDPA as a minimum (“up to” 21 mbit/s) you have to pay at least 100+ Euro…

        First you ask “Where in Europe?” as if you don’t believe me, now Switzerland is not representative for Europe… Maybe the 51st State on that small island where your example is coming from is not representative for Europe… :-)

      • Will

        France, Spain, Italy, Romania all have great deals with free or cheap high-end smartphones. Just because your country has a monopoly in the telecommunication industry does not make Europe a bad place to get a contract. A basic McDonald’s meal in Switzerland also costs 10 euros there.

        It is simply ignorant to class Europe as a place where you don’t have choices, when it has the best infrastructure and greatest competition when it comes to mobiles contracts.

      • vbonline

        Do you actual read what I’ve written? Swisscom has a marketshare of 67%. This is NOT a monopoly.

        Telekom in Germany charges 69,95€ per month for a 5c with contract, Vodafone in Germany 69,99€ and those two are the market leaders in Germany as well

        Be VERY happy with your £30 offer… But £30 is NOT the typical offer for contract including a current iPhone in Europe. At least not in german speaking countries…

      • Will

        Well there you go! You can get subsidized iPhones :)

      • vbonline

        Yes, but no one who got past 4th grade math successfully does (or at least should)… Because 70€ * 24month + 99€ for the phone compared to 8€ * 24month + 599 for the phone?

        That’s why I’ve written about TCO over 2 years in the original post…

        But go ahead with your contract plus subsidized phone. I won’t stop you…

      • Will

        Do you argue for the sake of arguing?

        All I said is that you CAN get the iPhone for £30 (or 50 euros like you said) and people DO buy it like that.

        Of course it’s cheaper to buy the phone directly but spreading payments over two years is convenient even if you pay more.

      • vbonline

        No, but I stated in my original post:

        “TCO of a 2 year contract WITH a current iPhone is close to 2.000 Euro, Buying the phone from Apple and get a cheap card is regularly under 1.000 Euro for two years)…”

        You doubted that… Now let’s do the math:

        70*24+99 = 1.779 (close to 2.000)
        8*24+599 = 791 (regularly under 1.000)

        … and this is Germany (biggest market in Europe), Switzerland is worse…

        So I still stand by my original post. You you still doubting?

      • phoner

        I think roaming charges are a big differentiator between big countries (US, China, India, Canada, Russia) and small countries (most of Europe).

        I went in to dispute a €60 add on to my bill because they had “accidentally” removed my free number, and the guy could hardly believe I had bothered. He said most people (who are on business tariffs) who come in to dispute bills have monthly bills in the thousands of euros.

        I never use 3G (or 4G if it exists) roaming – my operator’s roaming tariff is €300/month for 0.5 Gb and about €800/Gb thereafter ! Yup the price of an iPhone for 1Gb excess data. You can see why the EU has been having such trouble enforcing cross-border telecoms plans.

        Obviously people who use these data plans can easily afford iphones. But very few people can afford these plans and most people in the world try to avoid roaming charges if possible. I wouldn’t be surprised if at least 30% of Europeans lived within 100 km of a national border and crossed a border several times a year.

        Dividing markets has always been lucrative (in the short term) for businesses.

      • Will

        That is true everywhere where contracts exist! It has nothing to do with Europe!

      • eyez00

        I have that £30 deal in the UK, but somehow it costs me an average of £43 pm.

      • obarthelemy

        http://www.lesnumeriques.com/telephone-portable/samsung-galaxy-s3-p12749/comparatif-cout-revient-galaxy-s3-chez-operateurs-n28043.html , the brown bar is the TCO over 2 years: it’s fairly easy to get below 1,000€ /2 = 500€/yr.
        That’s jan. 2013 and a 500€-ish GS3, the GS4 is at the same price.

    • obarthelemy

      again: “I’m paying (France) $20 for unlimited everything (landline+mobile national calls and texts; landline to 80-ish countries; mobile to 40-ish countries; 3GB data then throttled; tethering; any protocol incl.VOIP, torrents….). I’m sure I could pay more… but what for ? With all the wifi around me, I’m using 500MB data/month….”

      • vbonline

        Congratulations! Keep the contract and don’t move to Switzerland (or Germany)…

  • TC03

    In this article you state that market are becoming islands. I doubt that. My observation is that the more pricing becomes an issue, the more market share Android tends to have. This is a global trend. Exhibit A: China. Exhibit B: Europe.

    So basically people favor Android (because of the price). In markets where Android is not topping market share, the market is somehow disturbed due to local influences. Examples are the subsidies in the USA, lack of 3G on China Mobile iPhones, etc.

    So basically the trend is globally the same, but the markets are subject to different constraints which can alter this trend.

    • mr.lee

      I think (and agree) it is all about the price. I think Horace sums it up best here: “It implies that there is still a great deal of friction in the market;
      friction which I suspect to be due to the peculiar nature of telecom
      economics.”
      - why would Apple forego the huge subsidy payments from telecoms in the US? (why wouldn’t Apple keep a death grip on the high-end, high-profit, mobile market? why wouldn’t Apple continue to upscale market iOS?)
      - if price were no object (or, it didn’t matter to me; and ignoring sects with any quasi-religious obsession with Apple or anti-Apple), I think it’s obvious iOS wins. Case in point: for my 80-year-old Mom, iOS would be the only phone I would consider for her.
      - I’m not arguing that Apple could aggressively attack the middle-to-low market with a low-end iOS device, and likely take huge market share; however, it appears to me at the significant cost of profits (plus alienating telecoms, as it becomes a free-for-all as there’s no “lock-in” to their service plans). They might make more money that way, or might not. And might subject themselves to monopoly scrutiny, as well.

