Analysing Apple Analysts’ Arithmetic
Leading analysts say a million iPhone sales may be missing from Apple’s next accounts, while Wal-Mart could sell as many iPhones as the Apple store itself. But both forecasts are based on some pretty large assumptions.
If you’re a regular MAC.BLORGE reader (or follow most Apple news sites), you’ll be familiar with the names Shaw Wu and Gene Munster. They are analysts for financial advisors Kaufman Brothers and Piper Jaffray respectively and regularly brief clients on how market conditions will affect firms such as Apple – and thus whether it’s best to buy or sell their stock.
The forecasts both men make lead to a lot of press coverage (including on this site) because they are so influential, which in some sense creates a circular effect. But it’s interesting that although both men are under immense pressure to get things right, with millions at stake in both client investments and advisor fees, many of their forecasts really come down to common sense and some very basic arithmetic.
Take Wu’s latest comments for example: He’s forecasting that iPhone sales in the holiday season may be much stronger than expected because most estimates don’t take account of iPhone gift cards. Apple gets the cash straight away for these purchases, but accounting rules mean they can’t log the phone as sold until the recipient activates it. As this will be after Christmas in most cases (and into January in many), such sales may not show up until the accounts for the first quarter of 2009.
All of this makes perfect sense and seems a pretty valid point. The problem is that Wu, perhaps inevitably, can’t make any more specific forecast than that “several hundred thousand to one million” sales will be affected in this way.
Meanwhile Munster has put out an estimate of how well the iPhone will do in Wal-Mart, concluding sales ‘could’ reach the same total as those in Apple stores. It’s a lengthy explanation, which looks quite impressive, but here’s what it really comes down to:
There are just under seventeen times as many Wal-Marts as Apple stores in the US. So to achieve the same total sales, they’ll have to sell around one-seventeenth as many iPhones in each Wal-Mart as they do in each Apple store. Munster thinks this is possible.
Not only is this incredible simple logic with a huge margin of error, but the conclusion is very disputable. One-seventeenth doesn’t sound a high proportion, but you have to remember that buying an iPhone is one of the very few reasons to go into an Apple store in the first place.
With only a couple of dollars off the retail price, most Wal-Mart iPhone sales will rely on impulse buys from general shoppers rather than people making special trips. And a $197 product that requires a lengthy service subscription really doesn’t seem like an impulse buy. It’s also worth remembering Munster’s logic only holds up if the iPhone is stocked in every Wal-Mart store throughout 2009.
None of this is meant to be a condemnation the analysts: it’s easy for writers such as me to criticise when we don’t have multi-million dollar client fees at stake if we make incorrect forecasts. But it’s worth remembering that when we report on such analysis, the reasoning behind the conclusions is often much simpler than you’d think, for better and for worse.
Related Posts:
