iPhone sine qua non

Last week I made an attempt to measure the iPhone’s manufacturing cost given new data points from the Foxconn field trip. The post generated a great amount of new knowledge and the feedback was very valuable.

The main value to me came from stepping back and looking at the entire cost and value structure for the iPhone. Putting costs into perspective is as valuable as knowing what they are.

The following diagram shows my estimates for this cost structure for the fourth quarter given both bill of materials estimates and the other parts of the cost of goods sold and operational expenses and even ancillary sources of revenue.

Source for BOM estimate: iSupply.

There are several observations easily made from this view:

  1. The manufacturing cost ($15 may be very generous i.e. high) but it still amounts to a tiny fraction of the device cost and hence value-add. The direct labor part of that cost is tinier still.
  2. Depreciation on capital equipment (Apple owned) amounts to $10 per device, probably higher than the direct labor cost.
  3. External developers as a group receive more app income than the salaries and costs associated with Apple’s own developers working on the product.
  4. The gross margin percent is 58%.
  5. The operating margin is estimated at 51%

The high margin that the product generates is even more surprising given the industry average as seen in this summary view of the industry’s performance over three years’ fourth quarters.

The operating margin is the ratio of solid colors to the entire bar.

As someone pointed out on twitter: one of these is not like the others.

The question should be: How is this possible? What does this product have that gives it such a pricing advantage? Note the the ratio was preserved through the three years shown and has persisted for nearly five years.

The answer to this paradox is in data that is not visible in any of the diagrams above: The largest revenue attached to the iPhone comes into the mobile network operators. We don’t have global averages but the revenue the iPhone generates over its lifetime is probably two to three times the revenues that Apple directly receives. The ARPU iPhones are attached to allows for a substantial subsidy, but more importantly, loyalty (or churn)

As seen in the small but clearly visible US market, being an operator having  iPhone exclusivity generates above-average growth and being the only operator to be excluded from the product generates above average subscriber losses.

It quickly becomes obvious that ranging the iPhone has moved from being an advantage to not doing so becoming a serious disadvantage. The product and its high up-front costs are “hygiene factors” for operators. The sine qua non status means that it can still be priced with the extraordinary built-in margin shown above.

  • Walt French

    “The answer to this paradox is …[that t]he largest revenue attached to the iPhone comes into the mobile network operators.”

    This plays nicely into the observation that, at least in the US, essentially all phones except Apples’ are specified, purchased, branded and distributed as wireless carrier products. Motorola just happened back in 2008 to be an independent company when Verizon was using it to build the Verizon Droid brand; the proper unit of measurement for growth, profitability, etc., is by carrier.

    Except for Apple, which figured out how to avoid the carriers’ “barefoot and pregnant” oligopsony policies of treating manufacturers. Maybe all those Jobs jabs at “orifices” were more important in defining the company’s future success than we realized.

    • Louis

      I think it isn’t quite so simple.  What Apple figured out is that AT&T and VZN are willing to pay a great deal in subsidies to defend the “long expensive contract” business model.  With less-generous subsidies, more people would just buy and unlocked phone from Apple and then start looking for lower-priced prepaid or unbundled plans.  (As is the usual setup on the Continent, where the iPhone also has worse market share.)

      The iPhone’s desirability makes the relationship with Apple more important than that with the MNOs for customers, which should drive down prices.  In the US, what happened instead is the MNOs handed over a lot of their revenue to Apple as a way of avoiding a price war.

      • Guest

        I really have to agree with this, you make a good point. I think Apple stands to gain in an environment where phones are sold subsidized rater than at full cost. Essentially, the high price of the phone is “hidden” from the consumer, because it’s hidden in the monthly contract you agree to. It makes it extremely tempting to keep re-upping your contract every two years to get the newest and exciting iPhone. It’s actually quite a brilliant psychological play by carriers and Apple.

        So my question is: do iPhone owners in the U.S. have a more frequent iPhone-purchasing rate than European iPhone owners? It seems like the answer would be yes, since the average consumer finds it easier to stomach $200 every two years than $600+ every two years. Which is precisely why it seems to me that Apple must prefer the U.S. environment over the European environment, all things being equal.

      • r.d

        Except, Apple originally sold the iphone as unsubsidized
        with low rate sub.  Three months later reduced the price.
        a year later AT&T renegotiated and Apple relented.
        and now Carriers are crying about giving their profits to Apple.

      • I don’t know how common this is with other users, but I didn’t upgrade my iPhone 3G until the 4S came out (about 3 years), and my wife only upgraded her original iPhone at the same time (about 4 years), so AT&T made off just fine with all our extra subsidy money. Just because we *could* upgrade our phones didn’t mean we actually *wanted* to.  This may be an example of Horace’s “good enough” in action.

      • Walt French

        Presuming you want to treat your carrier commitment as permanent, the CW is that as soon as you can upgrade, you do so for your $200 or whatever, then turn around and sell an only-slightly-depreciated 3GS for more than $200.

        Your strategy may be good if your time to sell the old one is worth more than minimum wage, but if you have an easy buyer (e.g., a child who will gratefully accept it, even as an iPod Touch), the subsidization model encourages over-consumption.

      • Partly I was hoping to see some more carrier choice, but as it turned out, a choice of black, black, or black for basic plans isn’t much incentive to switch carriers.

        I wound up sticking with AT&T mostly because their data service is faster, at the moment. I frankly don’t like any of the major carriers, but we’re unlikely to get much choice here in the US unless there’s an unseen disruptor in the works who can bypass the major carrier lock on spectrum somehow. Not holding my breath on that one.

