Proctor & Gamble (PG) was the third Online Stock Study at the Better Investing website and was led by Gretchen Hurt, a volunteer educator, who is VP of Education for the BI Volunteer Advisory Board and VP of the OKI-Tri State Chapter of BI.  Thanks Gretchen for a most excellent study.

Gretchen’s audio-visual presentation and related PDF slides are both available for downloading.  Other materials are a completed SSG in PDF format, PERT-A Worksheet, Fact Sheet from PG, and a thorough background paper on PG prepared by Gretchen.

Company Background:

This last handout explains that PG changed its business model some 7 years ago to focus more on its “billion dollar” brands.  It now has 23 such brands, with Pampers the largest at $7B in yearly sales, up from 10 big brands in 2000 and 17 in 2005.  It also has 18 other brands with annual sales of $500M-$1 B.  Beauty products account for 30% of sales; Gillette, acquired in 10-05, makes up 12%; and its largest customer, at 15% of sales, is Wal-Mart.  PG will aggressively divest brands that have a low profit margin or little growth potential.

According to Reuters.com, PG had $79,740M in sales during the last 12 months, some six times more than Colgate-Palmolive.  Close competitors, not peer group members, are Johnson and Johnson, Kimberly Clark, and Unilever (private) as per Hoovers.com and YahooFinance.

SSG Comparisons:

Here's the Better Investing Consensus SSG compared to my SSGs, one with S&P data and the other with Hemscott data (but with the same judgments) and to Take Stock.

* I updated the Consensus SSG with current data and price on 4-23-08, but did not change any of its judgments.

Discussion -  BI Online Stock Study:

- Study participants were asked to make 5 judgment calls about the next five years after each issue was discussed by Gretchen: Sales Growth, Shares Outstanding, EPS Growth, and Low Price.

- To estimate future Sales growth, participants were given 5 choices: 8.3%, 10 year historical average; 15.2%, 5 year historical average; 12.1%, last year’s growth; 7.5%, last quarter’s growth; or open-ended “other.”

- Estimating Sales growth was especially important to the consensus SSG because it used the Preferred Procedure (PP) to estimate future EPS growth.  The participants choose 8.3%, the 10 year historical average, which became the first factor of the PP which also requires estimates for the next 5 years of Pre-Tax Profit Margin, Tax Rate and Shares Outstanding.

- They were then asked to estimate the number of shares outstanding in the next 5 years as part of the PP: either 3105M, the TK5 default which is last year’s actual; 3132M, Value Line 2007 actual; or 2950M, VL’s estimate in 3-5 years.  The consensus was 2950M, VL’s estimate.

- To decide future EPS growth, participants chose from: 10.0%, 10 year historical average; 10.2% from the PP; 8.7%, 3 year historical average, 7.1%, 2 year historical average; or open-ended “other.”  They selected 10.0%, the 10 year historical average.

- Forecast High and Low PEs were not voted on because PG has such a consistent track record: there were no outliers, almost no volatility, and competitors looked similar [Also compare Take Stock’s choices, which always uses the lowest 5 High & Low PEs in the last 10 years (23.1 & 19.0 here) with the my SSG s High and Low PEs which used the last 5 year average.  The numbers were nearly identical with Hemscott or with S&P data.]

Discussion: Armin’s SSGs and Take Stock:

- When I did my SSGs, six different analysts were estimating long-term EPS at around 11.5-12.5% with Reuters high at 13.18% and Value Line low at 9.5%; Zacks was 11.61%, FactSet CallStreet was 12.00%, First Call was 12.14% and S&P was 12.50%.  Reuters less one Standard Deviation (which the website provides) was 13.18 - 3.13 = 10.05% and that was the basis for my SSG’s 10.00% EPS estimate.

- VL was estimating 11.00% EPS when the Consensus SSG was prepared; however, four months later when I did my SSG, VL had lowered its EPS estimate to 9.5%, the lowest of all the analysts I checked.  Strangely, VL continues to estimate 100% Earnings Predictability even though it has revised its EPS estimate in its last three reports for PG.