      • charly

        S4 is about as expensive as an iphone and has a larger marketshare so claims that ios would win if price were no objection are probably false.

        Apple maybe could have got the market if they attacked the cheaper price points two years ago. Now it is to late and they don’t have to worry anymore about monopoly scrutiny

      • evidence

        “S4 is about as expensive as an iphone and has a larger marketshare so claims that ios would win if price were no objection are probably false.”

        This is a huge claim that isn’t backed by any evidence at all.

      • charly

        It was true when it was introduced, especially if you consider world except US

      • isitjustme

        Point us to a link that said so.

      • claimchowder

        Impossible, because such a link does not exist. Those stupid trolls…

      • captialism
      • claimchowder

        Sorry I wasn’t specific enough, so let me explain.
        What has been missing from Samsung is hard data on the number of high-end (e.g. Samsung Galaxy series) units sold. All that is ever published are:
        - Android activation numbers (not Samsung’s)
        - Samsung revenue and overall handset sales (not just Android, though most of their handsets are by now)
        - but most notably NOT the number of Galaxy S# or Galaxy Notes sold. So how can anyone claim that Galaxy market share is higher than iPhone market share?

        I do think they occasionally publish how much they stuff into their channel, which then tries to get rid of them by BOGOF-promotions, bundling them with TVs and fridges etc. But not the number of Galaxy handsets that anybody actually paid for.

        And most certainly, given the huge amount of cheapskate low-end Android phones they are shipping, or maybe even selling, there is BIG reason to assume that NOT most of them are Galaxy high-end models.

      • obarthelemy
      • obarthelemy

        Actually, price is no object for the phones I buy for myself, and I choose Android because I need a big screen, an SD slot, and FM radio. Price was also no object for the phone I bought my 70 yo mom, and I got her an Android because Big Launcher was required for her to be able to see and handle and understand the phone.
        iPhones look very good, are very small, and take very nice pictures. That’s not an all-around skillset, lots of people don’t need that, but need other things. Widgets, SD cards, large screens… even a silly old built-in USB port are killer features.

      • claimchowder

        “I buy for myself, and I choose Android because I need …”

        But nobody cares about what you buy.

        And I’m sure your mom will make full use of all the widgets, SD cards, USB port, and ssh client in your L3E7-63EK-Phone.

      • obarthelemy

        It’s just than some people keep claiming that the only issue with iOS and iPhones is price. it’s not: a lot of features and functionality is missing too.
        BTW, my brother SSHes into his Raspberry Pi server from his iPhone too, so SSH is not one of the iPhone’s missing features :-p

      • mr.lee

        > claiming that the only issue with iOS and iPhones is price
        >

        it’s obviously “more than price”: there’s cachet (marketing), there’s easy-of-use, and, actually “being useful” (accomplishing needed tasks, and doing tasks that you didn’t even know before).

        if you’ve ever seen a non-computer literate person using an iPhone/iPad, compared with the same using Windows, you’ll know what I mean.

        There is an analogy with cars: the car we buy, we think is the best choice (based on cachet, features, functionality, performance, and *most* importantly “value”!). This value is entirely subjective and economic based: a Honda Accord might be the best car for one segment, but if you’re making US$1M annually, a Tesla looks better. There is the “cheapest” valuation that some use, but actually most wouldn’t touch it. As with most analogies, this eventually breaks down due to the cash outlay. But my point which I stand by, is that the pricing is very similar for an iPhone and Samsung, and the iPhone is hands-down winner for my own Mom. (And I look around on the subway, and my sampling suggests on the Boston subway that iPhone wins 75% of the time. Disclosure: I myself have a Nexus4, but I had a iPhone for a couple years: I concluded it really was better, but not enough for me to pay that, so it was my *own* value choice.)

      • Kizedek

        Then your research seems to have been lacking and biased.

      • Kizedek

        You’ve been told before that iOS has built-in features for Accessibility.

        iOS also has better and more consistent touch targets, due to its UI and touch technologies, that make it both easier to handle and understand as a phone, provide better response and feedback, and is more accurately predictive about user intentions as well. Not to mention having a better variety of apps of all categories to suit every palate and need.

        Therefore, iOS is very well received and reviewed by the elderly, young and the handicapped.

        You may say that your mother preferred a larger screen than provided on the iPhone. You may state that that preference, and your previous ignorance, were the reasons you bought what you did.

        You may not now pretend that iOS is any less suitable for your mother to “be able to see and handle and understand the phone” in any way beyond physically focusing on the actual real estate of the phone screen at a particular distance.

        You may not continue to state and act as though iPhones don’t have an “all-around skillset”, particularly in the area of “handling and understanding the phone”, as this is patently not the case.

        But, of course, after the very next article, you will say much the same thing yet again.

      • obarthelemy

        You’re so biased and illogical it’s funny.
        I y not criticize your pet iOS because I’ve only had to try out the usability features for 5 mins to realize they didn’t make up for
        - small screen
        - no widgets
        - inconsistent interface
        - still a whole lot of swiping going on
        - no synching of tabs and bookmarks with the desktop
        - …

        Yet you are fully qualified to not only dismiss my conclusion but tell me what I may and may not say… I guess because having no knowledge whatsoever of what Big Launcher does and being peremptory based on that nil knowledge seems logical and justified to you ?