      • Walt French

        Don’t have a link but look up Martin Geddes (sp?) maybe at GigaOm for “the end of telcos” type story. I think it fits Horace’s understanding, too, that telcos have very expensive cap ex but they aren’t really in command of “value” that people look for, any more. The argument is that our spectrum allocation is really FUBAR (haven’t evaluated it myself but I’m receptive) so carriers over-spend to keep users rather than provide service. Sound familiar?

      • It appears this is exactly the case, as the WSJ wrote yesterday: iPhone’s Crutch of Subsidies

    • Having seen VZ from the inside a little bit, I’m pretty sure there were dead VZ executive bodies on the floor before they accepted Apple’s terms for the iPhone. They do *not* play nice with vendors….

      • Walt French

        I, of course, do NOT have that inside info, but it sure sounds right.

        Incidentally, I had a nice chat with an airplane seat mate a while back; he said that some half dozen VZ staffers took especially early retirement after turning Apple away back in ’06. Yes, the reaction to Jobs & Co. was probably “don’t let the door hit your butt on the way out,” but the iPhone caused a major shift in competitive positioning in wireless in ’07 & ’08; had Android not been available it could’ve been EXTREMELY expensive to Verizon.

      • I wasn’t very inside… I wasn’t in the telco itself, but in a small R&D appendage where I got to deal with some of the mainstream VZ folks a little, and saw a bit of how the management thought of things. (I will also note in passing that I think quite highly of their engineering team, in general.)

  • Guest

    Hey, dumb question: does “memory” refer to NAND flash storage or RAM? I’m assuming the former. If so, we’re does RAM costs fit into this? Is it lumped into “processor” since it’s (I believe) part of the system-on-chip?

    • Walt French

      iSupply’s original iPhone tear downs showed RAM and Flash as approximately equal costs, adding up to MORE than the CPU, IIRC.

      • Guest

        Ok, so for anyone else interested, it looks like iSupply (per Horace’s link above) groups both the NAND flash and the RAM under “memory”.

      • Walt French

        There must be some good reason you thought your question important … care to share?

      • Guest

        Sure, but it’s not too interesting: I selfishly would like some more RAM on my next iPhone 🙂

        MobileSafari on the 4S (and all previous iPhones) kicks webpages out of memory too frequently for my liking. More RAM would be nice, and I wonder how expensive of a proposition that would be for Apple. Admittedly, I don’t hear average Joe user complaining about this.

      • Walt French

        I can’t/wouldn’t disagree with your thinking. However, …

        Doubling RAM might cost an additional $25 or so per handset. The maybe 10% of us who keep lots of apps/pages going might notice an improvement, while 90% wouldn’t notice any benefit, meaning that it would have to provide $250 of benefit to the one person out of ten who’d be subsidized by the other nine.

        If Apple employs any logic roughly like this, we can understand why their choice. I don’t think that *I* would pay an extra $100, but maybe. Apple has always been extremely tight with RAM on its iPhones and I think this exemplifies why.

        Now, as to whether people would pay an extra $50 for upgraded memory? That’s more interesting. It could have the effect of encouraging app developers to write for a much bigger RAM partition, with effects up and down the system.

      • I suspect it’s not a cost issue — more likely it’s a battery issue. Flash doesn’t take power when it’s not being accessed, but DRAM does, and the more memory that needs to be refreshed, the more power it takes. I believe many of the Android phones have more DRAM memory than the iPhone (possibly because of a larger OS footprint), and this may be one of the reasons why their battery life often isn’t as good. (As an aside, I think this DRAM power observation first surfaced over on AppleInsider.)

        The power issue isn’t a simple linear relation, due to various design tradeoffs and process effects, but DRAM from the same technology generation will generally use more power the larger it is.

        Apple is a master of design tradeoffs between software, hardware, cost, and function, and I expect this one is quite deliberate, but not particularly cost-driven.

      • I believe that Apple has constrained the RAM on all iOS devices not so much because they are cheapskates but because RAM burns power. We’ve finally hit the point where the amount of RAM that can be kept alive given the overall power budget is not painfully low, but I don’t think you’re going to see Apple throwing more RAM in devices just because “what the heck” for a while. With mobile, you ALWAYS have to think of the power costs of any addition you make.

        Now, if Apple or ARM could add features to a CPU+chipset that somehow changed this equation —the ability to simply power off blocks of RAM, or run them much slower, the tradeoffs would change. But with current technology it seems to me this would also require changes in how RAM is delivered — something like individual independent sub”blocks” of RAM on a given DRAM chip — and that is a substantial change from what the DRAM vendors are set up to deliver today.

  • Luis Masanti

    So, building an “user desirable” cell phone, with an [almost] “user desirable” refresh rate (yearly) while keeping the OS updated for a [certanly] “user desirable” period (3+ years) and not paying attention to telcos is a good business for [almost*] everybody! (even to the telcos!)

    * Exclude competence from this “almost.” 

  • Louis

    In Germany, you can buy an iPhone for €650 or so from Apple.  There are then any number of €10/month prepaid data plans, putting your total cost per month, if you don’t talk much on the phone and use iMessage, in a zone where Apple gets more revenue over two years.

    So the paradox resolution seems to depend in a fundamental way on the US market being split between two serious players who offer only expensive long-term contracts for data.  They are willing to heavily subsidize buying the iPhone as a way of protecting the business model, which has to be much, much more profitable for them than competing against 10 different MVNOs.

