Apple: Exploding iPhone Profits

iPhones in Europe are not the only thing exploding at Apple (AAPL). Recently, Bernstein Research and RBC Capital Markets predicted 2011 iPhone sales of 50 million and 54.7 million respectively with RBC also predicting 2012 iPhone sales at 82.1 million, about a 3-fold increase from 2009 levels. The above estimates are calendar year estimates, ending in December, while Apple’s fiscal year ends in September. The below discussion is referenced to Apple fiscal calendar, so Q4 2009 means the quarter ending in September, 2009.

While Bernstein provided an Apple price target of $185, RBC’s target was $250. This left me wondering why their price targets are so far off from each other when their unit sales estimates were fairly close? There could be many reasons including differing assumptions for P/E (or discount rate), revenue per iPhone or gross margin, GAAP vs. non-GAAP earnings. The financial model and stock price targets presented below are based on the above projected unit sales.

In order to produce an iPhone centric Apple profit forecast with some degree of reasonableness and technical correctness, you need to model the following: 1) GAAP subscription accounting, 2) iPhone unit sales price, 3) iPhone gross margin, 4) iPhone units sold and 5) a simple model for the rest of Apple’s business.

Under GAAP, iPhone revenues use subscription accounting and are amortized over 2 years, so there are two ways of looking at Apple’s iPhone earnings, GAAP or non-GAAP. GAAP subscription accounting smoothes out EPS because it is based on a running average of past revenues. The non-GAAP EPS fully accounts for all sales in the quarter they actually occur, but are very choppy due to seasonal sales variation and iPhone refresh patterns. As required Apple reports GAAP and voluntarily provides a limited amount of non-GAAP financial information. If you own Apple or are thinking about buying, it is a good idea to understand what both numbers mean and why they are different. The below graph illustrates the iPhone effect on Apple’s pro forma quarterly earnings. The non-GAAP EPS spikes are from the annual iPhone refresh in June which super charges Apples Q4 sales each year. The GAAP EPS is much smoother since it is largely made up of an average of past iPhone sales. Apple recognizes sales, general and administrative (SG&A) costs as they are incurred, so when iPhone sales spike, SG&A costs also spike, putting a dent in the GAAP earnings.

Apple does not directly provide current quarter iPhone sales or profit margin. When I say “sales”, I mean actual sales for the current quarter. What they do provide is the iPhone GAAP revenues consisting of 1/8th of the actual sales from the current quarter plus 1/8th of the actual sales from each of the previous 7 quarters. Apple also provides numbers to reconcile GAAP to non-GAAP iPhone revenues, cost of goods sold and net income. See the bottom of Apple’s Q3 2009 earning press release. Be sure to read the footnotes.

If you add the GAAP iPhone revenues and the GAAP to non-GAAP revenue reconciliation number, you get the current quarter’s actual iPhone sales. Apple also provides the number of iPhones sold in the current quarter, so you can calculate actual revenue per iPhone. From the Q3 2009 financial statement, GAAP iPhone revenues were $1,698M. Add to the above adjustment of $1,405M and you get total non-GAAP iPhone revenue of $3,094. Apple also disclosed unit sales of 5.208M. Actual Sales Prices (ASP) = $3,094/5.208 = $594 per iPhone. Applying this to the prior three quarters, starting with Q4 2008 gives ASPs of $666, $660, and $636. iPhone ASP is on a downward trend.

It would be nice if gross margin could be calculated directly, but this is not the case. You can not use the GAAP to non-GAAP reconciliation revenue and net income numbers above to calculate iPhone gross margin for the current quarter (or any quarter). The reconciliation numbers are marginal numbers, thus the percentages in the above table are not useful. For example: If a widget is sells for $600 at a profit of $200, the profit margin is 33.3%. The price is then lowered to $500 and 3 are sold for a total profit of $300 on $1500 of revenues, a 20% margin. Revenues increased $900 for an extra $100 profit. It would be incorrect to say that the profit margin on the latest batch of sales was 11.1% ($100/$900).