      • Kizedek

        In this particular instance, I am no more illogical or biased than you. I was perhaps guilty of reading into your comment as much as you have just read into my reply.

        Because I know your biases, I interpreted your comment like this:
        “I got her an Android because Big Launcher was required for her to be able to see and handle and understand a phone.” Any phone.

        This isn’t about Big Launcher at all. I don’t particularly care about Big Launcher, because I know that the built-in Accessibility features of iOS are more than adequate for most 70-yr olds that care to try them. The elderly are very happy with it. That is where you have read into my comment.

        Now, if you want to say that she had to have Big Launcher on that particular phone that you bought her, in order to understand and use that phone, no problem.

        I have not acted as though Big Launcher doesn’t exist. On the contrary, I am sure there is a big need for it. I simply directed your attention to the existence of Accessibility features on iOS. I was defensive, certainly; but not biased, illogical or ignorant.

        Rather, I was commenting on yet another throwaway line of yours (lack of “skill set”, etc.) by which you aim to constantly dismiss or ignore elements and features of iOS…
        perhaps simply because you have made a peremptory judgement based on nil knowledge and having no capacity to be corrected, or (since that is difficult for most of us to believe) you are a likely troll with an agenda.

        You sure seem to feel you are entitled and justified to continually act that way in the vast majority of your comments about every aspect of Apple, iOS and the iPhone. Which on the whole shows a staggering amount of illogic, blatant bias and wilful ignorance on your part in the face of patient and repeated evidence to the contrary by a lot of posters here. Sort of what you just accused me of …only on a grand scale.

      • charly

        Don’t forget the keyboard.

      • obarthelemy

        That link is interesting: http://bgr.com/2013/07/05/iphone-5-vs-galaxy-s4-study/

        Online chatter at launch was mostly about brand for the iPhone (twice as much as the GS4′s chatter). That is reversed with features: the GS4′s launch shows about twice as much “features” chatter than the iPhone’s.

        One might start to conclude that the iPhone is about brand, and the GS4, about features ?

      • Kizedek

        You don’t seem to get something pretty basic — with Apple’s “brand”, there are actually concrete things to talk about. That is why Apple’s brand is so valuable! It has been thirty years in the making.

        Generally, “chatter about brand” doesn’t amount to much, I’ll give you that. I can’t imagine what there is to chatter about with a clothing brand or the latest social media network that is trying to get a name for itself in a world dominated by Facebook — maybe the chatter is about the latest photoshoot, or did you see such and such celebrity dancing on the tables at the MS Store and get one of the tickets they were giving away, I don’t know.

        So, who chatters about Samsung as a brand? I don’t. What is there to say about them except they are some mafia-esque ruthless family business that makes everything from refrigerators to WMD. That’s fun chatter.

        The only thing to talk about with Samsung is how your latest Galaxy has such and such a processor that runs x amount faster than the previous. Again, fun chat. Are users talking about stuff like how many great photos they took and processed last weekend? Sounds like they aren’t.

        But Apple? There is all kinds of chatter about concrete things that re-inforce its brand, and in turn are represented by its brand…

        Hey, how about a story where a guy takes his cracked iPhone into an Apple Store and walks out with a new one. Hey, how about the story of an Apple product helping a family cope with the special needs of their child because they were able to communicate and interact and learn together better…

        You get the picture. That’s brand chatter when it comes to Apple.

      • dreamfeed

        That’s the whole point: the iPhone is for people who want to be confident that they are buying the highest quality product without researching what “features” it does or doesn’t have. I have no idea how many cores the 5s has but I’m planning to buy it tomorrow because I’m confident, based on history, that it will be the best phone available for months to come. That’s what “brand” is all about.

    • obarthelemy

      Lumping under the “Android” banner phones from $50 to $1,000 is a bit like lumping under the “not BMWs” moniker cars from $2,500 to $50,000.
      I’m fairly sure GS4/HTC One/Xperia Z owners behave much more like iPhone owners than like FeiTeng Mini N9300 owners.

      • howmany

        How many of them are there at each price point?

      • obarthelemy

        Actual models ? Don’t know, don’t care… How’s that relevant ? Unit sales, revenues, and profits for each segment are relevant, but actual models available, I don’t see the point, especially since a lot of them at the low end are rebadges. For the purely Chinese producers,

        http://www.geekbuying.com/category/SmartPhones-315/?c=&v=

        probably gives a broad idea of the prevalence of certain SKUs/models of phones (by screen size, RAM, cores, Android version…), though I’m sure not even unit sales match those, let alone revenues nor profits.

      • howmany

        I was talking about sales (pick units or revenue), hopefully obvious from context.

  • peter

    There is an interesting natural experiment. The iPhone is an exceptionally high margin product that relies on distribution by and subsidies from Carriers.

    Given that iPhones share network effects and ecosystems with the iPad and iPod Touch, it should be possible to do some comparative analysis because the iPad/iPod have lower margins (and the price gap with Android is smaller) and do not rely on Carrier distribution or subsidies.

    For example:
    1) the iPhone is weak in Germany but strong in the UK — does the same difference exist with respect to the iPad/iPod?
    2) the iPhone is weak in India — is the iPad/iPod also weak in India? etc.