    • You can buy an unlocked iPhone in the US as well. Not many do.

      • Louis

        Since there is no cheap data contract, why would you?

      • Canucker

        This is the anomaly. The only benefits to buying an unlocked device in North America is the ability to buy local SIMs when travelling and to be able to change carrier in North America if you want to. The data plans are identical whether you buy locked or unlocked. This is clearly a carrier incentive to lock-in subscribers. The up-front subsidy is relative peanuts compared to the 2 year data subscription return. The cost to accrue a new subscriber (in other words the cost in losing a subscriber) is also significant, compared to the loyalty costs of subscriptions.

        Few consumers do the math on subscriber costs! 

      • Michaelndn

         Actually my wife bought an unlocked iphone  in the U.S., and she signed up for T-Mobile with a $50 plan with unlimited data and minutes. The iphone can only use the older Edge network, but now that most people have moved over to the 3G networks, the Edge network is very usable.

      • I considered it, but since the only carrier the unlocked iPhone is fully-usable on is AT&T, and they don’t allow different plan options (i.e. they’ll charge you for the subsidy even though you’re not getting it), so there’s little point to buying the unlocked phone unless you travel to Europe a lot.

        Once the cross-carrier LTE iPhone is available, there may be more incentive to customers to buy it, but for now carrier-hopping to get better deals is a non-starter.

        Another problem is that the smaller carriers apparently can’t get decent data agreements with the national carriers, which means that smartphone users are pretty much stuck with the big oligopoly boys, and an unlocked LTE iPhone won’t fix this problem.

      • Louis

        In the US, carrier hopping will be a non-starter until there are carriers to hop to.  This may well be never, given the American regulatory environment.  Nobody would move around in Europe if there weren’t a wide variety of MVNOs trying to drive down the price.

        Horace seems to have completely whiffed in his response to my other post: nobody—by which I really mean the Chinese people lined up at the store who buy all of them—buys unlocked iPhones because you pay the same monthly charge anyway.  You are, essentially, giving away $450 for little in return.The point with respect to the analysis from Horace is that the iPhone’s consistently high ASP reflects, mainly, it’s dominance in markets where *any* smartphone *usage* is expensive.  Because the iPhone is the best smartphone, Apple is then able to demand, via subsidies, an outsized chunk of the resulting revenue from the MNOs.  (Hence eroding margins from AT&T and VZN.)What I do not see is the kind of conflict between the MNOs and Apple that some are imagining: both have benefitted quite a lot from markets where the MNOs can demand $80+/mo just to have a smartphone.  It is the revenue stream that’s being fought over, but not the basic setup.Further, if we look at Apple’s new China deals, the announcements all seem to be that the distribution model isn’t based on unlocked sales by long-term contracts.On the flip side, the iPhone is not nearly as dominant in Germany—though my informal U-Bahn experience is that it’s the most popular phone—but the MNOs aren’t collecting nearly as much money from smartphone users either.

  • I would like to see side-by-side with iPad, both with and without 3G if significant.  I would think the margins on iPads are notably thinner, since iPads have more pressure to be cost effective since they are not subsidized by the carriers.

  • Anonymous

    In Europe carriers are trying to push subsidized iPhone too, rather than the prepaid model. That’s why in the Continet Apple is playing the same carrier’s game, and this is not good for customers that are forced to enter into carrier_walled_garden for getting an iPhone without spending too much money.

    • germanguy

      Well, something’s gotta give. You can either pay a lot upfront (unsubsidized iPhone) or along the way (subsidized iPhone). There is no such thing as “an iPhone without spending much money”. Or, for that matter, any other high-end smartphone; I’m sure the margins for e.g. Samsung’s top-of-the-line models are much higher than their average margin. Apple’s margin looks so extreme because they are the only brand that sells nothing but high-end models.

      • Anonymous

        I think there is a wrong perceptions of price when it comes to buy a subsididez plan. I bought an iPhone 3GS with a monthly plan in the past but I realized after few months that I was spending more money than how much I would have spent without a plan.
        So carriers are pushing for a two years contract because people prefers not to give the whole cost at once, but if someone change their mind after one year, he cannot, unless he’s ready to pay a additional feed for breaking the deal.That’s way in Continent prepaid is so strong. 
        I think this is the big difference between US and EU Continent: in US the market is controlled by carriers but Apple is the only firm who has some power over carriers because of its strong brand. Without the iPhone, in US your offer is handicapped. 

        But in EU carrier are not so strong as in US, so Apple is helping them to become stronger pushing for the same subsidized model that he can influence in US. If in US this influence could be a good thing for customers, because it is showing some weakness about US carriers, I think in EU it is a bad thing because it is helping carriers to become more relevant in the relationship between customers and carriers.

      • germanguy

        And how do you suppose Apple is “pushing” the subsidized model in the EU? I don’t see that. Actually, I think the opposite is true, at least where I live: at first the iPhone was T-Mobile exclusive and the only way to get one was a T-Mobile contract. Later, you could get a free (unlocked, unsubsidized) iPhone from England or Italy, but import wasn’t easy because the online Apple stores would ship only to their respective countries. Today, the iPhone is available from all german carriers and also unsubsidized and unlocked from Apple.