If you did use the marginal GAAP to non-GAAP numbers, you would incorrectly be using a gross margin of 71.5%, a tax rate of 20.8% (very generous) and a profit marine of 50.7%. Various tear-down analysts (Google it) estimate 60% gross margins on the iPhone. So, where do you go to find a profit margin? Turley Muller, The Financial Alchemist, has done detailed bottoms-up price analysis on iPhone gross margins and comes up with 58.4%. Another good independent Apple analyst is Andy Zaky. I did my own accounting based analysis and got 57.3% for the four quarters ending in Q3 2009. I used the deferred CSG from Apples quarterly statements and used the solver function in MS Excel (on my MacBook). to figure out the actual non-GAAP CGS going back to Q3 2007 by re-creating a deferred CGS table. I already figured out how to calculate non-GAAP revenues for the last 4 quarters. Subtract the non-GAAP CGS from the non-GAAP revenues and you have gross margin for the iPhone. Using the above method, iPhone Gross Margin for the trailing 4 quarters was 48.4%, 55.3%, 59.7% and 56.8%. I’m close to Turley Muller and other tear-down reviews, so I’m satisfied that I’m in the ball park with 57.3%.

The below table of projected iPhone unit sales estimates is loosely based on Bernstein Research and RBC Capital Markets projections while trying to be consistent with prior years patterns of about 50% of iPhone sales occur in Q4 and 20% in Q1. There is also a big jump from Q4 2009 to 2010 to account for the expiration of the exclusive AT&T (T) deal.

With all of the above input, the below table provides pro forma GAAP and non-GAAP sales and earnings estimates with target stock prices for 2009 through 2012. The following assumptions were made:

• Conservatively estimated non-iPhone revenues grow at 10% per year and have a gross margin of 30%. Actual 3-year growth rate for this segment is 16% and profit margins have been historically higher but have been trending down from the 34-35% level in 2008 to 31.5% in the most recent quarter, so I feel this is conservative, but reasonable. My analysis of the non-iPhone part of Apple was purposely simple because I’m trying to isolate the effect of iPhone revenues and gross margins on the bottom line.
• iPhone unit revenues declining from $600 to $500, causing gross margins to decline from the current 57% to 48% by Q4 2012. I assumed that international sales with lower ASP and increasing competition will erode margins. Again a conservative approach.
• Apple’s reported deferred revenues lumps Apple TV in with the iPhone. I am going to ignore Apple TV and assume the deferred revenues are solely due to the iPhone for the following reasons: 1) Apple does not provide Apple TV unit sales numbers, 2) An informal survey of articles on Apple TV leads me to believe profit margins are thin to none and 3) iPhone revenues and earnings completely overshadow any Apple TV revenues and earnings.
• No assumptions are made about the revenue and income effects of future Apple products such as an “iTablet”, iPod refresh, Apple TV subscriptions, etc.

A few of points worth noting:

• By the end of 2012, iPhone non-GAAP revenues are almost half of Apple’s Sales and comprise 63% of total income. This is the explosion.
• There may be quarters where the GAAP revenues are higher than the non-GAAP revenues, creating a negative EPS adjustment. This can happen as a big non-GAAP revenue quarter creates a stream of large deferred revenues and is followed by one or two quarters of moderate non-GAAP revenues. This effect can be seen in the quarterly pro forma graph presented earlier. This has not happened yet, but I think it eventually will. I also think there will be many who misinterpret the event.
• The GAAP based target stock price for 2010 is fairly close to Bernstein’s target of $185. Could be a complete coincidence. RBC’s financial model and assumptions for a $250 target price are necessarily more optimistic. They may have assumed a higer P/E and/or higher margins. Maybe they accounted for the “iTablet” or used a DCF model and estimated accelerating revenues out past 2012. The difficult part of evaluating an analyst’s price target is that they normally don’t come with all the assumptions, although I’m sure if you paid enough money, you could get that data too.
• The deferred iPhone revenue of $40,381M in 2012 equates to $15.70 in deferred EPS. This was calculated by constructing a deferred revenue table and applying the profit margins from the quarter of actual sale to create a deferred gross margin table. Since SG&A has already been expensed, you can just apply a 30% tax rate to the gross margin to determine deferred EPS.
• Total GAAP gross margins are backward looking. I like to call them “apparent” gross margins. They are based on a weighted average of past iPhone revenues and CGS plus the current non-iPhone revenues and CGS. The non-GAAP gross margin is the real number, and I estimate that it will be near 41% for 2009. In my model, I assumed decreasing iPhone ASP and gross margins, which is the reason for the lowering non-GAAP gross margins in 2011 and 2012.

