Samsung’s profit center

Will there come a time when Samsung will earn more profits from the iPhone franchise than from its own Galaxy product line?

The problem for Samsung is that although it still sells the most phones[1], and the most smartphones, the price and margins for these products are collapsing. The pattern is shown in the graphs below:

Screen Shot 2015-07-30 at 7-30-8.20.29 PM

Screen Shot 2015-07-30 at 7-30-8.20.22 PM

Phone operating margins peaked in Q1 2014 at 20% but are half that level today. These margins have dropped to levels Samsung had in 2009, before the Galaxy launched and before they had any substantial revenues from smartphones.

In contrast, the semiconductor group is growing both revenues and margins. Margins and operating profits are both 50% higher than those of devices.

We also know that Apple is Samsung Semiconductor’s single biggest customer. We can’t be sure how much of the total revenue/profit comes from Apple but if the pattern continues then Apple could be the greatest contributor to Samsung’s profitability in the near future.

How could this be? Wasn’t Samsung supposed to “disrupt” Apple?

The reality is that Samsung’s own smartphones are being disrupted by good-enough Android devices, typically made by Chinese brands. This low-end disruption is also affecting LG, another phone maker and Apple supplier.

Unlike Samsung and LG, Apple is less susceptible to low-end disruption. What Apple offers is a brand promise, an ecosystem, associated products and services and what amounts to a new market. It’s this parallel value network that competes with Android/Google, rather than with Samsung.

Samsung is a supplier in the Apple network and a licensee in the Google network. As a result, it has a near monopoly position with a platform controller (Apple) and a commodity position with a platform licensor.

Neither position is one of control, but the current market stage benefits Samsung within the Apple ecosystem more than  in Google’s ecosystem. It’s a situation analogous to one where Intel made both x86 chips and PCs that ran Windows. Long term it’s likely that the sole-source chip business does a lot better than the multi-source systems business.

That should be a warning to others who hoped to survive as Android OEMs. They don’t have semiconductors to sell to Apple.


  1. over 80 million but we don’t have a precise figure []
  • Sony is in a similar position. Though their phone sales are tanking (undoubtedly due to Apple eating Sony’s home market), their profits on sensors, like iPhone cameras, are way up.

  • After Apple dropped Samsung for the A8X and split manufacturing with TSMC for the A8, Samsung went entirely in-house for the Galaxy S6 CPU instead of a mix with Qualcomm. That move likely helped bolster their chip profit. It’s also a reaction I anticipated on an earlier, related topic.

  • Aksam

    Hi Horace,

    Great article as always.

    you wrote “In contrast, the semiconductor group is growing both revenues and margins. Margins and operating profits are both 50% higher than those of devices.”

    But do we know what could cause the fact that the semiconductor division grew that much? Is this all attributable to Apple (or most of it)?


    • Ray

      It is not all or mainly attributable to Apple, but it is a factor. The main drivers are memory products, Samsung dominates both global RAM and Flash supply while other competitors are struggling, and premium smartphone processors, where Samsung has now almost 90% market share. They supply now to the two smartphone OEMs that dominate the premium segment, as they have the most advanced process in the world already at 14nm in mass production.
      The big loser here is Qualcomm which is having a rough year as
      they are being squeezed in the low-end by MediaTek and others and in the
      high-end by Samsung.

      • This does bring in question why Qualcomm is not doing well amongst a thriving components market.

        How much of it is due to Apple, Samsung and how much of it is due to the shrewd focus on core number by MediaTek’s marketing?

        The way I see it, Intel’s marketing is the prime example of how component supplier marketing (ingredient marketing) should be done, and MediaTek has been doing a superb job of this.

      • Ray

        Qualcomm has tried to follow Intel’s marketing playbook, they’ve been advertising their Snapdragon brand very heavily, but the smartphone industry has not turned out (at least yet) to be like the PC industry. The main players are highly integrated (Apple and Samsung), and these players, even if they’d use Qualcomm processors at some point, do not want Qualcomm to advertise “Qualcomm inside” (or “ARM inside”) as they have learned the lessons from the PC era where Intel captured most of the profits.