      • Anonymous

        Most of people that do wants the iPhone are not willing to spend at least 659 euro at once; if bought unlocked, the iPhone is indeed the most expensive phone available, and it is a bit frightening to spend so much money in one shot. So a customer prefers to pay it 20 or 30 euro per month. 
        Maybe in German is different, but here in Italy the majority iPhone owner I met, bought it susidized because with subscription it seems less expensive.
        Obviously I do not know every single iPhone owner, but this is my perception.

      • Kizedek

        I am not sure how you can complain about this: it is called called having “options”, where one of those options is a “contract”.

        Regarding contracts for financing a purchase: how is this exclusive to Apple, and therefore to be viewed as somehow unfair, unusual or negative?

        You can go to IKEA and buy a sofa “all at once” for 200 Euro. If you want a 2000-Euro sofa, and find that a little steep to pay in one shot, you would not be alone: a lot of people would look to get that financed. Maybe there would be a deal where you don’t pay anything for a year.

        But a contract’s a contract. You change your mind about your sofa in a year, you are still held to the payment plan or paying off the total remaining. If you don’t like it, get the 200-Euro sofa.

      • Anonymous

        I am not complaining about having options; I just think that this a less interesting option from my point of view, because it puts too much power in the carrier’s hands and at this moment the iPhone is the most efficient phone for getting this goal, and obviously carriers are taking advantage of it.

        Personally I’ll never subscribe any contract if I can, just because I like to be free to change my mind whenever I want.

        This is my own point of view, I have no problem with people that choose to buy a phone with contract. But I think that this option gives more power to carriers and this is not a good thing for customers. Then, if you cannot pay XXX euro at once, glad they can offer it to you with an option that suits your needs.

      • Louis

        The point is more that in Europe it is actually possible for Android to compete on price.  In the US, contracts are always about the same and very costly, so any cost savings to the MNO from Android aren’t getting to the end user.  This makes Apple’s job quite a lot easier.

        As long as Apple is selling everything it can make, there’s no pressure on it to chase market share in the markets where the operators have to compete with each other, though.  In these, it seems clear that the best deal is just buying the iPhone directly from Apple, as one might expect from the usual micro considerations.

  • Guest

    I have a comment, and also a question:

    It seems like the key to the iPhone’s financial success has been the partnerships with carriers, and especially the carriers’ willingness to dish out the dough to purchase high-priced iPhones, since they return such high ARPU. This seems like the perfect disruption strategy for Television as well: high priced TV sets, subsidized by cable/satellite companies, and sold to end consumers at a price that is lower than market expectations. 

    And now my question: Horace, you said “The ARPU iPhones are attached to allows for a substantial subsidy”. So is it a reasonable generalization to say that Android phones, in general, return a lower ARPU than iPhones? If so, I guess I’m not understand why that would be the case. Don’t Android data plans cost the same as iPhone data plans? How is it true that an iPhone user purchases more carrier services than an Android user? I’ve seen stats somewhere that show that iPhone users browse the web more often (not sure if that data is still current), but does that difference account for the massive difference in ARPU? My apologies if this has been explained in a previous post. If anyone can help me understand this, I’d appreciate it. 

    • Canucker

      I like the intimation of TV disruption through a cable provider subsidy model. The problem is that the business is more mature and people do not buy their TVs via their cable companies – but they do buy subsidized PVRs….  Wouldn’t that be nice for Google/Motorola to have the set-top business carpet pulled from under them? Of course, the cell phone business was also mature – it was the smartphone transition that afforded new incumbents and economics.

      As to your question, my feeling (i.e. can’t point to data) is that Android buyers tend to have cheaper plans. In Canada, at least, the smaller, budget-focussed operators (e.g. Wind) offer cheaper Android devices.  While the overall plans and tiers for data are similar between iPhone, Android and Windows Phone devices, my bet is that the iPhone plans choices tend to be at the higher data caps. Part of this may be the greater browsing activity but I think iPhone users (at least for the latest model – not so for the older models on sale) attached greater value/affinity to their device and so have greater tolerance for more expensive plans. This is clearly not an easy generalization. I am sure there are some Android users who pay for the highest tier data plan. 

    • Anonymous

      I like disruptive thinking when it comes to the potential for the TV Set market.

      Cable/Sattelite PayTV companies would need to up their monthly service plan fees to consumers to cover the increased costs of the TV Subsidy, so would be a rejigg of their business models (Whereas telco companies already have handset subsidy costs built into their business model).

      I think it would be more likely broadband service providers (some of whom may also be pay tv providers) that would like to entice people into long term contracts in return for a hardware subsidy – at present they have no “lock-in” with customers – a subsidised internet connected TV would be a nice option for them to get people locked into a 2 or 3 year service contract, and I presume they would team with apple and provide uncapped montlhy download allowances to the apple servers that consumers will be downloading/streaming from.

    • At least for US carriers, I believe nearly all smartphone buyers are forced into the same set of plans as the iPhone buyers are. I.e., you can’t have them on the network with a pre-paid SIM, and you *must* buy both a 2-year basic voice plan and a smartphone-specific data plan. Certainly this is true of AT&T, and I’m pretty sure it’s true for VZ and Sprint as well. They only seem to vary in exactly what they put in the bundle, and in the prices of the unlimited voice/text plans.

      I think one of the major US carriers is said to be playing with a pre-paid plan with some low-end Android model, but most of the pre-paid action in smartphones seems to be on the MVNOs and specialized/regional carriers, who apparently can’t offer decent data roaming outside of their core coverage — they’ve been complaining to the FCC about this, so far without much luck.

