Predicting The Second Quarter

The last quarter of 2016 was Apple’s biggest ever. $78.4 billion in revenues. 78.3 million iPhones. Both records. Earnings of $3.36[1] and cash reached new highs. The growth was modest but Services is now not just the second largest revenue but also the fastest growing, on track to doubling in four years.

The reason for this is the vastness of Apple’s user base coupled with the loyalty the brand engenders. The company reached one billion active devices more than a year ago and is quite likely to have nearly a billion users. Not just any billion either–the best billion most probably.

The realization that Apple benefits from extreme quantity and quality of customers has led some observers to defer the imminence of Apple’s demise. Share prices have risen lately to new highs and the ratios between these prices and earnings have started to come closer to the average of other large companies.

But enough with the reductio ad absurdum. Let’s look at the next quarter. The company has been very precise with its own offered predictions (guidance) so it’s a simple task to make an accurate forecast. The history of guidance vs. outcomes for revenues is shown below:

Placing the pipper on the upper edge of the guidance range was a safe targeting method for 11 of the last 17 quarters. We have never seen a drop below the bottom of the range but had a few overshoots, some huge. Doing this for the latest quarter and working backwards to the individual product contributions to revenue and using historical margin and cost patterns we can get the following core financial performance metrics:

Fiscal Q2 2017:
Rev ($B) 53.4
EPS ($/share) 1.96
iPhone (thousand units) 53100
iPad (thousand units) 9900
Mac (thousand units) 4100
Watch (thousand units) 2300
Services ($ million) 7129
Other products ($ million) 2300
Gross margin (%) 37.8

I sent these estimates to Philip-Elmer DeWitt and I recommend looking at his survey of the other analyst estimates on his excellent Apple 3.0 blog.

Overall, the quarter is shaping up to offer slight growth y/y and indicating a plateau formed after the surge from the iPhone 6 surge in ’15. All eyes are on the “super cycle” for the next iPhone.

The reason it’s “super” is that the iPhone has a 2+ year cadence for form factor changes and the user base updates on a similar cycle. The demand is vast due to the user base and accumulated age of devices in use. Expectations are that supply will be offered to meet this demand.

If history is a guide then the next iPhone will be the best iPhone ever.


  1. On shares priced at the time around $105 or 3.2% yield quarterly earnings []
  • obarthelemy

    Again, the main reason for last quarter being Apple’s biggest, is, literally, that it was Apple’s biggest ever, with an extra week. If you take that extra week out, it was a down quarter (+13% duration, +8% sales).
    I find it weird that this never gets mentioned… on some sites.

    • Sacto_Joe

      What you’ve left out is the imbalance between production and demand, especially for the exceedingly popular 7+. Had production capacity reached demand, the sell-through would have been much higher. This demand was almost certainly pushed into the following quarter, which also saw continued demand for the new AirPod that far exceeded supply.

      The bottom line is that neither we nor Apple truly could predict demand for Q2 fy ’17. But we have a history, as Horace points out, of Apple being able to estimate the MINIMUM demand quite well.

      That is not to minimize the importance of that extra week. For non-constrained products, it did help. But for the actual results of last quarter, we will, as usual, have to wait and see.

      • obarthelemy

        Indeed, the imbalance was left out of that quarter, as it was left out of all previous quarters. I’d be glad to take it into account, if the info is available.

        The other naysayer above kind of implies that sales go down later in the quarter, which doesn’t mesh with the product being constrained earlier in the quarter. Which is it ?ANy facts as hard as “hey, we made our quarter 13% lnonger , and we have sales 8% bigger, you guys think you can sell that as a record quarter ? – Suuuuure !”

        Edit: Apple has liability, so they took great care to publicize and clarify the longer quarter in their earning call and info. Apparently, “analysts” have more leeway in doing away with inconvenient facts…

      • klahanas

        Analysts, bookies, and other lower level primates.

      • mjoecups

        Don’t forget Pundits. I am one, I should know.

      • klahanas

        I’m just a lower level primate. But I’m terrestrial. For some real shameless BS, you should check out extraterrestrialis pithicalis…

      • rattyuk

        Well, one assumes they were buying into the run up to Christmas… not sure what the extra week added. You can’t just take the number of weeks and divide them down. Not all weeks in the run up and after Christmas are the same.

