Apple Makes $831 On Each AT&T iPhone

Ponca City, We Love You writes "The NYTimes reports that Gene Munster, an analyst at Piper Jaffray, has studied Apple's financial statements and come to the conclusion that AT&T is paying Apple $18 a month, on average, for each iPhone sold by Apple and activated on AT&T's network — up to $432 over a two-year contract. This shows how much incentive Apple has to maintain its exclusive deal with AT&T rather than to sell unlocked phones or cut deals with multiple carriers. Last week Apple disclosed that 250,000 iPhones had been purchased but not registered with ATT that Apple thinks are being unlocked so Apple has now taken action to curb unauthorized resellers by limiting sales of the iPhone to two per customer and requiring that purchases must now be made with a credit or debit card — cash will not be accepted." The latter article links to a US Treasury page explaining the incorrectness of the widely-held belief that cash cannot be refused for any transaction.

Comments

that math is wrong

[G Fab]

as correct as that explanation is for the 831 number, the math is wrong.

apple doesn't get iphones from fairies. They pay money to build them.

Re:that math is wrong

[mr_matticus]

The math is not wrong. Apple gets $831 from each iPhone. That doesn't mean it profits $831. It has a number of costs, from raw materials to labor to prorated warranty costs, all the way to packaging, shipping, inventory management, advertising, and ongoing software development; there's always then burden-shifting among products--some products may subsidize others and therefore have an apparently increased profit margin to cover the lower margins on a different product. Deducting all of these would be impossible for an analyst to do without intimate knowledge of Apple's overall operation.

It's better to report the total without them taking wild-ass blind guesses as to how much of that is profit (like iSuppli's crazily inadequate "what it costs" figures). Even if those numbers are right (and sometimes they just pull costs out of their ass because it's "close enough" to something they've seen before), that still only gets you to gross profit. And at the end of the day, gross profit is nowhere even close to the much smaller net profit.

Re:that math is wrong

[mr_matticus]

They're not reasonable figures, though, is the entire problem.

When they don't have data on a particular component, they use something they deem to be relatively similar. They extrapolate an approximate price based on what they feel is an appropriate price at a given (assumed) volume level. They never seem to account for time or place of purchase, either, which can be significant factors in volatile markets. For example, they used a run of the mill touchscreen price for the iPhone, without multitouch and without the daylight-readable backlighting.

Each step of the game is an approximation adding further error to the calculation, and by the end, they almost invariably end up at a "cost" figure that is below reality, sometimes significantly. I have some experience in various litigation involving some of the products they've assessed, and based on what we get in discovery, iSuppli's numbers are, in comparison, highly conservative and geared toward getting the highest possible gross profit rather than providing the most accurate figure. They generate the biggest stir when people think that actual manufacture costs peanuts, so it makes sense from their perspective, but it does a disservice to everyone.

Re:that math is wrong

[mr_matticus]

What is the value in utterly unreliable numbers? It provides absolutely no insight--products cost substantially less for the pieces than the finished good and retail price. Shocking!

Without being able to get within 20% in some cases of the actual materials cost, it doesn't inform any conclusion about the product. The general gross margin range they report is 25-50%--practically that entire variation is within their margin of error in reporting the figures in the first place. Thus, the assessments, apart from being nerd porn, are perfectly vacuous.

I think most people can figure out that almost nothing is sold without a gross margin of at least 20%, and that 50% isn't terribly uncommon either. Unless iSuppli shows up with a 75% margin one day, there's nothing useful about it.

Re:that math is wrong

[dedazo]

The math is not wrong. Apple gets $831 from each iPhone. That doesn't mean it profits $831

Then perhaps the obvious flamebait headline should have been crafted to reflect that simple fact. But in the age of one-liner evangelism, Apple Makes $831 Revenue (Though Not Really Profit, Mind You) On Each AT&T Phone Although That's Pretty Much Irrelevant To Everything, We're Just In It For the AdSense Revenue just doesn't work.

I'm having trouble trying to understand the mindset of people who think $831 or $8,311 represents "greed". If the market will bear it, that's the correct price. Otherwise Apple would have sold 1,000 iPhones instead of 100,000 or however many they've shipped so far.