      So I think your point stands, the carrier should be reaping more revenue from selling the cheaper Android models. The carriers also seem to be selling some of the high-end Android models with up-front prices above the iPhone’s base model, I think this is mostly with the 4G models.

      I also expect the carriers to soon add a “4G premium” fee on top of the normal data plans since they are providing a higher data burst speed (with admittedly new capital expenditures required). I don’t think they’ll do this until a) they get most customers locked into 4G data usage and b) they mandate 4G on all smartphones. (I note that VZ has apparently already mandated that all new smartphones for them *must* have 4G.). AT&T did this when the 3G iPhone came out — you had to pay a $10 premium for 3G service (with the same data cap as 2G) if your phone had 3G, even if your area wasn’t covered by 3G service yet. In their defense, they actually did roll out 3G quite aggressively, so the “pay without coverage” issue didn’t last long.

    • There are now more and more Android phones being sold on prepaid plans (i.e. no handset subsidy), where the carrier’s ARPU is generally lower (no need to make up the cost of the subsidy). 

      In the prepaid market, iPhones are generally sold for $400+, while Android phones start at $70 (when I last looked in Dec 2011; could be even cheaper now). Many of these phones, made by companies like Huawei, ZTE, and Samsung, are quality-wise no better than good featurephones but they do have enough juice to run Android.

      I think the term “smartphone” has been stretched so far that it’s becoming meaningless in terms of measuring marketshare. There’s possibly a larger difference between a low-end “smartphone” and a high-end smartphone, than between a mid-range featurephone and a low-end “smartphone.” I think we’ve reached the point where, when we must discuss marketshare, we should just use the whole cellphone market.

  • Canucker

    The consistency of profitability of the iPhone is staggering. At MWC this year, there is a slew of announcements of new Android models (not much on Windows Phone yet but Microsoft will presumably talk about that in a couple of days as they preview Window 8/Windows Phone 8). The consistency of Android OEMs from this announcement is bigger, faster, better. The screens are humongous and there’s LTE, higher resolution, more cores, etc, etc. They are competing with one another on specs and are each self-fragmented in their ranges. Instead of one new model, they each have multiple models (even the HTC “One” comes in three very different formats). Each of these has the latest components/technologies. The relative limitation in volume of each model (even within the expanding Android market) coupled to expensive components and direct head-to-head comparisons between models from other OEMs leads to the PC model of pushing specs – at the expense of margins. Samsung is perhaps the exception as its Galaxy S devices have hit high volumes (i.e. greater than 20 million units) and it is making the lions share of Android OEM profits.

    My mental vision is Apple basking in a comfortable pool of thick seafood chowder feeding at leisure next to a larger Android pool of sardine consommé stuffed with sharks who are flipping every other sardine  into the tiny pool next door to them into which Steve Ballmer is tightly squeezed (sitting in a Nokia inflatable chair). He doesn’t need to swim regardless of the chair since the food comes to him. He does have construction crews at work expanding his pool (they’re quietly dumping the landfill into the shallowing RIM pool).

  • What is “box contents” and “sales package”? Does that refer to the charging adapter and cable? Seems high compared to the rest e.g. battery

  • Anonymous

    The former editor of the Microprossor Report estimated that going by die size, and unknown elements on the die, Apple’s SoC’s could cost as much as $25, whereas other top chips may go for $15.

    In addition, iSupply and iFixit are not good sources for costs of parts, as they have no direct knowledge of what Apple pays. Apple’s costs in this area could be anywhere from -20% to +20% from their estimates.

    I don’t know why their numbers seem to be taken so seriously.

    • Anonymous

      Agreed, iSupply’s figures could be a source of inaccuracy, but, since (I presume) they use the same metrics to cost all devices, it serves as a useful baseline across all the makers. If it’s higher – it makes other manufacturers look inefficient, if it’s lower, then life is reeeally… tough down there for everyone else.

    • The flip side of that report was because Apple owned much of the integration IP of the A5, the actual cost to Apple was cheaper than a chip that went to (for example) HTC because Qualcomm got a good sized cut. As a result, Apple could do more powerful chips (as could Samsung) because they did not use COTS parts with additional markups built in.

    • If you have alternate sources of data (preferably with direct knowledge of what Apple pays) I’d be happy to cite them.

  • Pingback: iPhone sine qua non | asymco | iPhone And iPad News()

  • Rob Scott

    What is killing other OEMs among other things is cost erosion. Every quarter they have to lower their device cost prices. iPhone’s contribution margin (revenue less subsidy)/revenue is similar to other phones. So subsidy paid for iPhone is same or similar to other phone (this is mostly true, regardless of what you have read before) When not, the iPhone’s generates higher revenue and profits per phone (also true), thus same or similar contribution margin%. Put another way: when the iphone attract higher subsidy than other phones it also generate higher revenue and profits. In high end packages where all phones make similar revenue, iphone subsidies are the same as other phones. Subsidy on iPhones is usually recovered in 4 months or less. In short, the consistency in GM is to be expected.

  • sl149q

    The point that is missed when people criticize Apple for manufacturing in China is that only a small percentage of the profits ends up there. The vast majority of the iPhone (and iPad) profits accrue to Apple (and its mostly US shareholders) and in the case of the iPhone generates an equal or larger profit for the carriers (in whatever country it is sold in.) And finally an almost equal amount of dollars (to the manufacturing cost of salaries in China) goes to developers selling Apps. 