    With a couple of these comparisons, it should be possible to start identifying who or what is the culprit behind these country/regional differences (price level, subsidy level, language specific App selection, monopoly power).

    • charly

      In India a lot of people have a tablet but not a pc. That isn’t the case in Germany. That adds a whole new level

  • Luis Alejandro Masanti

    Horace:
    Can you relate the “insularity” of markets with the economic recession/status of the different countries and dates?

    Here in Argentina, were I live, the “cost” of an iPhone has gone so high with respect to our “pockets” that became like a Ferrary-to-go-to-the-supermarket.
    In the same way, old glories of Blackberry are held due to incapacity to change the cellphone.

  • Chaka10

    If I may refer back to a point Horace previously made (S is for Service), smartphone consumers purchase and pay for mobile connectivity services that is enabled by, and is really a bundle of, both (1) the device and its technological capabilities and also (2) carrier mobile network services. I believe that realizing this key difference in the smartphone situation (vs PC) etc., ] is the key to understanding the apparent puzzle of the different smartphone markets.

    After all, it seems quite obvious from this bundled services view of mobile connectivity services that the shape of the smartphone market in any country would depend very materially on the status of the mobile telecom services market in that country.

    Why would smartphone markets be the same in countries with 3G penetration in the twenties (China; single digits in India) as in countries with 70%+ 3G penetration?

    Why is it surprising that smartphone markets are different in countries that are well into 4G adoption from countries where that is just beginning or are facing structural impediments? It has been recently broadly reported that Europe has proposed sweeping telecoms reforms (e.g., VentureBeat’s piece, “Europe’s most ambitious plan yet to fix the telecoms market”). A recent report from the GSM Association said “Europe…now lagging…in the deployment of mobile broadband, particularly in 4G/LTE”.

    Ben Evans pointed out in a piece last year that the US mobile market has significantly higher ARPU and usage than the EU (I do not presume to reproduce, but refer readers to, the second chart in Ben’s piece “iPhone Pricing and US Market Share”). ARPU levels provide a simple metric for the status of the mobile telecoms market of different countries, and seems to have the strongest correlation to iPhone share (and I would guess, more broadly, high-end smartphone share).

    In theory, the relationship between smartphones and mobile telecom usage should be a virtuous cycle:

    higher ARPU means phone cost less significant to total cost of ownership (TCO) and carriers can better afford subsidies,

    lower phone cost relative to TCO and better carrier subsidies drive high end smartphone adoption (iPhones),

    increased high end smartphone adoption (iPhones), drives higher ARPU, feeding back into 1 above.

    However, many factors go into the development of the mobile telecoms market in any given country (including VAT and import duties that can be the biggest factor in inflating device prices). Vendors (Apple) can’t start or accelerate the cycle themselves, certainly not only with pricing.

    Importantly, this does not mean to me that Apple should not target China (or other EM) with a cheaper phone. But, a decision to do so needs to be recognized as a decision largely to target emerging 2G-3G transitions and should be done on a targeted, specific fashion. NOT on a one-size-fits all basis, with a broadly distributed cheap 4G LTE product!

    Though telecoms markets are different, carriers world wide operate under two common realities — (1) they offer a scarce resource (their networks can support only so many users using so much capacity) and (2) their networks are sunk costs. That means the focus of carriers world wide is to gain as many high ARPU users as possible, and even more so, on a committed (contract) basis. That’s why they pay subsidies and suggestions that there’s some sort of slight of hand that fool users are exaggerated. By the way, this also translates in emerging markets, e.g., CHU and CHL pay high subsidies on high ARPU plans.

    • obarthelemy

      I’m not sure you can read too much into the ARPU. I’m paying (France) $20 for unlimited everything (landline+mobile national calls and texts; landline to 80-ish countries; mobile to 40-ish countries; 3GB data then throttled; tethering; any protocol incl.VOIP, torrents….). I’m sure I could pay more… but what for ? With all the wifi around me, I’m using 500MB data/month….
      Sure, it’s not 4G yet, and I’m not streaming stuff (that’s what the SD card and integrated FM radio are for), and I’m setup to do software updates over wifi only…
      I see things the other way: networks are sunk costs, marginal cost to support an extra user is nil, the game is to attract as many users as possible, whether high-revenue or low-revenue.
      I’d be curious to know how many sites are bandwidth-constrained either at the tower stage or at the back end.

      • neutrino23

        Slightly off topic, but how can your plan be so cheap? Roughly speaking we pay $60 or $70 a month for voice and data whether we get a phone from the cell company or supply our own. I’m not aware that the cell companies are wildly profitable. How can they support the infrastructure in France for so little money? Is the US more costly because of the size of the country and low population density in many areas?

      • obarthelemy

        First, all carriers use the same standards, so there’s infrastructure sharing in low-density areas, instead of 4 carriers setting up 4 distinct towers to serve 1,000 people and 10,000 sheep.
        Second, all the mobile companies also do ADSL, so I’m guessing the backhaul costs are minimal: the fiber is already there, as is the active equipment in the local network node. At least one of the carrier is actually using a tweaked version of their ADSL box in their mobile towers.
        At least one carrier is also using the EAP-SIM spec to switch mobile traffic to wifi whenever one of their customer’s ADSL box is in range (supposedly out-of-band so no impact on the customer’s bandwidth)
        And finally, I’m not sure technical costs in the US are much higher: there seems to be a whole lot of extra marketing, middle management, upper management, and shareholders costs, but I remember seeing investment plans form one of the top 3 and thinking it was a measly %age of expenditures. Sorry, no link.