        Looks great from the haters, when you bring facts to the table, not so much.

      • klahanas

        What you say is reasonable. Could you be neglecting using monies from returned ties, sweaters, and thingamajigs boosting sales the week after? Many people buy more the week after.

      • rattyuk

        I could be wrong but I’m not entirely certain the extra week was the one you mentioned.

      • klahanas

        Yeah, me either. I really should shut up about these matters…

      • Space Gorilla

        The extra week was at the beginning of Q1, since Apple ended the previous financial year on Sep 24, so Q1 started Sep 25 and that was where the extra week was.

        Crawford posted a good comment about this.

        I would think it susses out something like this:

        01 – Sep 25 to Oct 1
        02 – Oct 2 to Oct 8
        03 – Oct 9 to Oct 15
        04 – Oct 16 to Oct 22
        05 – Oct 23 to Oct 29
        06 – Oct 30 to Nov 5
        07 – Nov 6 to Nov 12
        08 – Nov 13 to Nov 19
        09 – Nov 20 to Nov 26
        10 – Nov 27 to Dec 3
        11 – Dec 4 to Dec 10
        12 – Dec 11 to Dec 17
        13 – Dec 18 to Dec 24
        14 – Dec 25 to Dec 31

        So unless we all think Sep 25 to Oct 1 was a HUGE week for iPhone sales, we can all relax about the extra week 🙂

      • Space Gorilla

        There’s some analysis that puts the extra week at about 1/14th of sales. But worrying about the extra week is missing the point. Apple has a billion customers now (or close to it), a significant Service business which is growing as well as high customer satisfaction, and the user base continues to grow. Apple has challenges and things it needs to improve, certainly, but Apple is doing just fine overall.

      • obarthelemy

        Well, if you’re comparing quarters, and one is 13% longer than the other one, I’d think a neutral analyst would at least mention it in passing; Apple took great care to mention it in their earning call, which is probably a sign that it is important.

        If you want to get fancy, you can try and justify that the week breakdown shouldn’t be proportional, and estimate objective impact of a trailing week. But if you’re unwilling to do that work, blindly crediting an extra week as a blockbuster quarter just doesn’t cut it. It’s PR, not analysis.

        My position is less flawed than the article’s. Yet I’m the one getting heat. Time to examine bias ?

      • Sacto_Joe

        “Indeed, the imbalance was left out of that quarter, as it was left out of all previous quarters.”

        That’s not a true statement. There was a bit of imbalance between production and demand in Q4 fy ’16, since the iPhone 7 series only sold a few days in that quarter. But there was no imbalance in the other quarters since the iPhone 7 wasn’t sold in them.

        “The other naysayer above kind of implies that sales go down later in the quarter, which doesn’t mesh with the product being constrained earlier in the quarter. Which is it?”

        First, as regards “the other naysayer”, that implies that I was “naysaying”. I wasn’t. I agreed that there was an impact – on non-constrained products.

        Second, asking me “Which is it?” implies that both answers can’t be right, when they can. Obviously, demand in the U.S. will slacken after Christmas is over. But it doesn’t disappear. For one thing, Chinese New Year was coming up. For another, it’s a simple fact that supplies for the 7+ and the EarPods were constrained all through the quarter, including that 13th week. In fact, last I heard, EarPods are STILL constrained, and here we are in April!

        Third, you’ve completely ignored my valid point, which is that, had the products not been constrained, then a LOT more money would have been made by Apple in that quarter. Yes, it could be expected that an extra week would also increase revenue, but not beyond the capacity for Apple to produce. As it was, Apple sold over 78 million iPhones, which is 3.5 million or 4.5% more than they’ve ever sold before in a single quarter.

      • obarthelemy

        The revenue chart isn’t about iPhone 7, but about revenues: it doesn’t care which iPhone is current.

