I recently browsed in an Apple store and it was a very fun experience: lots of innovative products, lots of customers, and lots of buzzzz.  Last week, the New York Times reported much the same experience.  You may have to register, but everything is free; go to: http://www.nytimes.com/2007/12/27/business/27apple.html 

The Times also reported that Apple’s retail stores are very successful (unlike Dell, Nokia and Sony) accounting for 20% of its sales, some $1.25B out of $6.2B total in its recent 4^th quarter, and up 42% from the year before.  A large part of its retail success is its special treatment of customers: lots of friendly staff, readily available technical help, and one-on-one in-store trainers.

The latest Value Line reports that AAPL, in its 4^th quarter, sold 2.1M Macintosh computers, 102M iPod music players, and 1.1M iPhones, and ended its FY with overall sales increasing 24% and EPS up 73%.  Despite its glowing report, VL without comment lowered its expected EPS growth from 31.0% to 26.5%.

Soooooooooo, I decided to revisit my SSG and here’s what I found:

Observations:

- Take Stock’s Forecast High Price in the next 5 years is less than AAPL’s current price and therefore TS forecasts a -7.7% Total Return; Take Stock’s $119.02 High Price is also substantially (some 28%) below the low end of Value Line’s High Price estimate of $165-250;

- The analysts are estimating long-term EPS at around 20.0% with Value Line high at 26.5% and First Call, FactSet, and S&P exactly at 20.0%.  Zacks is estimating 22.41%, Reuters is 21.67%, and Reuters less 1 Standard Deviation is 16.85% (where I got the 17.00% I used in Armin-1).  Take Stock, on the other hand, is estimating a measly 1.57% EPS growth in the next 5 years;

- AAPL has had a big run-up in price and, at $171.25 is close to its 52 week high ($202.96 high and $82.86 low) gaining 35% in the last 6 months;

- AAPL is not close to a SSG Buy and, with Hemscott data, would have to drop to $143 per share (down 20%) to satisfy the NAIC/BI Buy criteria even when I use optimistic judgments as in Armin-2, or drop to $151 (down 12%) with S&P data as in Armin-3.  To be a Buy with Take Stock, AAPL’s price would have to drop a whopping 65%, to an unrealistic $59.51 per share.

Armin Fields

Hi Armin,

Isn't this more an issue of why does the Business Model/Preferred Procedure calculation come up w/ an estimated 5yr EPS of forecast of 4.25 (1.57%) using AAPL data?  Take Stock, based on historical data, as noted in the TSSW Front Growth Stat% box predicts future sales/eps at 15.2 based on past performance, resultant in a 5yr eps forecast of 7.99 (and a more 'realistic' future high price).  But in the interest of the new investor, and again until we have the 'advanced' version available online where one can use their own judgments for future sales and eps, TS online currently defaults, as noted in red at the bottom of the tssw front "Forecast Earnings" section, to the lower of the business model/preferred procedure vs. the forecast numbers found in the Growth Stat% box.

So my question really is why does using the preferred procedure/business model EPS forecasting mechanism for AAPL result in such a wide discrepancy between historical results and forecast future results? And is there a good reason in your opine i wonder?

dave

Hi Dave:

You wrote:

“…why does using the preferred procedure/business model EPS forecasting mechanism for AAPL result in such a wide discrepancy between historical results and forecast future results? And is there a good reason in your opine i wonder?  I underst “

Well…the Preferred Procedure starts with a Sales growth estimate: if the beginning point is (too) low, the ending point also will be (too) low. 

Then, the Preferred Procedure involves three additional judgments about the next 5 years (Pre-Tax Profit Margin, Tax Rate, and Shares Outstanding) providing even more room for error and/or wide discrepancy.

Add to these factors the NAIC/BI guru advice to be conservative as a general rule-of-thumb (which usually means without regard to what the company & industry trends are indicating) and you have a formula for the Preferred Procedure to often result in a low-ball EPS estimate.

These are some of the reasons that led me to reassess using the Preferred Procedure and you can read more of my thinking at my blog, http://stock-analysis-with-ssg-and-pert-a.blogspot.com/ (scroll down to “Considering Kohl’s (KSS) and Reconsidering NAIC’s Preferred Procedure,” September 13, 2006.

No matter how I derive my long-term EPS estimate, I ALWAYS want to know how it compares to the current, company-specific EPS estimates from FactSet, First Call, Zacks, Reuters, Value Line, S&P and, of course, Take Stock Online.

What do you think?  Comments, of course, are welcome from anyone….

And, what were you thinking when you wrote (“I underst”) which got cut off?

Armin

Ps: I think the better question to ask, and one with a much different answer, is why does Take Stock Online so often come up with an unrealistic (or unreasonable) EPS estimate?

Hi Armin,

First of all I always appreciate your detailed feedback and analysis, and your keen eye for where (and how) in your opine we can strive to improve our offerings.

"(scroll down to “Considering Kohl’s (KSS) and Reconsidering NAIC’s Preferred Procedure,” September 13, 2006."

I will do so and hope to better understand why PP seems to come up with valid, imo, forecasts for some co's and seemingly invalid (again in imo) estimates for others so that i may provide a more worthy debate on this topic.

"And, what were you thinking when you wrote (“I underst”) which got cut off?"

I don't know/recall, it was an error i didn't catch before posting which is why i removed it.  I must have thought i understood something and then changed my mind? :)

"I think the better question to ask, and one with a much different answer, is why does Take Stock Online so often come up with an unrealistic (or unreasonable) EPS estimate?"

I think in most circumstances when this is the case (seemingly unreasonable forecast eps), Take Stock is defaulting to the bus. model/Pref Proc calculation (which is usually the lowest possible tmk), in the interest of having a new investor avoid risks.  Again once the advanced version is available the user will have the choice to accept these defaults or to make their own judgments, if upon looking at these historical data, and considering things like industry/company trends that fall outside of the historical data presented, they conclude that TS is being unreasonably conservative.

In cases where defaulting to the lowest possible eps estimates are not the result of using the PP/Bus. Model, i think i recall that ET discussed previously that there are regression algorithms that weight more recent quarters more heavily in this regard which is intended.  This is discussed here.

I hope to continue to research your feedback and hope that we can in the future come up with specific recommendations for improvements to Take Stock Online, as well as our other offerings and i appreciate your comments.

Dave