A target price between $345 and $399 by 2012 may seem high, but RBC’s predictions of 82.1 million units shipped in 2012 will give Apple a smart phone penetration of 17% and a total mobile phone penetration of a mere 2.6%. Their 2009 projections are 13.5% smart phone and 0.8% mobile phone penetration. In this context, these numbers seem reasonable and achievable. The battle between perceived value based on GAAP and value based on non-GAAP EPS will continue. This is really no different than other companies whose EPS do not properly represent their value and analysts turn to DCF or other valuation models.

One item missing from the valuation calculation is the effect of cash. Apple is a cash-generating machine. As of Q3 2009 they had $31B in cash and marketable securities, 20% of their market capitalization. To see how Apple’s exploding iPhone revenues are affecting cash, in a future article, I will expand this discussion to include the balance sheet and statement of cash flows.

I look forward to feedback on the above discussion. Remember, the above is a constrained analysis based on some recent 3 year sales projections with all else being constant.

Disclosure: The author is long AAPL and encourages all investors to perform their own research prior to making investment decisions.

Comments

[Federico:]

Turleymuller,

all I am asking is where in the CC did Tim Cook say this.

Your best buy price discount makes sense.

I have my doubts that accessories would be any part of the subscription accounting.

Although apple did stop the revenue sharing model, the revenue from these carrier payments are more than made up with the reduced costs.

Sep 04 04:52 PM

[t0000:]

seekingalpha.com/artic...

"Recognized revenue from iPhone handset sales, accessory sales, and carrier payments was $1.69 billion during the quarter, compared to $419 million in the year-ago quarter, an increase of over 300%. The sales value of iPhones sold during the quarter was $2.9 billion."

middle of page 2nd pg



"I have my doubts that accessories would be any part of the subscription accounting"

They are not part of it, but they are recognized at POS...You have to account for this number.....about 1/2 of the new revenue apple reports is from accessories/carrier payments and other fees charged by aapl....This is why it is so important to to understand very little of the iPhone itself gets recognized each Qtr.

Sep 04 05:04 PM

[Federico:]

t0000

thanks...

Sep 04 05:20 PM

[Stefan Sida...:]

I appreciate everyone's comments. I don't have time till next week to give them a due diligence, but I will.

Stefan

Sep 04 05:56 PM

[LT:]

Very good work everyone. IF the economy improves, I think APPL beats these estimates. It appears Analysts are beginning to recognize this .

Sep 05 07:40 AM

[mollytjm:]

thanks for a great article...Muller and Zaky are good analysts, much better than the pack. I can't see anything stopping Apple's incredible steady rise up the numbers ladder on all fronts...market share, profits, popularity, stock price, units sold, apps available, etc.

Sep 05 09:43 AM

[User 22489:]

I am pleased with my Apple holdings, but there are two niggling worries - the open source competition and the very human trait of letting success become your master.
Apple is a very "walled garden" company. It is worse than MSFT in trying to control its universe. While that works, life is dandy and you thow off boatloads of cash. But that tends to stop (for various reasons) and then the only person thinking you have a still brilliant company is you and maybe your mom.
The second issue is far more dangerous - once successful, you start believing in your own divinity. You really think you know it all and better than everyone else. You cannot understand why the world does not continue to adore that new brilliant product.
The two feed off each other - the PE drops and shareholders want returns, not cash being hoarded that they cannot get their hands on.
Apple is already possibly the most "they could have had it all" company in the world as they threw away opportunity after opportunity.Fortunately they now have a couple of great products (yes, only a couple as the computing range is nothing special) but will still have to contend with a Nokia that (maybe) finally opens its eyes or a Google that allows everyone in or an Acer that hires in a designer or two.
Cy

Sep 05 10:11 AM

[t0000:]

I re-read the article, a few things. iPhones ASP are not coming down at the rapid fasion you have laid out. You are using adjusted GAAP numbers to come up with your results. You need to back all the numbers out and start from scratch to find aapl's true iPhone sales revenue. When doing this the figure that apple reported of 1,698 is almost irrelevent in this procces.