        The industry structure of the premium smartphone market is out of Qualcomm’s control, and it doesn’t play in its favor right now: highly consolidated, there are only two major players that take ~90% of the global market (Apple and Samsung Mobile). Both have decided to have significant control over their processors, Apple just on the design (Qualcomm’s expertise), and Samsung both on the design and manufacturing. So Qualcomm is now locked out of the two main accounts that are willing to pay for high-end smartphone processors. As other OEMs (Huawei, ZTE, etc.) move to the high-end, Qualcomm might be able to regain some volumes in the coming years.

        On the low-end, Mediatek, Spreadtrum and others offer lower cost solutions (as their engineering and sales costs are lower – China vs. California salaries), so Qualcomm just cannot compete with them, at least with most of its workforce in California.

        This leaves effectively only the mid-range as the market where Qualcomm can compete. And even this segment is being now effectively targeted by Mediatek and others.

        Samsung has another advantage over Qualcomm (besides being connected to Samsung Mobile): it has its own fabs (like Intel) so it can optimize its new processor designs for new leading-edge manufacturing processes, both technically and in terms of implementation and go-to-market dates.

  • David Leppik

    Samsung isn’t exactly in a safe position as Apple’s supplier. I could easily see Apple switching fabs, or building an in-house fab. Just as Android is a hedge for Google against Siri taking its search customers,* Samsung’s phones are a hedge against iOS gaining a monopoly position in phones.

    *Siri is much newer than Android, but Google’s fear of being locked out long predates actual, viable competition arriving.

    • Joe90

      What rubbish, Androids job was to ensure MS didn’t replicate its OS domination on smartphones and use that to attack Googles search business. The proof is in the documents the filed with the SEC. As for Apple gaining a monopoly in phones, that would require them to care about 85% of the market that provides 15% of the profits. Clearly they don’t. In this Android helps Apple by forcing everyone else to compete with each other on price, making it very hard for them to justify the investment needed to compete with Apple.

      • Juan

        Actually that 85% of the market provides about 8% of the profits and shrinking

  • Sacto_Joe

    Oh, the irony! Samsung joins Google as a company whose profitability is becoming increasingly dependent on Apple.

  • Exploited

    I find it amazing how Google managed to seduce so many OEMs into working for them by making them play the “everybody loses” hardware cutthroat arena with a “free” OS, when Wintel has already showed their true colors for over 2 decades…Guess sheep will remain sheep?

    • Ray

      Not “everybody loses”. Samsung has made billions in profits from Android, in fact more than Google itself so far.

      Not only on smartphone sales, as Samsung actually makes two margins on each flagship Galaxy phone it sells: the margin of the device maker (OEM), and the margin of the component maker, as they produce the key and most expensive components themselves (OLED displays, Flash, RAM, application processor, battery, etc.). This second margin is not included in the typical figures about smartphone market profits, but are significant additional profits for Samsung that Apple and other OEMs do not have as they have to buy their components from external suppliers (one of the major ones being Samsung coincidentally).
      It’s rumored that actually there is a third source of profits (according to WSJ), as Samsung might be taking a cut of advertising revenue from Google, the percentage not being known.

      Sure in the long run margins decrease as more new entrants find their market niches, but the long run is very long and one can make a fortune in the meantime…

  • jameskatt

    And Samsung has just cut its own profits by lowering the price of the Galaxy S6 to try and stimulate sales.

  • opaqueid

    I’ll buy the secondary point, that Samsung is being “disrupted” by the rest of the market. But the primary assertion seems a good example Betteridge’s law ( Questions in headlines can always be answered “No” )
    The core argument is that Samsungs Android profits are being commoditized away. Sure, that seems reasonable, but it’s highly unlikely that the either of the commodities that Samsung sells to Apple are likely to be **more** profitable then a commoditized phone line. Some related points:
    * It seems at least likely that Samsung would just drop it’s phone line if profits truly were that poor.
    * The assertion that “Apple is Samsung Semiconductor’s single biggest customer” although likely true, is misleading. At best Apple is a small chunk of Samsung’s total business, despite our natural tendency to equate “largest” with “majority”. In displays alone I’d be sorely surprised if the total Apple volume is greater then the total Android volume.
    * Being a supplier to a monopoly provider does not typically get you access the the monopoly provider’s profit margins. Apple can and presumably does drive a hard bargain, why should Apple give anything more then commodity profit margins to Samsung.