      • http://fundless-sponsor.blogspot.com/ Capitalistic

        Mobile devices and telecom hardware are heavily subsidized in the US compared to Europe.

      • charly

        24 months to pay of a 600 phone costs at least $25 a month for the phone company. And that is without interest and deadbeats.

        And obarthelemy is probably talking about 20 euro/months but that is close to $20 without VAT. I also think that European operators make more from roaming charges outside the home country. And they are less profitable. But the main difference is that they don’t have local monopoly power like they have in the US

      • obarthelemy

        that’s 16€, with VAT included. Actually 15.99€. $21.61 it seems.

      • handleym

        Here’s one possible answer:
        http://gigaom.com/2012/04/26/free-france-churn/

        To be fair, I have no idea how relevant this is today, 16 months later. ATT of course was hit with a wave of similar complaints, but are pretty good today. Of course to BECOME pretty good from where they were required investing a LOT of money — which presumably is not an option for Free.

        If I had to guess, I’d say obarthelemy is in the position where he does not need to demand much from his mobile data network, but he persists (even after being corrected repeatedly) in claiming that everyone in the world has the same limited needs as him.

      • Chaka10

        US has higher usage, not just ARPU. So it’s not a matter of pricing as I’ve seen suggested.

      • Walt French

        In many parts of the US, mobile usage expanded so rapidly that carriers were overwhelmed and have had to add extra capacity. Although I’d assume similar restrictions in Europe, NIMBY and other red tape means a new tower can mean 3 years of delay and (partly as a result) no small expense.

        Meanwhile, people were unable to get connected when they’re within a block of the cell tower and see 5 bars. I’m not talking about Podunk Cellular in Kentucky, but AT&T in NYC.

        That problem has been greatly ameliorated in the past 3 years, and I don’t think it’s due to magic. Rather, carriers have violated all your assumptions and forced subscribers to limited data plans (because competitors couldn’t afford to offer it, either) and built towers or added higher-speed virtual circuits. These of course have genuine marginal costs.

    • davel

      I think you are right. The robustness of high speed data networks and the relative wealth of the population has a direct relation to what kind of mobile services are popular.

      I am not sure why there is such a difference in Europe and USA. I do know the out of pocket costs differ and perhaps the value the consumer places on the services differ as well.

    • Oak Grove Capital

      ” … carriers world wide operate under two common realities — (1) they offer a scarce resource (their networks can support only so many users using so much capacity) and (2) their networks are sunk costs. That means the focus of carriers world wide is to gain as many high ARPU users as possible, and even more so, on a committed (contract) basis.”

      Exactly. Carriers worldwide have high fixed costs in the ground or will. Payback can only come from selling the data that the infrastructure delivers. If the iPhone remains the best data salesman, it bodes well for Apple over time across geographies and markets.

      Many of the carriers worldwide are burdened by monopolist (and sometimes statist .. see China Mobile) cultures that resist/are slow to change. Ultimately, in this global economy, business/capital dynamics win. NTT Docomo, Verizon, Sprint, T-Mobile, China Telecom, and soon, China Mobile have no choice but to accept Apple’s terms to attract the best data consumers. If they don’t they get no payback on their high fixed capital costs. It is a good story for Apple and I’m sticking to it. :)

    • Oak Grove Capital

      ” …. importantly, this does not mean to me that Apple should not target China (or other EM) with a cheaper phone. But, a decision to do so needs to be recognized as a decision largely to target emerging 2G-3G transitions and should be done on a targeted, specific fashion. NOT on a one-size-fits all basis, with a broadly distributed cheap 4G LTE product!”

      The view that Apple needs to move its price point down to grow now is flawed. First, we, in the west, don’t seem to be able to comprehend the sheer numbers in China and India. A smaller percentage of a much larger number is a bigger number. That is point number one. Second, EM is not homogeneous. Too many categorize EM as if it were. Across developing markets today there are multiple more upper middle and higher end consumers than Apple serves in the US. Growth of that segment in EM will be robust for Apple. It has just begun.

      • Chaka10

        Yes. As I posted recently elsewhere:

        “China is interesting because while they have 3G penetration only in the twenties, that small percentage translates to big numbers (300 mm 3G users, reportedly) and they are about to launch 4G. That means that there’s a very sizable part of the market (small proportionately, but large in absolute terms), that is “there”. Unfortunately about half of that market is China Mobile. It’s important for Apple to do well in the available addressable market in China, and fortunately I believe they are in fact doing well. But that is lost entirely, however, if you look at total share numbers because, again, the addressable market for Apple small in RELATIVE terms. This distortion is so big that it (and other large EMs) results in the unfortunate reports of alarming declines in Apple global market share.”

        I would refer you to some of my other posts if you’re interest (I give some views on Europe etc). I think it’s easy enough to lookup on Disqus.

      • Oak Grove Capital

        Yes. We agree. :)

      • Chaka10

        Yup. Another excerpt below from an older post I put on Asymco (S is for Service):

        “SO MY SECOND MAIN POINT IS, while the iPhone had a 9% share of total smartphone shipments in China for 1Q13 according to IDC, I estimate that it had a 30% share (30%!!) of total 3G smartphone shipments by China Unicom and China Telecom. That seems like a better reflection of iPhone’s competitive performance in China.
        Assuming China Mobile is signed, and assuming Apple can reach, even with it’s existing offerings and pricing, the same 30% share it has of China Unicom and China Telecom 3G smartphone shipments, that would imply an additional more than 8 mm units. Again, that’s just assuming that China Mobile’s users react the same way to Apples existing offerings (without any discounts or cheap phones). That would result in Apple’s share of total China smartphone sales (even including any 2G “smartphones) reaching 19%, more than double it’s current share.”

