Pouring Over Pepsi (PEP)

PepsiCo (PEP) has 18 global mega-brands, each with annual sales of more than $1 billion, and they include Pepsi Cola, Gatorade, Mountain Dew, Fritos, Lay’s, Doritos, Tostitos, and Quaker.  Pepsi is a global beverage, snack and food company and makes a wide range of snacks, carbonated and non-carbonated beverages, and foods that it sells in some 200 countries.

Company Background

Sales in PEP’s North American beverage segment fell 7% in the second quarter of this year while Coke’s sales fell only 1% according to Morningstar.  However, Pepsi’s snack food segment showed solid growth.  Its North American snack business is PEP’s most profitable sector, generating 33% of its total sales and 44% of its profits.  Through its Frito-Lay division, PepsiCo is the world’s largest snack food company, controlling almost 40% of the U.S. salty snack market and around 30% of the non-U.S. market.

PEP’s latest 10Q quarterly report shows that Frito-Lay is the largest of its six segments accounting for 30% of Net Revenues for the last 12 weeks and 33% for the last 24 weeks.  Sales of Pepsi Cola and other beverages by the Pepsi Americas segment amounted to 25% in both periods.  Bottle Case Sales, a common measure PEP uses for all of its soft drinks, declined 6% “reflecting continued softness in the North American liquid refreshment category.” (7-22-09, page 34) Worse, non-carbonated beverage volume in North America dropped 14%, primarily due to double digit declines in Gatorade sports drinks and Aquafina water.

And, PEP’s sales have declined for four consecutive quarters.

In August, PEP announced it had reached a deal to buy the outstanding shares of its two main bottlers for $7.8 billion, ending a months-long disagreement over the price.  Before the deal was final, Value Line thought the purchase would contribute to long-term growth of sales and EPS.  However, with declining sales of carbonated and non-carbonated soft drinks, I don’t understand what Pepsi hopes to concretely accomplish.

Below, I compare and then discuss AnnC’s SSG, which I got from BI’s First Cut page, with mine and with Take Stock.

PepsiCo                    (PEP)	AnnC, from BI’s First Call	Armin	Take Stock
Date	7-13-09	7-17-09	7-16-09
Data	S&P	Same	Hemscott
Price	$55.52	$56.66	$57.28
52 week High & Low Price	$72.25 & $43.78	Same &             Same	Not Included
Last Quarter of Reported Data	Q1 ending           3-31-09	Same	Same
Software Used	TK 5	Same	TS Online
Project Growth From End of	Last FY	Last Q	Last FY
Sales Growth	6.00%	8.00%	8.50%
EPS Growth	7.00%	8.00%	-5.1%
High PE	20.0	21.0	22.5
High EPS	$4.85	$5.13	$2.47
High Price	$97.00	$107.20	$55.49
Value Line Estimated High Price = $90-110 as of 5-1-09 and 7-31-09
Low PE	12.0	14.4                   (from 2008, lowest in last 10 years)	17.3
Low EPS	$3.49 (TTM)	Same	$3.17
Low Price	$41.90          (Low PE x         Low EPS)	$43.80             (Recent Severe Low Price)	$57.28               (same as Current Price)
Upside/Down	3.00	4.0	Impossible to Calculate
Total Return	13.8%	15.6%	2.2%
Final Result	SSG HOLD	SSG BUY	DON’T BUY
SSG Buy Under	$53.00	$60.85	$31.39
RV/PRV                 (no outs)	77.6/72.5	79.0/73/3	Not Included
Quality	N/A	A+	2.2 (Unacceptable)
PTPM – 5 yr ave	19.3%               Trend down	Same              Same	18.9%                      Trend N/A
ROE – 5 yr ave       with End Equity	33.4%               Trend down	Same                Same	Not Included
ROE – 5 yr ave      with Start Equity	N/A	33.8%               Trend down	33.9%                  Trend N/A
Debt to Equity –  5 yr ave	N/A	27.8%                Trend up	Not Included

DISCUSSION

1. EPS Estimates:

(A) AnnC’s SSG

** Ann used the BI/NAIC Preferred Procedure to estimate 7.00% EPS growth. Her PP was based on the following five estimates: 6.00% Sales growth [less] 18.0% Pre-Tax Profit Margin (overriding the 19.3% default) [less] 26.7% Tax (default) [less] -0- Preferred Dividends (overriding the $2.0 per share default [divided by] 1556 M Shares Outstanding (default) [equals] 7.2% EPS. Ann used 7.0% EPS.