    • question

      There’s no question in the headline.

    • There are hundreds of phone makers with no profits. The profitless nature of the industry has not deterred those hundreds in the past from entering.

      • Ray

        It’s the lack of barriers to entry. Google gives you a free mature OS, Mediatek gives a full chipset solution. So if you have identified a niche in the smartphone market it is very easy today to start producing your own smartphones and enter the market. What a big company considers low profits could be a fortune for a small startup. The smartphone market is the largest consumer market in history, soon close to 2B devices sold per year, so it is relatively attractive for small players with an interesting niche.

    • it’s highly unlikely that the either of the commodities that Samsung sells to Apple are likely to be **more** profitable then a commoditized phone line.

      I’m not sure why you think this way. I think it is very wrong to assume that Samsung is selling “commodities” to Apple. On the contrary, it is more likely that Samsung is selling the world’s finest, most performant, and highest quality components. Why else would Apple buy from Samsung? Wouldn’t Apple insist on the highest quality components for their products?

      Apple buys camera components from Sony. Not because they are cheap, but because they are the best in the world and even Chinese manufacturers use them in large quantities. Apple buys semiconductors from Samsung because they are very good at manufacturing the highest quality semiconductors in the world. The list goes on and on.

      This Wall Street Journal article should provide perspective.,d.dGY

      • tmay

        Samsung might end up doing very well as a component supplier, but it would be fair to note that it probably wouldn’t drive it back to the peaks of revenue and profits in the heady days of its smartphone business.

        The larger question would be whether Apple will “stray” further into the “branded” component business with its HomeKit partners, possibly including a CarPlay initiative.

      • Is HomeKit a component businesses? I’ve never thought of it that way. It seems more like a branded communication standard to me.

      • tmay

        Broadcom, Marvell, and Texas Instruments sell the Apple certified communications chips, without which HomeKit compatible devices will not work.

      • Oh, so you mean it’s a components business for them. I thought you meant for Apple.

      • tmay

        No, Apple doesn’t build them, but Apple selected the three manufacturers, sets the spec, and certifies compliance. That’s a lot of control.

    • Ray

      It seems there are two underlying assumptions in your argument:
      Electronic components = commodities
      Commodities = low/no profits
      Both identities are wrong. Many electronic components are highly differentiated and profitable, so differentiated, unique and hard to produce that sometimes only 1-2 companies in the world are able to offer that product (e.g. Samsung is the ONLY company in the world able to mass produce power-efficient 14nm smartphone processors).
      Also, even commodities can be very profitable when demand exceeds supply and there are barriers to entry. In fact some of the most profitable companies in history sell commodities (e.g. oil).

  • Ray

    Samsung first made a fortune mining the newly discovered
    smartphone mine, and will next make a fortune selling mining tools to the dozens
    of existing and new smartphone miners.

    • I would wholeheartedly agree, if it were not for the fact that Japanese electronics (chip) manufacturers were obliterated by the Koreans.

      I have read that the Japanese failed to reorganise their business processes around the new technologies and trends in photolithography and manufacturing. This I think illustrates that disruption can come from many facets of a business, not only end product commoditisation but also from manufacturing, workflows, organisation structure, etc.

      • Vladimir

        Their electronics divisions were obliterated, but they are still in the game thanks to their other branches (insurance, construction, heavy industry, chemical industry and what not), which is very different from Nokia and Blackberry, who were on top with no net to catch them if they fall. So, Sony et al might not make any profit now, but they are still there hoping for a come back.
        Now, Samsung is not only doing good with Apple’s chips, but they also made dent in Intel’s and Qualcomm’s business (arguably). I don’t know how much potential is there but that could be maybe interesting to analyze. Before, Samsung would buy chips from Qualcomm and put in their flagships, now they make better chips in the house. And that means design, not only manufacturing. Same goes for the camera modules, memory, displays, and other crucial parts of smartphones.

      • The point that I was trying to make, and what I understood from reading about Japanese chip makers, was that business processes are typically optimised depending on the requirements of a product. Hence unless Japanese chip makers change their business process and workflows to adapt to the new requirements for modern chips, they will have a difficult time competing with the Koreans. Since these tend to become part of company culture, they can be very slow to change.

        So, Sony may have bought time, but that does not necessarily improve their chances of success relative to Nokia or Blackberry. Failure just becomes more prolonged and costly.

        Samsung may have trouble if for example, a new process emerges that eliminates the advantages of scale or quality in chip manufacturing.

      • Vladimir

        I agree with that, I just say we should expend what is considered by “Samsung” and “Sony” in context of the upper post even if we talk just about their chip manufacturing. I think that Sony’s chance for recovery comes from being able to identify, finance and ride the next wave if it ever comes, and that comes from less profitable but more stable sections of the whole company.

      • Ray

        I agree that Samsung could have trouble, just like any company in the world, if there is a disruptive change in technology or business processes. However the industry trends seem the move in the opposite way, the tech industry is increasingly reliant on Samsung components. Future trends and growing markets (upcoming smartphone market share wars, IoT, wearables, data centers, drones, etc.) play on their favor.

      • Ray

        No company is immune to disruption of course, there is simply a lower probability that Samsung will get disrupted vs. Apple or Google, because regardless of who wins consumer and platform wars, Samsung will keep having the cutting-edge components needed to satisfy the demand of the winners.
        Samsung has a strong and increasingly dominant grip in a moderately growing industry with increasing barriers to entry (semiconductors) – the huge amounts of capital and cutting-edge processes and equipment needed keep rising, and make it impossible for new entrants to get the right to play. Texas Instruments, Infineon, Freescale and many other companies simply haven’t been able to generate the sales and capital required to be able to stay on the race.
        One could think that China is the only country in the world that has the capital and incentives (growing consumer electronics industry) to change this and “do what Korea did to Japan”. However, due to geopolitics China is not allowed to import the equipment needed to produce high-end semiconductors. So China, even with enough capital, cannot do what Korea did: invest heavily on science & technology education of its population, on importing cutting-edge equipment and on building extremely expensive fabs. They can do it in the low end, but in the high-end China actually increasingly relies on Korean and Japanese components as their consumer electronics brands are trying to move to premium segments.

        On the demand side, the markets are demanding increasingly smaller and low-power components (as computing is increasingly mobile and wearable), which makes leading-edge process semiconductors even more valuable than they were on the mainframe and PC eras. Samsung and TSMC are uniquely positioned to satisfy this global demand for years to come.

        So the position that Samsung and its main US and Taiwanese competitors have in the semiconductor industry is relatively safe in the mid-term (relatively compared to Internet and software companies). If you look at Samsung revenues and profits from semiconductors, they’ve been increasing over the last two decades (except in the 2008 global economic crisis), and they keep doing so even if their Mobile division is not doing as well as it was three years ago.

      • I enjoyed all the details and information that you gave me, but I think that you might be underestimating the importance of new processes and technologies. You seem to be making your discussion based on an expectation that most innovation will be “sustaining” not “disrupting” (in Christensen terminology).

        It’s not only the Japanese giants that have been disrupted. Intel is also having a very hard time despite having at least an apparently unsurmountable advantage as Samsung is enjoying today. This all happened in the scope of 5 years or so.

        Disruption can happen to any company, and when it happens, it can be really fast. As far as I know, the strength of a company’s present position is no indicator of whether it will be disrupted.

        Christensen’s “law of conservation of attractive profits” suggests that the relative power of a specific stage in the value chain will change depending on the market situation. The current market conditions in smartphones favour component suppliers over assemblers, but there is no guarantee that this will continue to be the case in the future.
        Things change very quickly.