        And I see the “constrained” issue as mainly separate. It is indeed a concern, and a recurring one, but doesn’t work as an excuse for a longer quarter. As someone else mentioned on techpinions, it’s also hard to evaluate impact especially when doing comparisons, previous launches might have been earlier or later, more or less constrained, or with a different profile, … Again, taking that out of the hat to justify not mentioning the extra week is fake: earlier launch quarters were impaired too, there has been no effort to assess by how much, and it is an ongoing issue endogenous to Apple so why even mention it as an excuse ? And the constriction makes the extra week bigger because catch up.

      • Sacto_Joe

        (Sorry for the time it takes to respond – lots of demands on my time these days.)

        First off, if we’re talking about revenue, which we are, then it’s very relevant to talk about the iPhone 7+, since the iPhones are such a big chunk of Apple’s revenues.

        Second, I agree that the extra week should have been mentioned – and not just as a way of making sense of the stellar sales. It’s just as important, or maybe even more so, for next fiscal year. Thus, in the December quarter of 2011 (Q1 fy ’12), we also had an extra week. In October of the following year, Apple guidance for Q1 fy 2013 was for revenue growth of 12.23% over the prior-year period but a decline in yoy eps for the quarter of 15.28%. In contrast, analysts were expecting a decline in eps of 4.10%. Management tried to remind analysts that the year previous quarter had gained from that extra week, and so year over year comparisons were out of whack, but to no avail. The stock began one it’s worst retractions of the last decade, dropping about 45% in value by the middle of 2013.

        Of course, there was more going on. Year over year comparisons have always been a bit unfair because of the difference between a major upgrade year and an “s” or minor upgrade year. So comparisons are more meaningfully taken between the 2 year previous quarters.

        Still, it’s nice to see you acknowledge that there was a significant impact from that extra week. I expect that you won’t be grousing about unfairness when, after Apple’s guidance in October for the December quarter, we note the negative impact on the year over year comparison due to the LACK of that extra week!

    • nonuniform

      Sales aren’t uniform across a quarter, especially one that includes Christmas. I find it weird that this never gets mentioned…by some posters.

      • obarthelemy

        Please do elaborate. How do sells vary, and how did that impact this quarter vs last year’s ?

      • nonuniform

        Who knows how they vary, companies generally only provide financial results at the level of a quarter, which is of course the integral across the quarter. So any model is going to be essentially untestable. Perhaps you could extract some information from when the launch timing of the iPhone has changed, but I’m not inclined to do your homework.

      • obarthelemy

        It’s your homework actually, me I’m just saying that comparing an 8-week quarter to a 9-week quarter is flawed.
        edit sorry 12 and 13.

      • nonuniform

        No, that is explicitly not what you are saying. To quote
        “If you take that extra week out, it was a down quarter”
        This is obviously not supportable with any model other than the obviously incorrect one (that sales are uniformly distributed throughout a quarter).

      • nonuniform

        To be more clear, you’ve supported it with the obviously incorrect model by saying “(+13% duration, +8% sales)”. I misspoke when I said it is not supportable by any other model, which is obviously incorrect.

      • obarthelemy

        This is obviously supportable by many models, as long as the sales in the extra week are not down by a very large amount. Which is your contention, so it’s yours to prove.
        The sales can be down a fair bit in the last week, and still explain why the quarter was “record” (I use the term loosely, which seems to be the standard here).

        Edit wanna bet that one of the first thing that will be said about next year’s quarter will be “and don’t forget the performance seems worse because we’re comparing the last year’s abnormally long quarter” ? To me, that’ll be the litmus test of intellectual honesty. Anyone doing that, but not this, will be not only dumb, but dishonest.

      • nonuniform

        I already addressed that below, so I don’t need to again. Obviously this is not “my contention to prove” since I didn’t make the assertion about it being a down quarter, you did. Please follow your own side of the conversation, keeping track of the things you assert.

      • obarthelemy

        Congrats on Apple for a record quarter ! I suggest they keep adding a week to each quarter going forward, that shoull keep the trend going.And lower reporting costs, after a while.

    • Space Gorilla

      I think it might be constructive to point out that you’re missing the point of the post, which I think is summed up nicely here:

      “The realization that Apple benefits from extreme quantity and quality of customers has led some observers to defer the imminence of Apple’s demise.”