** Compared to Ann’s 6.00% expected Sales growth, Zacks.com estimated 10.17% while Morningstar Premium estimated 4.00% internal growth.  However, Mstar recognized that PEP’s nearly 10.00% historical Sales growth in the last 5 years included several acquisitions and also the advantageous effects of a weak dollar.

** Unlike Ann, I no longer use the PP and think it involves too many estimates and too much guesswork. Moreover, if you start with a low estimate of Sales growth, you usually wind up with a very low EPS estimate. See: Pondering the Preferred Procedure, http://arminfields.wordpress.com/2009/03/28/pondering-the- preferred-procedure/

(B) Armin’s SSG

** When I did my SSG, the six analysts I always check were estimating long term EPS at an average of 9.775% with Zacks.com high at 11.53% and Value Line low at 8.00%. First Call via YahooFinance was 9.47%, Reuters.com was 9.65%, and S&P and FactSet CallStreet via CNN Money were both 10.00%.

** Three analysts contributed to the consensus at Reuters and ranged from 10.0% to 8.9%. The three analysts at FactSet ranged from 11.0% to 9.0%.

** I used 8.00% EPS based on Value Line’s estimate which was the lowest of all the estimates.  To know which is the lowest, we must check all six.

** PEP’s historical EPS has been steadily declining for the last 5 years from 11.0% in 2004 down to 2.7% in 2008.

(C) Take Stock

** TS estimated -5.10% EPS (that’s a negative 5.10%) which was a whopping 14.80% less than the average of the six analysts and 13.10% less than VL, the lowest of the six estimates.  I consider that unreasonably conservative and, once again, we need to check all six analysts to make that judgment.

 2. Forecast High Prices:

** Ann’s Forecast High Price was $97.00 which was close to, but not below, the low end of VL’s $90-110 estimate.

** I got $107.70 which was close to, but not above, the high end of VL’s estimate.

** Both forecasts seem reasonable to me as my rule of thumb is to never substantially exceed or fall below VL’s estimate, at least not without a good reason. See:     Determining What’s Reasonable and What’s Not: An Update, http://arminfields.wordpress.com/2009/07/15/determinung-whatsreasonable-and-whats-not-an-update/

** Take Stock’s Forecast High Price was $55.49 which was a huge 38% below the low end of VL’s $90-110 estimate and which, once again, seems unreasonable and irrational by comparison since PEP was then currently selling for $57.28.

3. Pre-Tax Profit Margin and Return on Equity

(A) PTPM

** With S&P data, which Ann and I both used, PEP’s 5-year average PTPM was 19.3% and trending down.  Down trends are usually a red flag indicating poor performance. However, S&P places PepsiCo in the Soft Drinks Industry whose 5-year average PTPM was 11.6%, substantially worse than PEP.

** With Hemscott data, which Take Stock used, PEP’s average PTPM was 18.9% while its industry average, this time in the Processed and Packaged Goods Industry, was 8.1% and again substantially worse than PEP which ranked third of 61 companies (as of May 6, 2009).

(B) ROE

** With S&P data, PEP’s average ROE was 33.4%, trending up, and its industry average of 20.9% was substantially worse.

** With Hemscott data, PEP’s average ROE was 33.3% and again its industry average of 21.0% was substantially worse. PepsiCo ranked 7 out of 61 companies.

 4. Final Results :

 ** Ann got a SSG Hold with a 3.0 Upside/Downside ratio and a 13.8% Total Return which did not satisfy the minimum 15.0% TR criteria.

** I got a SSG Buy with a 4.0 U/D and a 15.6% TR.

 ** Take Stock does not use the Buy, Hold or Sell criteria nor the U/D concept and it seems likely that TS would say Dump/Don’t Buy PEP.

** S&P rated PepsiCo’s quality as A+ while Take Stock rated PEP a 2.2, unacceptable.

 5. Final Thoughts:

** Ann projected future growth from the last FY while I used the last Quarter.  If she had projected from the last Q, her U/D would have been 3.1 (instead of 3.0) and her Total Return would have been 14.8% (instead of 13.8%). That is an almost SSG Buy and, as the FY progresses with one or two more reported quarters, is likely to be a solid SSG Buy.