    A vast amount of dollars going every where except China… 

    • Anonymous

      Good points sl149q.

      Additionally Apple employs over 60,000 people, of which around 40,000 are in the US.

      Also they don’t take into account that Apple has sparked off “The App Economy” which according to a study by TechNet has generated 466,000 well paid US jobs.

      These jobs are well distributed around the country. Although following the usual “S” pattern for rapid growth, the number still grew 45% last year!

      To put the 466,000 in perspective, The App Economy already employs about half the number of people employed in Agriculture, but of course IT jobs are far better paid.It is probably not an exaggeration to say that Apple has made a significant contribution to reducing the recession and to pulling the US economy out of it.  

      TechNet, is a bipartisan political network of tech CEOs and Senior Executives, 

      This is a link to their report.

      • One thing I noticed is that we might be missing a point about employee headcount.

        Is that I think you cannot just say that Apple has 60,000 employees. Because more than half are retail workers.

        Which I don’t think you can use to compare to other technology companies, because most other companies don’t have a large retail operation that employees a large number of retail workers. Which I know are viewed differently from corporate employees. Even Apple separates them out on their own jobs board.

        The true numbers would be: Apple has 28,000 Full-Time Employees and 36,000 Retail Full-Time Employees.

        Then you can track Corporate workers more consistently:

        Non-Retail Full-Time (Derived from Apple Filings):
        2011: 28,000
        2010: 20,100
        2009: 17,800
        2008: 16,100
        2007: 13,700
        2006: 12,000
        2005: 11,127

        One other note is if you look at the growth chart I plotted, only within the last year has Apple significantly increased their rate of corporate hiring. From their previous growth curve.

        Just a thought,


      • Anonymous


        Thanks, good points and excellent data distinguishing Apple’s IT employees, from retail employees. Also a good chartHowever, to be fair to myself, I was responding to sl149q’s point about the economic benefit to the US from Apple’s profits and shareholder returns. I added the point that that Apple also created a lot of jobs, directly and indirectly in the US. I did not try to distinguish between IT and other jobs.Similarly the 466,000 jobs referred to TechNet report are not either all Apple jobs or all IT jobs.

        But there is no doubt that Apple is making a very significant contribution to the US economy, probably substantially exceeding the contribution it makes to the Chinese economy.

      • I totally understand,
        My point was less about your direct comment, but more about how some people are using the 60k as the number of people that work on the development and creation of Apple’s products which is simply not true.

      • Anonymous

        Apple is not an “IT” company

  • Anonymous

    Hmmm… I don’t know a lot of these figures are pure guesswork. The one thing that is known is that Apple has huge margins on its iPhone (not 50% though? I thought it’s more like 40? still crazy high…)

    We also know that Apple has made lots of strategic investments in its supply chain. $Bns invested that have an extremely healthy – though hard to measure – ROI. Where did these $Bns go and how much did Apple profit? We also know that Apple has THE master of supply chain management as COO, and now CEO. 

    I think two factors determine Apple’s crazy margins:
    – One, it’s the most desirable phone on the market. I don’t think subsidies have anything to do with it. It commands the highest price of any mobile – in part because if other companies make a phone more expensive than the iPhone, it just doesn’t sell. Here in Thailand about $700. There are no subsidies here, and Apple still had trouble keeping the phone in stock ever since it officially went on sales here – that’s close to 3 years running. 

    Think about it – if your main problem is making enough of the devices – which has been the case for Apple up until a month or two ago – then you won’t lower your prices. It would be counter-productive: You can’t make enough as it is, there is no need to get more customers.

    – Two – best supply chain – the advantages of Apple’s strategic investment and general supply chain tactics are hard to understand. What we can understand is the outcome: Amazing profits. 

    The iPhone 4S just became available in quantity here in SE Asia (first time of no wait), and Apple immediately lowered the price by almost $100. That proves they’re on the ball, and they know their margins are too high. 

    Conclusion: They will live with more production and lower margins from now on. I imagine operating margin will trend towards 30%, depending on their ability to saturate the market. They’re close now, having almost all countries, and most carriers… keep in mind that this was not the case before the iPhone 4S… 

  • anonymous

    “solid” and “opaque” are not the best descriptions for the regions in the graphic, because in reference to color they mean essentially the same thing.

  • anonymous

    (you could use the pair of terms “solid” and “translucent” or the pair “opaque” and “translucent”, but in the context of a white background, “light” and “dark” would probably be clearest)

  • Pingback: Onderzoek: ‘Apple verdient 319 dollar aan verkochte iPhone 4S’ -

  • Gordon Shephard

    Horace – I’m wondering why you include “Payments to Developers” and “iTunes overhead” as part of the iPhone cost structure – aren’t those separate business units that run at roughly a break-even (or small profit)?  I understand why iCloud is part of the iPhone cost structure, I suspect that the iTunes business should not be, if I understand what you are trying to accomplish here.

    • Those are “above the line” revenues meaning that they are ancillary sources of income for apple beyond the device price. When Apple reports iPhone income they also include accessories income (which includes licensing of the “made for iPhone” label). They don’t include app income but I thought it might be illustrative to show that as well. Note that the cost of goods sold don’t get subtracted out of the ancillary income.