      • Oak Grove Capital

        Agreed but, 30% of 150 million 3G subscribers is 45 million units. What is the basis for your 8 mm units. Assuming all else equal (admittedly a stretch), I would expect more than 8 mm. I understand that both at Verizon and T-Mobile, Apple reached more than 30% share in the launch quarter.

      • Chaka10

        That’s the danger of giving just an excerpt. I would welcome your scrub of my math in the complete piece, but I think the confusion is that I assume your 150mm refers to an estimate of China mobile installed base (50% of 300 mm), whereas my 8 mm (actually 8.4 mm) estimate is of unit shipments, specifically only for C1Q13 (calculated as 30% (ie, the iPhone share I estimate of CHU and CHL 3G shipments) times the 28 mm 3G shipments by China Mobile in that period, as reported by IDC).

        Note, this is an estimate, but it’s actually checks out pretty well against other rough estimate methods. For example, reported iPhone shipments of 37.4 mm for C1Q13. China represented ~20% of Apple’s net sales in that period. Apple doesn’t give iPhone sales numbers in China, but assuming it’s proportionate to net sales, ie ~20%, that would give ~8 mm units of iPhones shipped in China for CHU and CHL, and if you assume China Mobile is about equal to CHU and CHL aggregated, that checks out against the 8 mm incremental units estimated using IDC numbers.

        Again, these are all estimates, but the implication is that APPLE COULD DOUBLE ITS CHINA IPHONE SHIPMENTS BY ADDING CHINA MOBILE, which by me estimates could be an additional 25 mm units on a full year basis applying a 16% China share of Apple iPhone sales to 150 mm total units for Apple’s fiscal ’13. Let that sink in … ~25 mm units incremental (for perspective, that compares to 32 mm total units for all of Apple in the last quarter).

      • Oak Grove Capital

        Thanks for the clarification. I understand your assumptions now.
        Here is a clarification of my math process. I had read that the iPhone immediately (ie in launch quarter) attained a 45% share of all Verizon subscribers, not just of sales in the 1st quarter. If that is the high water mark boundary, then assuming China Mobile has ~150 million 3G subscribers, the backlog of subscribers who have been waiting for the iPhone – including the approx. 40 million that already have been using an iPhone – might cause an initial launch spike in conversions to the iPhone.

        If a spike were on the order of Verizon (admittedly unlikely for many reasons), the number in the 1st quarter would be 45% of 150 mm or 67.5 million! Even assuming 1/3 less than Verizon, the number is 45 million.

        I think those numbers are grandiose but, the sheer magnitude of the numbers combined with the precedent for the math working with carriers in the past means that estimates for 1st quarter spike just might be too low. Even if, the conversion rate is one year and 30% of 3/4G subscribers convert, the 1st year unit sales could be well north of 45 million. Imagine that.

      • Chaka10

        “the iPhone immediately (ie in launch quarter) attained a 45% share of all Verizon subscribers”. Can you share a link to that data? Seems a bit incredible but would be great.

      • Oak Grove Capital

        I found it in a tweet from Ben Evans:
        User Actions
        Following

        Benedict Evans
        ‏@BenedictEvans
        $AAPL When VZW started selling iPhones, sales didn’t ramp slowly – they went right in at 44%. Outlook for China Mob? pic.twitter.com/2OwIcJMIkg
        Reply Retweet Favorite More

      • Oak Grove Capital

        Just realized that he might have meant they went from 0 to 44% share of new sales in the launch quarter instead of 44% of entire subscriber base in launch quarter.

      • Chaka10

        Well, even my more conservative estimates aren’t so “bad”, showing a potential doubling of Apple’s China iPhone sales by addîng China Mobile. Have you tried to estimate EPS impact from an additional 25 mm iPhones a year?

      • Oak Grove Capital

        Very rough, back-of-the-envelope, calc, assuming 52% G/M and $600 ASP would be a min. of $2.50ish up to as much as $4.20ish. That also assumes 900 mm shares and an isolated net margin on the iPhone of between 15 and 25%. This last part is the roughest part of the calculation. Do you have any input on what the iPhone net margin is? The ASP and G/M assumptions depend on the mix and margins of the 5s, 5c, 4s, and 4 in China. Citicorp estimated that the 5c G/M will be around 56%. That is higher than it has been for the high end phones.

      • Chaka10

        Well, op ex consist of r&d and sga. No reason to think adding CHL would increase r&d and I’d think there would be operating leverage, ie sga wouldn’t increase exactly ratably, but assume it does. That’s 7% of net sales in Apple’s last quarter. Other than that, it’s just tax which seems to be ~20% of EBIT. Using your assumptions on GM and share count, that’s an add’l ~$5+ to EPS. Right?

      • Oak Grove Capital

        Right, assuming 900 mm shares, 52% G/M, $600 ASP, and a net margin of 30%, $5/share.