The way apple got to that figure was to divide current D/R by 4 = 5,467/4 = 1,367, that is the base number that apple would have reported if all iphone sales stopped 1 year ago. Now add 331 to 1,367 gets you 1,698 and what was reported. So apple added 331 to recognized revenue from Q3.......And only about 50 was from the actual iphone, the rest is from acc/carrier/other.

One more thing is your FY '09 EPS is way to low. You have apple only reporting 1.29 for Q4...Thats 7% under the street already....Thats is saying aapl is going to drop revenue by 4.5% from Q3 to Q4. Apple will be reporting about 1.65 bringing FY '09 to 6.11.

I don't want to try and figure FY '10 out right now because it is just a waste of time. It is to hard to say how aapl is going to account for China iPhones for this Q4 or Q1, and how much of the supply constraint effected iPhones sales elsewhere....Apple could report 12 million phones this Q4 or they could report 4 million. I need to see more on the China story before I start looking at FY '10.

By the way not trying to be malicious or anything just trying to help out and if anyone has any questions I would be happy to try and help. As I know appl's accounting is not cut and dry.

Sep 05 10:17 AM

[107Sid:]

Good luck Stefan. I'm fairly certain that you will find the sort of position that you are looking fro when you leave the Navy, even though it may take a little time. It's good that you are demionstrating your skills and intelligence on Seeking Alpha. Keep up the good work.

Sep 05 04:21 PM

[Turley Muller:]

t0000-
Dividing the current DR by 4 to find next quarters DR recognition is slightly flawed.

Current DR comprises of revenue to be recognized over next four quarters, but each quarter the amount isn't equal. There is only one period of 3Q07 sales left to recognize, and 2 qtrs of 4Q07 and 3 of 1Q08 left as well.

So these amounts (sales from Q3,Q4 '07 ,Q1 '08) that are included in current DR will obviously not be recognized equally over the next 4 quarters.

I don't know if that makes sense. I didn't realize this phenomenon right off, and it really is only important when trying to ascertain the mix of iPhone revenue reported for the quarter. Take a look at this link.

spreadsheets.google.co...

According to my estimations, Apple recognized 1,474 in DR for Q3. So 223 was from current period sales. Your estimate of 50 from handset sales is about what I have. Leaving the balance to accessories etc. as you mentioned. 2,900 in Q3 handset sales (50 recognized) is 362 per quarter (2,900 / 8 qtrs) with 7 periods of 362 and the final being 312 (362 - 50)


On Sep 05 10:17 AM t0000 wrote:

> You need to back all the numbers
> out and start from scratch to find aapl's true iPhone sales revenue.
> When doing this the figure that apple reported of 1,698 is almost
> irrelevent in this procces.
>
> The way apple got to that figure was to divide current D/R by 4 =
> 5,467/4 = 1,367, that is the base number that apple would have reported
> if all iphone sales stopped 1 year ago. Now add 331 to 1,367 gets
> you 1,698 and what was reported. So apple added 331 to recognized
> revenue from Q3.......And only about 50 was from the actual iphone,
> the rest is from acc/carrier/other.
>

> By the way not trying to be malicious or anything just trying to
> help out and if anyone has any questions I would be happy to try
> and help. As I know appl's accounting is not cut and dry.

Sep 06 07:06 AM

[Stefan Sida...:]

Sorry. I thought about this and decided not to share for several reasons:
- as pointed out in the comment here, I do have some errors in my iPhone ASP calculation.
- It is not well documented and would be hard for someone else to use. Heck, if I go to long without looking, I may forget what I did.
- I am happy to share my methods as I did in this article, but I feel like the actual spreadsheet is my intellectual property. Hope that doesn't sound selfish. I'm hoping others who have developed their own, will get different or same results to either confirm we are correctly analyzing Apple, or come up with a technique/insight someone has missed.


On Sep 04 12:12 PM user1010101 wrote:

> Would you be able to email me your spreadsheet?

Sep 09 10:47 PM

[Stefan Sida...:]

iPhone ASP correction:
Thanks to t0000 and Turley for pointing out my error with calculating the iPhone sales. I made some tweeks to my model and got some very good results. I am using the reported iPhone and Apple TV current and non-current deferred revenues to reverse engineer the actual sales for each quarter going back to 2007 by using the solver function in Excel. I got it to work great for the deferred cost (<1.5% error for any current/non-current value), but was having trouble with deferred revenues. The problem: I was trying to force iPhone actual sales into the model presented in my article. When I deleted that requirement, I got the deferred revenue error (current and non-current) down to less than 1% for any given quarter back to Q2 2008. The output is: actual iPhone and Apple TV sales, (actual iPhone sales in from CC)
2009
Q3 3102 (2.9B )
Q2 2321 (2.2B )
Q1 2552 (2.6B )
2008
Q4 4332 (not reported)

The difference is Apple TV sales running around 150m per quarter. For Q1 2009, the iphone sales from the CC exceed my calculated iphone and Apple TV sales. I have no explaination at this time. I could force Excel to a number greater than 2.6B, but I haven't yet.

WRT the (current deferred rev)/4 equations I am seeing in the comments. I am with Turley and do not agree that this is correct. The below is a cut and paste of the 2009 Q2 current deferred revenues for Q3 2009 through Q2 2010. Each column represents a quarter. It adds up to 5551 vice 5467 due to the reverse engineering, but it is close enough. This shows that the current deferred revenues is made up of differing amounts from the 4 forward quarters, so you can't just divide current deferred rev. by 4. A table needs to be built.
0
117 59
45 45 22
134 134 134 67
48 48 48 48
542 542 542 542
319 319 319 319
290 290 290 290

Also, I found that I get better results if I defer revenues as follows over 9 quarters: 1/16 1/8 1/8.... 1/8 1/16. This makes sence since the sales from the iPhone sold during the last week of August will not contribute 1/8 of its sales to the 4th quarters revenue, while the iPhone sold on the first day of the quarter will contribute 1/8 of its sale to the current quarter. Thus on average (assuming smooth sales of iphones across the quarter, which isn't always the case) iphones contribute 1/16 of sales to the current quarter and the ninth quarter away. This is why the numbers in the rows of the above table drop off.

Stefan

Sep 10 12:40 AM

[Stefan Sida...:]

On Sep 05 10:17 AM t0000 wrote:

> I re-read the article, a few things. iPhones ASP are not coming down
> at the rapid fasion you have laid out. You are using adjusted GAAP
> numbers to come up with your results. You need to back all the numbers
> out and start from scratch to find aapl's true iPhone sales revenue.
> When doing this the figure that apple reported of 1,698 is almost
> irrelevent in this procces.

What I did (which I admit was flawed) is really a mixture of GAAP and non-GAAP. As pointed out the 1698 GAAP (iphone + carrier + accessories) + 1405 GAAP to non-GAAP adjustment gives a mixture of actual iPhone sales + accessories + carrier revenues. I adjusted my model and have decided to re-publish the iPhone sales/ASP portion.


> The way apple got to that figure was to divide current D/R by 4 =
> 5,467/4 = 1,367, that is the base number that apple would have reported
> if all iphone sales stopped 1 year ago. Now add 331 to 1,367 gets
> you 1,698 and what was reported. So apple added 331 to recognized
> revenue from Q3.......And only about 50 was from the actual iphone,
> the rest is from acc/carrier/other.

I don’t agree with you here. As Turley mentioned and I put more detail in my previous comment posted, 5,467/4 is not a meaningful number, thus the 331 is also not meaningful. I’ll detail my method more in my follow-up to this article.

> One more thing is your FY '09 EPS is way to low. You have apple only
> reporting 1.29 for Q4...Thats 7% under the street already....Thats
> is saying aapl is going to drop revenue by 4.5% from Q3 to Q4. Apple
> will be reporting about 1.65 bringing FY '09 to 6.11.

I realized this and need to do some more work on my model. What caused it was the spike in Q4 2009 estimated iPhone unit sales to 11,000. Apple recognized SG&A as it occurs, so a big iPhone sales spike will spike SG&A, but most of the revenues are deferred, so GAAP earnings take a hit. I need to do some more modeling of how SG&A costs are tied to fluctuating iPhone sales. Any comments on this would be appreciated.