      • Gordon Shephard

        So – I spent 5 minutes carefully reviewing the chart.  You are trying to look at the entire cost and value structure for the iPhone.  Where I’m having trouble understanding the chart, is that on the left side, you have the Average Selling Price (ASP) of $630, and on the right side, you have the Cost Structure + Margin.  I understand why the ASP = Cost Structure+Margin, by definition.  What I don’t understand is why the Ancilliary costs (iTunesOverhead+PaymentstoDevelopers) are incorporated in the cost structure, without, at the same time, capturing the value of that anciliary revenue in the ASP somehow – I.E. If we are representing the _cost_ of ancillary revenue, should we not also represent the _value_ of ancillary revenue?

        As a useful mind-experiment; if we were to have an increased revenue of $2000 in content (Apps, Music)  and a cost overhead (Payments+iTunes Ovheread) increase of $319, then, as the chart approach stands now, and entering the new ancillary cost, there would be a  margin ($630-630)/630 =   0%.  If, instead, we added the $2000 to the value column,   the margin would now be ($2630 – $630) / 2630 = 76%.  With real numbers it’s less obvious, but the same principal applies.

      • Gordon Shephard

        Upon further reflection – perhaps I was missing the point of including the ancillary costs of iTunes+Payments to developers – you aren’t necessarily trying to accurately reflect the value positioning (beyond what we already by showing by ASP) – rather, you’re trying to make it clear that, in the big picture (A) these costs are really minimal in the context of the actual margins – I.E. Apple could sustain these costs simply to promote the iPhone as a platform, even if it weren’t making a profit and (B) You want to further demonstrate how little the manufacturing costs are, in the context of costs such as iTunes Overhead/Payments to developers.

        It took a few hours to meditate over, but I think I understand, now, what it is you are trying to get across.  

        Great Presentation – thanks Horace!

      • Thanks. I strive to make it simpler than this but I had a dilemma with what to do about the “ecosystem” value “on top” of the hardware value.

      • The entire value structure would need to include the operator income and cost structure but I left this out because of space and resolution constraints. Also note that it’s Ancillary revenue (not costs) that is in the second stacked bar chart. The costs don’t begin until the third column.

      • Gordon Shephard

        How is “iTunes Overhead” and “Payments to Developers” considered revenue?  Sounds like costs to me.  And, why would you place revenue in the second column instead of the first column?  I’ve spent the better part of an hour looking at this diagram now – and I’m reasonably adept at understanding data visualizations and Balance/Income sheets…

      • Gordon Shephard

        Ahh.  I see what you’ve done, in the Revenue Column, you started stacking _above_ the $630 ASP.  So – I understand now how you have revenue in that column – and, from that perspective, “Revenue” may mean to Apple, or it might mean to vendors or data centers – Regardless, the numbers in the second column are not taking away from the Apple Margin – and are shown for comparison purposes of scale only.  

        It took me a couple days and about three hours, but I think I finally understand this chart.  Not the easiest one to comprehend, but once you finally grasp it, it’s a masterpiece of communicating a lot of things!

  • Anonymous

    “being an operator having  iPhone exclusivity generates above-average growth”..
    if you knew as little as using Google, you will find that this statement is DEMONSTRABLY FALSE.
    iPhone has been historically a loss causing entity for carriers. You can do a regression analysis on increasing iPhone sales vs profit for carriers and see the co-relation yourself.

    Is it a good phone or a bad phone? I dont know. But from a journalistic standpoint it’s your duty to report facts and not what you believe to be true.

    • As a reader, it’s your duty to read complete sentences.  If you had, you’d see that Horace is correct and that you’re wrong.

      “As seen in the small but clearly visible US market, being an operator having  iPhone exclusivity generates above-average growth and being the only operator to be excluded from the product generates above average subscriber losses.”

      The rest of the sentence makes it clear that Horace is referring to subscriber gains and losses. AT&T, Verizon, and now Sprint all gained valuable postpaid subscribers with iPhone; while Sprint, T-Mobile, and US Cellular lost those subscribers when they were/are without it. Both Sprint and T-Mobile management have emphasized that repeatedly in their statements.

      A company can increase profit by increasing revenue, or maintaining/cutting costs, or both. Those without iPhone can remain profitable (via reduced costs), but as we can see with Verizon and Sprint, there was only so far that they could go with that strategy, before choosing to let costs increase, in order to get the increased revenue (via increased subscribers) from iPhone.

    • jawbroken

      You seem to be confusing a temporary hit to margins from the upfront subsidy with the subscriber (particularly smartphone subscriber) growth that sentence clearly refers to.

    • davel

      The article ( and many like it ) that you quote from CNN is such BS.

      Just because something is printed in a reputable source does not make it true. I recently read a post someplace where they used rough estimates to show that even with subsidies a carrier pays off the cost of the iPhone in about 5 months.

      So yes, obviously there is an up front cost to acquiring an iPhone customer. Just as there is a cost to acquiring an Android or other customer. However there is a benefit to the carrier in keeping existing or new customers who spend on additional features ( like a data plan ) over a 2 year period. There is also a benefit to not going out of business. Look at TMobile who lost 700K customers that the company attributes to not having the iPhone. Or to the 5 year high in subscribers that Sprint reported immediately after acquiring the iPhone.

      So your statement “iPhone has been historically a loss causing entity for carriers.” is patently untrue.

    • orthorim

      Please. Don’t take the carrier whining seriously. That’s just hot air, and everyone knows it. If they really took a loss on the iPhone – guess what? They would not sell it! 