** Regardless of what my SSG shows, I’m not impressed with PepsiCo.

- Armin

[please rate this post using the new, mouse-over star system at the top and/or leave a comment below]

Projecting Growth at Patterson (PDCO)

Mark Robinson, founder and General Manager of Manifest Investing, recently posted his SSG of Patterson Companies (PRCO) at BI’s First Cut page.  Mark used several extraordinary SSG methods: he added two years of estimated annual data, source and details unknown,  which extended his projection of growth for the next 7 years out to 2015. 

Also, Mark’s projected High EPS shows as $2.05 of the SSGs front page, but mysteriously as $2.50 on the back. The additional estimated data changed the 5 year average and trend of PTPM as well as ROE, and also led to a projected Low Price option, Low PE x Low EPS, that was higher than the current price even after he lowered his projected Low EPS by 40%.  All of this is seriously strange and decidedly questionable..

In January, I critiqued Mark’s SSG for Stryker (SYK) and his SSG methods, http://arminfields.wordpress.com/2009/01/31/studying-stryker-syk-and-mulling-over-methods/

Here, Mark has some surprises which I discuss after the following comparative table.  Armin-1 uses all of Mark’s judgments except it does not add any estimated annual data.  Armin-2 is identical except it uses 10.00% estimated EPS instead of Mark’s 7.00%.

Patterson Companies (PDCO)	Mark        Robertson	Armin-1	Armin-2	Take            Stock
Date	7-10-09	8-7-09	Same	8-11-09
Source of Historical Data	S&P	Same	Same	Hemscott/ Morning-  star
Price	$21.01	$25.00	Same	$21.89
52 week High & Low Price	$33.85 &      $15.75	Same &         Same	Same &     Same	N/A
Last Quarter of Hist Data	Q4 ending       4-30-09	Same	Same	Same
Software Used	TK 5	Same	Same	Online TS
Years of Hist Data	2001-2008	1999 -2008	Same	Same
Years of Esti-mated Data	2009 & 2010 ESTIMATED	NONE	NONE	NONE
Source of Est Data	UNKNOWN	None	Same	Same
Project Growth from	UNKNOWN	Last Q       of Hist Data	Same	Last FY of    Hist Data
Sales Growth	6.00%	Same	Same	2.90%
EPS Growth	7.00%	Same	10.00%	0.30%
High PE	22.0	Same	Same	25.6
High EPS	$2.50	$2.37	$2.72	$1.66
High Price	$55.00	$52.10	$59.80       (9% > VL)	$42.56
Value Line Estimated High Price =$40-55 as of 5-29-09 and     $35-45 as of 8-28-09
Low PE	14.0	Same	Same	15.4
Low EPS	$1.75(down from $2.90)	$1.70            (TTM)	Same	$1.69
Low PE x         Low EPS	$24.50                (> current price)	$23.80         (< current price)	Same	$26.03         (> current price)
Final Low Price	$16.00(“other” option)	Same	Same	$26.03        (> current price)
Upside/Down	6.8	3.0	3.9	Impossible to calculate
Total Return	21.2%	15.8%	19.1%	11.3%
SSG Buy Under	N/A	$25.03	$26.95	$21.28
RV/PRV          (no outs)	69.9/67.3	60.5/56.6	Same/55.0	N/A
Quality	N/A	B+	Same	1.1 (un-acceptable)
PTPM – 5 yr ave	11.3%               (end est    2010, trend even)	11.7%           (end 2008, trend down)	Same	11.2%         trend N/A
ROE – 5 yr     ave                 End Equity	17.1%             (end est  2010, trend down)	17.4%           (end 2008, trend even)	Same	N/A
ROE – 5 yr ave                Start Equity	N/A	19.3%         (end 2008, trend up)	Same	19.3%         trend N/A
Debt to Equity – 5 yr ave	N/A	30.5%           (end 2008, trend up)	Same	N/A

Discussion

- Mark added two years of estimated annual data (2009 and 2010, source and details unknown) to his SSG which meant that 1999 and 2000 were eliminated because our Toolkit software only holds 10 years of data.  Then, he eliminated three more years (2001, 2002, and 2003) as outliers to arrive at a 6.00% quasi-historical sales growth rate that he used as his projected sales growth for the next seven years to 2015.

- Projected sales growth is important to Mark because it is the first element of the BI/NAIC Preferred Procedure which he uses “almost exclusively” to determine his projected EPS growth.  Sadly, he did not set forth the other elements of his PP (Pre-Tax Profit Margin, Tax Rate, and Shares Outstanding) which, in Mark’s case, all had to be projected for the next seven years.  A simple command, Alt-R, could have set forth the PP on the SSG’s front page. 