        And really, I don’t think arguments about which country is more educated based on some score has any place in these discussions. If you did that, most western countries would have a lot of trouble with the dismal scores that they get. I hate it more when you assume that we dismissed Korea because it was a former colony. Well USA occupied Japan even more recently and I’m sure they made fun of us too, but we never assumed that was why the Big Three found themselves in the predicament that they did. There’s surely more in business theory than that.

      • Ray

        We might forget but Intel or Samsung Electronics have been around for half a century. They’ve been thru many disruptions and new market creations in the high tech industry.

        Disruptions might be fun as an observer (so the more the better for us), but each industry has different dynamics, “company-bankrupting” disruptions happen more frequently in some industries than others. The mining industry for instance does not experience the kind and frequency of disruptions that the software industry does. The semiconductor industry is somewhere in the middle, definitely more stable than consumer electronics, software or Internet services industries, where every 5-10 years we see companies go from high-growth and/or leadership positions to near-bankruptcy or fire sale.

        In semiconductors Moore’s law has remained unchallenged (we’ve been hearing about the end of Moore’s law for over twenty years). And even Intel, which as you pointed out was disrupted by the chip design licensing + fabless business, is very far from disappearing (as opposed to Nokia or Blackberry). By the way, ARM processor licensing business model which disrupted Intel started not 5 years ago but over two decades ago, it’s been long in the making and was already highly successful in the early 2000s:
        This is the only main disruptive innovation at the business level that has happened in the semiconductor industry in 50+ years, and it hasn’t affected yet all markets (Intel is still dominating stagnating PCs and high-growth Data Centers).

        My point about education is that the barriers to entry in the semiconductor industry are extremely high (as opposed to the smartphone OEM industry for instance). Large amounts of human and financial capital are required, Korea was only able to do so because they invested for decades in their citizens’ education. Most Samsung executives have been educated in top universities around the world (e.g. Samsung’s CEO has a PhD from Stanford.). The rise of Korea has been more about constant long-term sustainable innovation, in Christensen’s framework.

        I meant that Japan dismissed Korea as a former colony, not the US (the US didn’t run these countries for long, but Japan occupied and ran Korea for several decades in the early 20th century). I meet many Japanese executives and even today they still dismiss Korea as a former Japanese colony and underdeveloped country. That attitude was widespread in the 1980s when Samsung was starting to be competitive with Japanese companies. In the meantime today Korea is beating them in many industries, and not only electronics: automotive, energy, music, TV, etc. In fact, the unthinkable (for many Japanese executives I meet at least) is just happening this year: Korean cars have just achieved the highest quality score in the world, surpassing Japanese, European and American brands:

        I know this is a blog focused on disruptive innovation but even the current success of Apple is more based on sustaining innovation. They created the iPhone almost a decade ago, which was a disruptive innovation then, and they have been since then perfecting it with sustaining innovation (improving camera, size, materials, key apps like maps and music, etc.) to protect and expand their market share.

        What really favors Samsung is that the world has been increasingly demanding more and better electronics and looking at future trends this seems to continue being the case. IoT will essentially require expanding the semiconductor market into “things” that traditionally did not have any electronics. This is a trend that bodes very well for Samsung, beyond the current smartphone wars that also as you mention are shifting profits towards cutting-edge electronic component providers like them.

        Yes, it’s not impossible that Samsung Electronics gets disrupted, but its diversification and particularly the industry they depend on (semiconductors and electronics applications) has been relatively stable and keeps expanding and more importantly increasing its barriers to entry. McKinsey estimates that due to this huge capital and technology requirements only three companies in the world are going to be able to produce leading-edge semiconductors in the coming years: Samsung, Intel and TSMC.

        The barriers to entry in the smartphone industry, to the contrary, keep getting lower and lower and that’s why you have now thousands of smartphone OEMs (there are over 1,000 Android licensees) and tiny startups in China and India are able to enter the market from nowhere and challenge major global players like Apple or Samsung Mobile division.