      • Chaka10

        I assume you saw that IDC came out with a report yesterday which concurs with my est that Apple could double its iPhone sales in China by adding China Mobile?

        http://blogs.wsj.com/digits/2013/09/24/idc-sees-apple-doubling-china-smartphone-market-share/

      • Oak Grove Capital

        I did. I assume you will agree with me that IDC probably is underestimating Apple, as it typically does. :)

      • Chaka10

        Yeah, rather than just continuing the drumbeat of “Apple declining share in China” mantra, they themselves could have come up with this earlier than I did (the analysis is based on their data after all). Shame they didn’t but better late than never (and the delay is an investment oppt)

      • beidaren

        should we really call 2G “smart phone” smart? isn’t much of the Android offering the new dumb phone?

      • Chaka10

        Yep, that’s part of my point. Importantly also is the carrier angle — as China Unicom’s presentation slide shows nicely, there’s a big ARPU bump with mobile broadband (3G, 4G) adoption, which means carriers are better able and incentivized to pay more subsidies on high-end smartphones (iPhones in particular), which then drives more mobile broadband adoption, data usage and ARPU, which cycles back …

      • charly

        Single digit market share means that a lot of apps are not made for your OS. This isn’t so much a problem in the paid market where total number of users is most important, but it is a killer in costumer service apps.

      • Oak Grove Capital

        Apps developers want to be paid. Note that the best apps are written for iOS first.

      • charly

        The must haves for most people are apps like banking, shopping etc. Those developers are being paid just not by the user but by the bank, shop etc. And they may be written first for IOS users in the US but i would be seriously be surprised if that was true in Western Europe, let alone in markets like India or China

      • Oak Grove Capital

        It is just the category of apps you mentioned that do not favor android. Glitchy, malware vulnerable, insecure, fragmented mobile software platforms are not where the app developers want to be.

      • charly

        They may not want to be there but it is where their users are. Besides in markets like India Samsung android 4.2 probably has a larger market share than all IOS combined

      • Oak Grove Capital

        Unless you can cite a link, I doubt it. But, that isn’t the 1st problem with your argument. Look at global mobile web usage data, iOS dominates for a reason. Mobile commerce and mobile functionality that improves people’s lives follows the mobile web. Without it you have potential. WIth it, you have potential realized. iOS wins when the potential is realizable.

      • charly

        What is apple smartphone market share in India? I’m to lazy to look it up and i will be the first to admit that those data are probably wrong grey imports and such but i will eat my hat if mobile phone networks see more than 5% of their smartphone users use it. The biggest company version of andriod has undoubtable a higher market share.

        About world usage. If you target the local India market you don’t care about world usage but Indian usage. There is also every indication that Andriod will strongly lead that market in the very near future. There is also the OSX riddle which always had a much higher share of web usage than one would expect from even a very optimistic analyses of Apple market share in the pc market.

      • USisthePotOfGold

        “About world usage. If you target the local India market you don’t care about world usage but Indian usage”
        PRECISELY!!
        What is by far the biggest market for generating app revenue?
        What market has, by far, the most app developers?
        In what market is iOS the only platform that saw growth last quarter?
        The answer to all: the U.S.
        I’m not saying that Apple wouldn’t love to serve the emerging markets, but let’s get real here. The pot of gold is still the U.S.

      • charly

        The pot of gold may be in the US but the paycheck is local. The way the average profession developer earns his living is by making custom software not shrink wrap.

      • Walt French

        I suppose there are some down-market firms that cater to lower-income consumers.

        But probably not too many banks or the fancy retailers who like to have apps for their best customers.

        Meanwhile, these apps are to attract the high-end customers, just as iPhones are “sales agents” for high-speed data. Could very well be that one higher-income user satisfied this way is more valuable than a dozen lower-income ones.

        In fact, if fraud comes into the picture, banks and others might want to encourage only those customers least likely to have their devices compromised, exacerbating the split.

        No data, sorry, but a workable model worth considering.

      • GoWithTheFacts

        Bingo.
        People can theorize all they want that the simple hugeness of the global Android market will inevitably cause it to dominate the efforts and attentions of app developers.
        Thing is, the darn facts don’t support their theory. Today (not in some mythic future) Android has overwhelming global market share. And today, in the face of that Android dominance), iOS still clearly rules in the developer world.
        The fact that the situation is even better for iOS in the tablet universe will only help to preserve the lead for iOS.
        Facts are stubborn.

      • Chaka10

        Single digit market share where? China? I estimate that Apple had 30% of China Unicom’s and China Telecom’s 3G smartphone unit sales for 1Q13 (most recent data I have). India? India has single digit 3G penetration currently, and doesn’t really show up materially on Google Play, so what kinds of apps are being run, exactly, on Indian smartphones that iPhones are at risk of missing out on?

      • charly

        30% of 3G capable handsets? I seriously doubt that. You assume that only 3G users would buy an iphone, the most fashionable phone in the world, or that 2G network users wont buy a 3G capable phone. There is also the fact that 2G phones are almost substitutes for 3G phones. The only thing they can’t do as almost as well as G3 is video

        Kingfisher boarding pass app would be a prim example of an app Indian iphone users would miss out on. And those are apps that are the true killer apps for mobile phones.

        ps. I don’t know or even care if kingfisher has such an app. I’m not an expert on India, couldn’t even name a bank there but boarding pass, banking, public transportation, road closure etc. apps are the real killer apps in the smartphone world. Those are not made for 1% market share operating systems. Not so much because of money but of a limit of attention.