> I don't want to try and figure FY '10 out right now because it is
> just a waste of time. It is to hard to say how aapl is going to account
> for China iPhones for this Q4 or Q1, and how much of the supply constraint
> effected iPhones sales elsewhere....Apple could report 12 million
> phones this Q4 or they could report 4 million. I need to see more
> on the China story before I start looking at FY '10.
>
I figure that estimating something is better than nothing.

Sep 10 12:12 PM

[Stefan Sida...:]

On Sep 06 07:06 AM turleymuller wrote:

My DR recognition for Q3 2009 is 1,494 (close to yours). From the current quarter I have 194 recognized, the 7 quarters of 388 followed by a ninth quarter of 194. Now, this is for iPhone and Apple TV, thus it adds up to 3,102. Subtracting 2,900 gives 202 from Apple TV. The problem: 2.9B is anywhere between 2850 and 2949, so the Apple TV sales is +/- 50.

> According to my estimations, Apple recognized 1,474 in DR for Q3.
> So 223 was from current period sales. Your estimate of 50 from handset
> sales is about what I have. Leaving the balance to accessories etc.
> as you mentioned. 2,900 in Q3 handset sales (50 recognized) is 362
> per quarter (2,900 / 8 qtrs) with 7 periods of 362 and the final
> being 312 (362 - 50)

Sep 10 12:23 PM

[Turley Muller:]

Stephan... Hell of job here, fantastic work. Seeking Alpha needs more articles like this as it's of tremendous value. Not surprised though, being that you are submariner which are the brightest in the armed forces. Thanks for sharing your work.

Sep 11 07:39 PM

[Stefan Sida...:]

Turley, Thanks and appreciate the feedback.

Stefan


On Sep 11 07:39 PM turleymuller wrote:

> Stephan... Hell of job here, fantastic work. Seeking Alpha needs
> more articles like this as it's of tremendous value. Not surprised
> though, being that you are submariner which are the brightest in
> the armed forces. Thanks for sharing your work.

Sep 11 09:50 PM

[t0000:]

Regarding your information you provided in the comments section looks very good. Good Job and your model is much more accurate than any of the paid analysts on the Street.

Some of the figures I listed were just generalizations and you have more accurate numbers that you have stated in your comments. Regarding me not wanting to talk FY '10 numbers was in no way disrespect to you. I personally didn't want to explain any of the numbers in my comments. You are right, better than nothing. I have a model, just don't want to give it to much weight until I see some more data.

Good Luck

Sep 15 03:08 PM

[Stefan Sida...:]

t0000, appreciate all the feedback. It really helped me fix a problem with my model and allow me to post my newest article. Really need to pay attention to what is in each number Apple reports. There are numbers reported with varying combination of iPhone, Apple TV, carrier revenues and "related products and services" revenues.
If the FASB change comes out, I'll be able to throw away my spreadsheet.

Stefan


On Sep 15 03:08 PM t0000 wrote:

> Regarding your information you provided in the comments section looks
> very good. Good Job and your model is much more accurate than any
> of the paid analysts on the Street.
>
> Some of the figures I listed were just generalizations and you have
> more accurate numbers that you have stated in your comments. Regarding
> me not wanting to talk FY '10 numbers was in no way disrespect to
> you. I personally didn't want to explain any of the numbers in my
> comments. You are right, better than nothing. I have a model, just
> don't want to give it to much weight until I see some more data.
>
>
> Good Luck

Sep 15 06:41 PM

[pk de cville:]

I might be missing something about the last 2 lines of your last spreadsheet, but here goes:

Aren't the values for GAAP and non-GAAP Target Stock Price reversed? (The GAAP values are actually the non-GAAP values and visa versa.)

Great article. Thanks.

Sep 18 08:44 AM

[Stefan Sida...:]

You are missing nothing.
Thank you for pointing out this typo.


On Sep 18 08:44 AM pk de cville wrote:

> I might be missing something about the last 2 lines of your last
> spreadsheet, but here goes:
>
> Aren't the values for GAAP and non-GAAP Target Stock Price reversed?
> (The GAAP values are actually the non-GAAP values and visa versa.)
>
>
> Great article. Thanks.

Sep 18 09:43 PM