      Also – an iPhone bought at subsidy costs nearly 2x as much over 2 years than an iPhone bought at cost and used in a network with no subsidies. That’s money that goes straight into the carrier’s pockets. They’re pretty happy with the subsidy model. 
      On AT&T, the minimum amount you’ll pay over 2 years with the iPhone 4S is: $199 + $1319.76 ($54.99/month). $1520.Here in SEA I paid $650 for my iPhone + $480 for unlimited data over 2 years with tethering. = $1,140. This is $400 less than in the USA, with a way better data plan, and tethering. The closest comparable plan in the USA is the 2GB plan which will set you back $2240 ($89.99/month).The US carriers are raking it in with the iPhone.

  • Pingback: IphoneBreak.INFO » Sprint commits to buying almost 24 million iPhone units from Apple()

  • Pingback: Better Business Models Lead to Better Businesses « Innovation Leadership Network()

  • Pingback: IphoneBreak.INFO » Sprint commits to buying 24 million iPhones from Apple()

  • Pingback: Nieuwsoverzicht Appleweetjes » ‘Apple maakt 319 dollar winst per verkochte iPhone 4S’()

  • Lots of room for 3rd world iDevices at lower pricepoints… it’s easy to confront the law of large numbers when you simply open up markets that weren’t considered viable targets for your products before. At least “units sold”-wise.

  • Pingback: Nieuwsoverzicht Appleweetjes » Onderzoek: ‘Apple verdient 319 dollar aan verkochte iPhone 4S’()

  • Anonymous

    “Putting costs into perspective is as valuable as knowing what they are.”


    This is the most insightful comment I have heard in a long time. 

    Not related to the mobile device industry, but as an architect I would say that “putting costs into perspective is more valuable than knowing what they are”.

    Just think about those huge construction projects that ran many times “over budget”. Did they over-run because, for example, the budget included a price for reinforced concrete construction that was too low compared with the market rates?

    In most cases, almost certainly not. The cost over-run is the result of some factors not even considered when the budget was originally set. For example, the cost of an environmental impact assessment was not included in the budget for a housing development, and then the cost of extra work mitigating that impact, plus the legal costs and compensation suddenly add huge amounts.

    I think, in your field, “Putting costs into perspective” is what has set you apart from other commentators, and I think in your case it makes your commentary “more valuable than knowing what they are.”

    Keep up the good work …..

  • Morgan Michael

    Quick question….you put memory cost at $50…what memory capacity are you talking about here (16gb, 32gb). How will this change with different capacities

    • I used average cost for an average selling price.

  • kgelner

    iSupply numbers are complete fabrications.

    One way you can tell is simply by observing margins for competing devices.  They are built of roughly the same materials, or cheaper.  The manufacturing costs are probably a bit lower not having quite the same finish.

    Yet no other company has margins near those Apple has, even though devices from competitors are similar in cost.

    Something is thus very off in the material cost estimates.

  • kgelner

    iSupply numbers are complete fabrications.

    One way you can tell is simply by observing margins for competing devices.  They are built of roughly the same materials, or cheaper.  The manufacturing costs are probably a bit lower not having quite the same finish.

    Yet no other company has margins near those Apple has, even though devices from competitors are similar in cost.

    Something is thus very off in the material cost estimates.

    • A good friend of my is a sr. level exec at Samsung.  When the iPad 1 came out, Samsung put its engineers to work trying to build the exact same product, using essentially the exact same parts. Samsung’s costs were more than 50% of what Apple’s were.  There’s a reason you don’t have comparable products at Apple’s iPad prices.  Instead you have iPad competitors coming in at 200–400 above Apple’s retail price. 

      • kgelner

        If that is true then the article is still flawed, because it means Apple can have a margin $200 higher than any competitor at the same price point, which means Apple’s margins cannot be driven down.

        But how did SAMSUNG know Apple’s costs? All they have is guesses, in fact they too might be using the cooked iSupply numbers…

      • orthorim

        Yep, but the iPad has much lower margins than the iPhone. Tha’ts Apple’s strategy. The iPhone is the cash cow, the iPad is the long term play.

        Another reason iPhone margins have been so high – I think Apple called it unsustainable in one conference call – is that demand has been extraordinary. Apple was unable to deliver enough iPhone 4 devices throughout the entire lifespan of the product. The 4S is now finally catching up, and as a result Apple has lowered the price in South East Asia (official from the Apple store) by nearly $100. 

        Apple doesn’t want to be that much more expensive than Samsung’s high end models but as long as they can’t fill demand, lowering prices is impossible. It would just lead to even more demand, and even more customers that can’t get it, which would drive them elsewhere.

      • orthorim

        Also, I credit Tim Cook with the high margins – that’s Apple’s secret weapon. Tim Cook has taken supply chain management to a new level, and Apple’s size and cash has given them advantages that others just don’t have. 

        What other company is investing $Bns every year in their supply chain? And more importantly, where does this money come back to Apple, if we assume that these are wise investments with a healthy return? The ROI for these investments shows up in Apple’s margins!

  • Pingback: Horace Dediu on Apple’s iPhone Profit Margins | David Richard Greenwood()

  • Pingback: Apple, Samsung Top Smartphone Market Together, Begin to Settle Lawsuits « Mobile Industry News « VIP Cellulaire()

  • Pingback: Apple, Samsung Top Smartphone Market Together, Begin to Settle Lawsuits - News of 2012 | News of 2012()

  • Pingback: Calculating Google’s contribution to iPhone profitability | asymco()

  • Pingback: Lo stato dell'industria mobile al Q3 2012 | Hymn to Future()