- If this scheme wasn’t convoluted enough, Mark’s High EPS shows as $2.05 on the SSG’s front page, but as $2.50 on the back page.  It’s supposed to be the same so this is a serious and major discrepancy, there’s no attempt at any explanation, and Mark’s SSG cannot be replicated, at least not without the secret code.

- Mark’s PP resulted in a 7.00% EPS rate ($2.05 or maybe 2.50 estimated High EPS in 2015).  Unlike Mark, I no longer use the Preferred Procedure because it involves too many estimates and too much guesswork even under normal circumstances.  See: Pondering the Preferred Procedure, http://arminfields.wordpress.com/2009/03/28/pondering-the-preferred-procedure/

- In his First Call write-up, Mark insists:

“It’s not the EPS growth rate that matters. It’s the 5-year estimate for EPS [AF: in dollars] that forms the core calculation on the SSG.” He adds: “(For those of you who teeter on the precipice of a nervous breakdown over the EPS growth rate, it’d be something on the order of 7% — but get over it, it’s the 5-year EPS value that really matters and I derive that almost exclusively using the preferred procedure.)”

** Overlooking this needless hectoring, Mark’s core calculation is seriously scrambled because it shows as $2.05 on his SSG’s front page and $2.50 on the back (it’s intended to be identical);

** Moreover, at four major websites, long-term EPS estimates are expressed only as a percentage rate (and not in dollars): Reuters, CNNMoney, Zacks, and YahooFinance.  Two others show an EPS estimate as a percentage rate and as a dollar amount: S&P and Value Line.  No website I know of  makes a long-term EPS only in dollars.  See: Estimating EPS, http://arminfields.wordpress.com/2009/03/05/estimating-eps/

- Perhaps the biggest disappointment is that Mark never even tried to explain why it was necessary to add ANY estimated data since the norm is to project the next 5 years based on actual data.  How did he project our current recession?  Which web site did he rely on and for what data? Why two years of estimated data and not one or three years?

- Lastly, the default for the projected Low EPS is no growth at all which, for Mark’s SSG, was $2.80 estimated (entered as 2010 actual).  Mark lowered this by 40% to $1.75 but, once again,  offered no explnation. With 6% growth for his projected High EPS and -40% for his projected Low EPS, Mark’s growth projections seem flaky and foolish, especially adding two years of estimated annual data.

CONCLUSIONS:

(A) It turns out that every one of Mark’s machinations were unnecessary.  Armin-1 duplicates all of his judgments with one exception: it does not add those two years of estimated data.  We both got a SSG Buy and satisfied the minimum 3.0 Upside/Downside and the 15% Total Return criteria.  Armin-1 would look even better if I used the same $21.00 price as Mark did.

(B) Mark’s 7.00% projected EPS, after his many maneuverings, is not even close to what the analysts were estimating.  When I did my SSG, the six long-term EPS estimates I always check averaged 12.70% with Value Line low at 10.00% and Reuters high at 14.40%.   Thomson/FirstCall via YahooFinance was 11.75%, Zacks.com was 12.67%, FactSet CallStreet via CNN Money was 13.00%, and Reuters.com was 14.40%. 

** Armin-2 used 10.00% estimated EPS, the lowest of the six analysts and also satisfied the SSG Buy criteria.

(C) The substantial variation among these six long-term EPS estimates demonstrates, I think, the wisdom of not using any estimated annual data in our SSGs.  Unlike actual data, you can never be certain with any estimate and one website’s estimate is no more accurate than another’s.  Moreover, our SSG wants six pieces of info for each year of annual data which are tricky to estimate.  Note the different trends on PTPM and ROE that Mark got with his estimated data and that I got with actual data.

(D) Whenever we share our SSGs, at a club meeting or with BI’s First Cut or at another web site, all of us should try to make clear what we did and why, especially when we do something out of the ordinary.  For example,  Mark added two years of estimated annual data (source and details unexplained), used the Preferred Procedure (unexplained), got a High EPS of $2.05 on the front page and $2.50 on the back (unexplained), lowered his Low EPS from $2.90 to $1.75 (unexplained), and got a Low EPS x Low PE that exceeded the current price (unexplained).

(E) I think Mark’s SSG and his unique methods are unreliable, and neither helped me understand PDCO.

 What do you think?

[You can leave a comment below and/or rate this post by using the new, mouse-over star system at the top]

Armin