      • berult

        Any enlightened perusal of Apple, Apple as a phenomenon that is, must do away with the highly reductionist concept of iPhone coming onto the world scene as a ‘hit product’. It’s as much a hit product as ‘The Renaissance’, and as the ‘Industrial Revolution’ might also have been conceptualized in the history books as ‘hit products’.

        iPhone is the look one casts upon oneself, when one abstracts an obfuscating paradigm from what is the essence of oneself. iPhone is, all at once, pure Renaissance, and through concomitant millennium exegesis, pure Industrial Revolution.

        The Mac, the iPod, the iPad, …, emerge as epiphenomena. Hit products, if you insist. Borne out, and nourished through their umbilical cords, by either a counterclockwise Renaissance in-the-making, or a full-fledged, fast-forward one. We’re talking fluid-dynamics on a grand meta-scale here.

        Time-lapsed, and overtaken comic strip-like, I, you, and the Samsungs of this world, …aren’t we all…!? berult.

      • Ray

        I’m in awe.
        I had heard many superlatives about the iPhone but never putting it at the level of The Renaissance, or the Industrial Revolution. I guess soon someone out there might compare it to the Big Bang…

      • I appreciate your knowledge but the dismissal of the Koreans by the Japanese is exactly what is described in Christensen’s framework. It is the common response to disruption by the incumbents. You do not have to take in account colonisation or occupation, regardless of whether that was for a few years or for a few decades. Christensen’s theory is totally sufficient to explain the reactions of the Japanese executives that you met.

        Korea may have some advantages over Japan based on history or attitude, but I’m sure that you remember, not so long ago, the Japanese were also being praised for hard work, eagerness to study, etc. Furthermore, I’m pretty sure that the impact that Japan had on the U.S. In the 1980s was far more significant than what Korea has now.

        Just to clarify, I’m not making a comparison between Japan and Korea. I’m just saying that we do not have to include these arguments about which country is better at what to understand business dynamics. Christensen’s theories are sufficient. Furthermore, praise of a countries culture, education, etc. tends to flip flop. The very same Japanese attributes that outsiders admired during the 1980s, like stability of employment, suddenly became the scapegoats in the 2000s. The correlation between various cultural aspects and economic growth are hardly understood in concrete statistical terms, and hence commentary on these issues tends not to be no better than simple opinions. That’s ok if you don’t have any other theory to base the discussion on, but when you have Christensen’s framework, I question your decision to ignore it.

      • Ray

        Sure, low-end disruption that is initially dismissed. You could apply that framework. But frameworks are a very high-level / low-resolution ways to describe reality, they tell you a high-level idea but they don’t tell you how things happened. There are plenty of low-end Korean companies that did NOT disrupt Sony and other Japanese companies. So there is a much more interesting story about why or how that disruption happened. Why was it Samsung and not others? What’s especial about them?

        There are some strategic organizational choices where Samsung excelled over Sony. A key choice is that Samsung didn’t make the mistake of keeping its semiconductor division tied to the consumer division (as Sony did). This forced the semiconductor division to actually have to look for customers, grow, invest and improve to be competitive and world-class (instead of just a division with an assured in-house customer). Nowadays Samsung Electronics semiconductor division is the #2 semiconductor company in the world (in revenue), which allows them to continue investing in extremely expensive fabs, and thanks to that has caught up to Intel in the leading edge. To the point that now Apple and other premium OEMs rely on Samsung. The only alternative is TSMC, and they have just been surpassed by Samsung (Intel does not provide foundry services for chip designs). Essentially the leading-edge semiconductor industry is now a duopoly Samsung-TSMC, with Samsung now having the most advanced technology. Hence the great margins that Samsung now enjoys and will enjoy for the next few years.

        Also, disruption goes both ways. Vizio, an American company, is now disrupting from the low-end Samsung and Sony TVs which have a stronghold in the premium segment.

      • Yes. This is much more interesting.

        Current market conditions in smartphone SoCs does support modularity in semiconductor design and manufacturing, as has been discussed here before.

        The question always is, when will technology overshoot the market, and that is what may predict the likelihood of disruption. Of course, keeping in mind that market demands change rapidly.