      • Chaka10

        Don’t just doubt my numbers, they are not a guess but base on calculations. Look at my analysis and pick if apart but disagreeing based on your gut sense is not helpful.

  • obarthelemy

    If smartphones are partly fashion accessories (and gosh do they seem to be !), there’s no reason for smartphones tastes to be any more similar than fashion tastes. And I can usually tell someone’s country of origin as much from their clothes as from their complexion…. Even on the inside, I fully expect some cultures to be more sensitive to openness and egalitarianism, and others to looks and exclusiveness.

    If smartphones are partly tools, we can expect preferences to vary depending on the costs and benefits. ie, purchasing power, carriers prices and policies (subsidies, lock in…) and jobs to be done (movies are still unavailable in many countries, making LTE less important, but an SD slot to store torrented movies, more important…). Also, if desktop/laptop/tablet penetration is high, the jobs to be done by a phone are different than if the phone is the sole computing device. Etc, etc…

    I think having Samsung’s detailed sales (which we’ll never get) would be more interesting than doing an Apple vs ROW study: Samsung have better distribution, and a wide range of models available mostly everywhere at similar relative positioning, so local tastes would be easier to spot.

    As it is, Apple vs ROW amalgamates too many issues (high-end vs mid/low end, importance of looks, importance of openness, success of ad campaigns, Store density, carriers strategy regarding Apple/Android/…).I’m not sure what can be gleaned from that apart from “Some countries like Apple disproportionately, some dislike them as much”.

  • Sander van der Wal

    Regarding the network effect, it could be possible that a rich enough or big enough country is able to support a viable ecosystem. The PC conquered the world speaking American English. All the horizontal apps were in English, and the people buying them had English as a second or first language. Compare that to smartphones, where localisation has always been available.

    And secondly, one could sell apps for a small geographical area. Much easier to program compared to the first PC apps, and much cheaper to buy, so more people buying them.

  • Naofumi

    Horace, if you look at the market for Apps (look up the AppAnnie report on their blog in collaboration with IDC), local is even more extreme. While the number of free downloads roughly correlate with the population of users, paid apps are skewed towards high GDP per capita countries on both Google Play and the AppStore. Interestingly, google play revenue is unhealthily dependent on Japan and Korea with Western countries contributing quite little, whereas AppStore is a more natural distribution.

    I would argue that since app stores tend to have little local sales, marketing, distribution effects and are even more Global, there is no mechanism to buffer income disparity and hence local effects are more pronounced.

  • captialism

    Android is vastly more popular in India – Indian BUY mobiles. No wonder Android is popular since it comes at many price ranges. iPhone never had a chance to gain a significant share. WP would have to be severely discounted to gain some share.

    The question you should ask as a analyst would be the share of ios in USA if mobile were purchased instead of subsidized.

    • Mark Jones

      Of what value would be the answer to that hypothetical question? Wouldn’t an analyst’s time be better spent working with the real historical data from the USA, India, and other countries?

      • captialism

        asymco is not good with historical data. I believe he concluded once that android would not gain more then 20% share cause hew ignore markets where mobile were not subsidized!

      • Walt French

        You couldn’t possibly have made a worse argument. Between your fuzzy recollection, incoherent statements and even your strange screen name, your post simply screams, “a hurried, one-off ‘identity’ for the purpose of throwing out garbage.”

      • Chaka10

        I am loath to answer this poster, so if I may, I will direct my answer to you. There are so many differences between India and the USA. Subsidies are the least of them. If India were further along with mobile broadband adoption (it’s still in single digits on 3G penetration) and if it had higher ARPU/mobile usage (the two are related), I would expect that Indian carriers would pay subsidies for high value plans (just like China Unicom and China Telecom).

        Edit to add: Frankly, I think it’s an indictment of Android that its popularity is disproportionately in poor countries that are largely pre-mobile broadband.

  • Ian Ollmann

    How does divergence between countries dispute the theory that platform predominance is principally due to network effects? It seems to me the fact that most adoption curves for the most popular platforms correspond to a higher order (3rd or 4th power) curve would suggest that there are strong network effects at play here. That is, if three or four of your friends or family all have the same phone, you are likely to get that one. Isn’t the Nokia texting home run a prime example?

    Let us suppose for the moment that network effects are the prime motivator for most first time buyers, and platform loyalty (provided the platform is worthy) thereafter. Then we expect that platforms that take an early lead in a market may be able to persist despite superior competitors, and that which platform dominates a particular country may be subject to capricious whim like who offered the right BOGO deal, or a certain killer feature at a critical time when a new wave of platform adoption was taking off. Other markets for which the timing was not right would shrug off the feature and cling to something else when they were ripe. So, the finding that different regions (with different languages and different cultural barriers) pick different winners disputes network effects how?

    There is always the opportunity to unseat the incumbent as you so often describe, but it remains nonetheless extremely difficult, which is why when it happens it is fascinating and rare.

  • poke

    The PC market started out heterogeneous (before the IBM-compatible PC established dominance). There were pockets of different platforms around the world. Atari, Commodore, Apple and other manufacturers were more or less popular in various countries (I remember having to reluctantly give up my beloved Amiga for an awful Windows PC). It was a patchwork. I think the continuing heterogeneity of the smartphone market reflects the fact that it simply lacks maturity in platform terms. The two major platforms are only 5 years old. Plus, there’s still a large market for smartphones that are just phones with web browsers, and that’s distorting platform adoption figures.