Re:that math is wrong

[jollyreaper]

I'm having trouble trying to understand the mindset of people who think $831 or $8,311 represents "greed". If the market will bear it, that's the correct price. Otherwise Apple would have sold 1,000 iPhones instead of 100,000 or however many they've shipped so far. "Whatever the market will bear" is a two-way street. Businesses will try to whore their wares for top dollar but customers (I despise the label "consumer") will try to find the best deal possible. If they know that there's a shitload of profit built into a given deal, customers will try to beat down the vendor so they can pay less. And that's a fair move in capitalism. If a vendor I use is making a 15% markup, that may well be completely fair and just. If he's making a 75% markup, I want to know about it, and I'll certainly try to prevail upon his better nature or take my business elsewhere.

Re:that math is wrong

[Durandal64]

Except that iSuppli doesn't attach error bars to their estimates. They just state them as-is. Without error bars, estimates that have error are useless.

Re:that math is wrong

[eclectic4]

" And that's a fair move in capitalism. If a vendor I use is making a 15% markup, that may well be completely fair and just. If he's making a 75% markup, I want to know about it, and I'll certainly try to prevail upon his better nature or take my business elsewhere."

You don't understand... if something is marked up 75%, there most likely isn't business elsewhere. Is someone else selling iPhones besides Apple? Are they cheaper? Get the picture? If I told you Oil was being sold at a 100% markup, are you going to trade in the car for a bicycle as a show of "taking your business elsewhere?" No. You are either going to buy this "widget" for "this price" or you aren't. Supply and demand set the price, not it's known markup. If you are going to use markup in your purchasing decisions, know that it will have zero effect on everyone elses purchasing tends, and therefore will do nothing more than satisfy your strange needs to not give too much profit, even if demand supports it.

Good luck.

Re:that math is wrong

[mr_matticus]

Getting to within 10% of the cost of goods sounds fantastic to me. Within 20-25% still sounds not bad. It's a lot better than a total guess, which seems to be what you're suggesting. (gross margins of between 20 and 50 %). Except that it's not. A range of 20% is no better than a total guess, and a range of 10%, while narrowing the gross margin gap by about a third, does not exceed "total guess" for net profit (which ultimately is the only one that matters). In what way do you consider it superior to a total guess?

An error of 20% on a parts bill will move you from one end of the gross margin scale to the other. In other words, the error of the estimated cost rarely does any better than the standard margins on products. Then there is absolutely no guidance whatsoever on net profit figures, which are what really matter. The company itself can be assessed, but it's pure folly to try to do it for individual products.

Let's take a product. Retail price $100, iSuppli guess: $58 for parts. That's 42% gross profit, and we will use your conservative and overly generous 10% range for accuracy. The possible gross margins range from 47% to 36%. What does this mean? That they fall in the normal range of 20 to 50%. What did we learn? Nothing. We already knew it was almost certainly going to be in that range, and regardless of where it falls, it's unremarkable because it's the normal range. The only place it would be noteworthy would be if iSuppli found figures grossly outside that range (e.g. 75% or 5%), and that basically never happens.

What do you gain by knowing that the product is within a normal range of markup? The "smaller iSuppli margin" product could easily be the bigger-margin cash cow, and an attempt to minimize the pocket-lining of corporations can't be undertaken with the information iSuppli supplies, if you'll pardon the pun.

When your margin of error covers the most of the basic spread of possibilities, you're not providing a service. Trying to peg it down to some quasi-accurate Ouija-board figure without any real knowledge gets us nowhere useful. iSuppli rarely, if ever, has provided anything better than a 20% range on a 20% range, which means it has never demonstrated or even rationally suggested that any given product is a better "value" than any other. It relies on faulty analysis for people to make that claim and gives them a quasi-factual, half-true basis to do so. This can only cause harm.

The entire system is highly variable from company to company, and even among products from a single company. Without details, it is impossible to get any accuracy beyond a massive range. Gross profits are usually 20 to 50 percent. Net profits for self-sustaining (i.e. not loss-leaders) products are usually 5 to 20 percent. Anyone offering you any level of accuracy beyond that without specific documentation is lying.

useful information

[m2943]

People say things like "it's Apple's right" and "good for them". Of course, it's Apple's right to do those deals.

Nevertheless, where do you think this money is coming from? Do you think that AT&T is giving that to Apple because they are such good buddies?

No, you are paying for it one way or another (e.g., by paying a premium for their sluggish EDGE service).