- I almost never use the Preferred Procedure (PP) because it involves too many estimates and too much guesswork for me.  Here’s why:

 ** The PP depends on making four estimates for the next 5 years: future Sales, future Pre-Tax Profit Margin, future Taxes, and future Shares Outstanding.  VL makes an estimate for two: future Taxes and future Shares Outstanding.   However, the most important estimate is future Sales, and VL does not make an estimate for future sales, only for future Sales per share which is not used by the PP.

** I don’t mind relying on VL estimates, even though it has revised its EPS estimate for PG three times in the last 3 reports, but I don’t like relying only on VL estimates for the PP.  And, I especially don’t want to rely on historical Sales growth to estimate future Sales growth….deciding to use the last 2 or 3 year average, or last 5 or 6 year, or maybe the last 10 year average, seems like guesswork to me.

** My SSG’s default PP with Hemscott data results in 8.5% EPS, but if I use VL’s estimates for Taxes and Shares Outstanding, I get 11.8%.  With S&P data, the default is 10.6% which becomes 12.1% with the same two VL estimates. The differences are even greater if I start with the 8.3% Sales estimate from the Consensus SSG.  Then, the default drops from 8.5 to 6.7% EPS with Hemscott data, and from 10.6 to 8.8% EPS with S&P data.  For my estimated EPS growth rate, should I pick 8.5%, 10.6%, 11.8%, or 12.1%, or maybe 6.7% or 8.8%????

** All in all, the PP involves too many estimates that I find too unreliable.  And, this means too much dubious guesswork for me.

- I never want to substantially exceed or fall below Value Line’s estimated High Price, at least not without a good reason.  The Consensus SSG was below VL’s High Price estimate (which is good) and, four months later, my two SSGs were $5-8 per share above VL’s current estimate, some 5-8% higher (which is ok for me).

- However, Take Stock’s High Price of $95.15 was only fifteen cents higher than  the low end of VL’s estimate, probably because TS’s 6.3% EPS estimate was about half of what the analysts were estimating.

- Take Stock gave PG an unacceptable Quality Rating of 3.2.  TS requires a minimum of 3.4 to pass muster and 6.7 is desired with 10.0 the max.  S&P, on the other hand, gave PG an A+, its highest rating out of eight grades.

Conclusions about Comparisons:

(1) Comparing the original Consensus SSG to its update four months later, PG’s price has fallen 11% or $8.00 per share.  Although I made no changes in its judgments, the Relative Value and Projected Relative Value both dropped below 100%, the Total Return improved by 3%, and the Upside/Downside Ratio increased from 3.0 to a whopping 9.1.

(2) No matter how we determine our EPS estimates, analyst estimates can tell us if we are in the ball park or if we need to rethink this crucial SSG judgment. My 10.0% EPS projection was based on the Reuters estimate less 1 Standard Deviation, compared to the BI Consensus SSG’s Preferred Procedure of 10.2%, and both our SSGs decided to use the same 10.0% EPS estimate.

(3) Take Stock, by comparison, seemed off-the-wall: its EPS estimate was way low, almost 40% lower than the Consensus SSG’s and mine; its Forecast High Price was at the very low end of VL’s estimated High Price; and its Quality rating was “Unacceptable” compared to S&P’s A+.

(4) Data Comparisons: There were a few “inexplicable” differences between the S&P and Hemscott data: Take Stock’s price for PG was $2.76 more with Hemscott data than the Hemscott data file although both had the same date; S&P, like Value Line, reduced 2002 EPS by .19/share for non-recurring losses, but Hemscott did not; and the S&P data file shows preferred dividends, which are part of the PP, but Hemscott does not.

Armin Fields

The Comparing SSGs Workshop will continue in this forum on Monday, May 12 with a discussion of the BI Online Stock Study SSG for Colgate-Palmolive (CL).  Please join me…..