      Whether Apple’s last quarter was up, flat, or down doesn’t change this. Nobody is trying to hide anything about the extra week. It was reported on long before Apple released that quarter’s results, it was noted by Apple when they gave the results, and it has been reported on since. It seems like you’re just mad that people are calling it a record quarter.

  • Space Gorilla

    I see that the Resident Asymco Troll is back to remind us: No matter how much Apple succeeds, Apple is not succeeding. ™

    • rattyuk

      it appears that certain posters love to knee-jerk react to any actual facts with nonsense, dividing the total number of weeks by the income and then claiming that every week had the same sales, ignoring the fact that the sales into Christmas were probably significantly higher than those after.

      • Space Gorilla

        Yep, hopefully the trolls won’t pollute Asymco too much.

  • Walt French

    Having spent the last couple of hours speculating w an Apple employee about the evolution of understanding,* I’m intrigued that the format of these quasi-forecasts/explanations are very well-suited to sell-side research providers’ expected output, but much less insightful into what makes Horace’s work so useful and unique.

    No company exists without shareholders, who need this type of info, of course, but as you say, others do fine jobs of collecting & comparing the estimates and histories, across analyst, time and company.

    There’s lots of useful context here, to be sure. But Some spitballing by @ElkMovie this morning gets closer to the question of disruption, how businesses cope with forces bigger than themselves:
    “There’s no platform I’m aware of that’s in a potential-dripping nascent stage like that now; nothing 5 years away from conquering the world.”
    “For any expensive / rare consumer tech product, cut its price by 75% and improve its performance by 400%: would everyone then own one?”

    Those led us into a little chat which touches on what the value stack of automobiles will be (the only tech I could think of that warranted a positive answer to his question). Actually, I’d recently wondered whether @AsymCar wouldn’t be the BEST person to opine about the likely radical shift in auto economics, with the engine/wheels/seats/etc becoming as relatively important as the bottle is to a Coke.

    Changes to the iPhone, even possibly “radical” ones involving AR, can still sneak up on us and create an explosive new platform shift. Those should be (1) a specialty of this site, and (2) even easier to look at, given the likely stability of the other 95% of the iPhone platform.

    So, not a rant but a plea: these are exciting, useful stories to tell; there is information to be teased out & organized, and where else can I get informed opinion about transformative changes better than here?

    * did you know it was 256 years from Copernicus’s theory until the empirical verification of Newton’s Universal Law of Gravity that cemented the mechanisms?

    • macro pundit

      >> Actually, I’d recently wondered whether @AsymCar wouldn’t be the BEST person to opine about the likely radical shift in auto economics, with the engine/wheels/seats/etc becoming as relatively important as the bottle is to a Coke.

      Engines are already ignored by consumers. But wheels? Nah, for new cars people want big wheels, and big lights (flashy LED taillights and daytime running lights). Car companies charge extra for models that include these features, and people gladly pay to have it.

  • Crawford

    To the best of my recollection, amongst Apple’s many gifts, they did not give us December 32nd through December 38th.
    Week 13 of 1Q17 was, as always, 25 Dec- 31 Dec, 2016.
    Week 12 of 1Q17 was, again, 18 Dec – 24 Dec.

    Week 14 of 1Q17 was the group of days that forced the issue in the first place, 25 Sept – 1 Oct 2016.

    That particular week was not in any way associated with pre-Xmas, post-Xmas, year-end tax purchases, or any other major purchasing holiday.
    And so was most likely amongst the calmest, most low-volume, low-revenue weeks of the quarter.
    Please consider titrating any criticism of the achievement of 1Q17 with a true conception of that sleepy week; and thereby be kinder to each other here in this asymco space.

    • Space Gorilla

      Well said, and you’re right. From Apple’s filing info: “the fiscal year ended September 24, 2016 and its Form 10-Q for the fiscal quarter ended December 31, 2016 to be filed with the SEC.”

      Which means Q1 started on Sep 25, as you point out.

  • hannahjs

    If history is a guide, locusts will descend, jackals and hyenas will creep forward, and pundits will locate the flaw that dooms Apple.