Friday, 18 September 2015

Petsmart, The Fat Pitch, and Margin of Safety

I'd like to do a little backtracking here by revisiting an idea I had that worked out.  Trust me, I've had tons of ideas that haven't worked out, the key to success at this game is ensuring that the ideas that work outweigh the ideas that don't in terms of return (and risk).

There's an old adage that "Luck is what happens when preparation meets opportunity" (courtesy of Seneca). I'd go further and state that success, not just luck, is what happens when preparation meets opportunity.

And where better can the concepts of luck and success exist simultaneously (and as an aside, where better can luck be so perversely confused with success)? Why, where else but the stock market!

Back in the spring of 2014, I had begun my journey away from what I would describe as impulsive, unsystematic trading, the catalyst being my devouring of The Intelligent Investor (twice in a row, successively), towards a yearning for wanting to understand and analyze the actual  business fundamentals of the tickers I had been flipping indiscriminately for years.

The Intelligent Investor provided me with the beginnings of a conceptual framework for valuing businesses. But, at the time, I hadn't fully developed my own framework for analysis beyond what I had read and absorbed, courtesy of Ben Graham (I'll humbly offer that I still haven't fully developed a robust framework even to this day, as my current evaluation methodology is still a work in progress).

In May 2014, Petsmart (which was still public), missed earnings estimates and issued soft guidance, and if I recall correctly, the stock got demolished over the next two trading sessions post earnings. Here are the charts illustrating the reaction:

My initial impulsive response was to buy the stock on weakness for the following reasons:

  • I was familiar with the company, and I perceived the business to be an annuity type business, a one-stop shop for all things pet related housed in a nationwide chain. I categorized Petsmart as a Walmart or Costco for pets.
  • I had observed that a few local warehouse locations near me had closed, I equated the closures to proactivity on the part of management in getting rid of underperforming stores.
  • On a cursory basis, I didn't think the stock was hugely expensive before the earnings miss. It traded at 19x ttm earnings. Post earnings miss, it got down to just under 14x ttm.

Before I bought though, I analyzed the previous 10 years' financials courtesy of my initial Graham oriented framework. I'm reproducing my analysis below. You can see how rudimentary it was. In truth, it didn't really point me towards a conclusive decision, and in retrospect, although I identified that it looked cheap, I didn't understand how cheap it was, or why it was cheap.  I hadn't yet developed a conceptual framework to help me evaluate what my theoretical margin of safety was buying post Petsmart earnings .

First, my comparison between Petsmed Express and Petsmart:

I focused on the highlighted areas in arriving at my buy decision, namely, P/E at $65 per share of 16x ttm was now below the average of the P/E over the last 3 years, Greenblatt return on capital was sufficiently high at 43% relative to return on capital between 2011 & 2013, and my Earnings yield (EBIT/EV) was sufficiently high.

Shares o/s2099
Mkt Cap2646,4800.04
Pref stock--
Total cap (EV)2647,0000.04
Book value701,0940.06
Graham Book value701,0530.07
Book value/share3.511.01
Graham Book value/share3.510.6
Net Income184200.04
EPS, 20130.94.02
Avg EPS, 2010-20120.862.7
Avg EPS, 2007-20090.981.69
Avg EPS, 2004-20060.481.21
Avg EPS, 2000-20020.10.49
P/E (ttm)14.6616.230.9
Avg P/E 2010-201215.3424.130.64
P/S (ttm)1.130.941.21
Price/Book Graham3.776.150.61
FCF Yield = 1 / EV/FCF (ttm)5.12%6.51%0.79
Div Yield5.00%1.04%4.8
Net/Book Graham26%40%0.64
Current assets/current liabilities5.91.54
Working cap/debt06.5-
EPS growth
2010 to 20132%14%0.11
2007 to 2012-3%10%-0.26
2004 to 200915%7%2.23
2000 to 201320%19%1.05
growth 10 years 2002 - 2012760%452%Must be > 33%
ROE 201325.71%38.39%0.67
Avg ROE, 2011-201324.19%32.17%0.75
EBIT 2013286930.04
Average EBIT, 2011-201329615.670.05
Greenblatt Capital, current69.21,611.00
Avg Greenblatt Capital, 2011-201370.931,680.67
ROC 201340.46%43.02%0.94
Avg ROC, 2011-201340.88%36.63%1.12
EY = EBIT: EV 201310.61%9.90%1.07
Average EBIT, 2013-200928.5654.33
Average Int x, 2013-2009-54
Interest coverage:#DIV/0!12.12

Next, my 10 year lookback analysis:

I'm impressed that I was able to come up with this without having studied some of the more detailed valuation concepts I've since studied:

DATECompound Sales GrowthCompound EBIT GrowthCompound EBITDA GrowthAvg priceCompound Share GrowthAvg mkt capAvg EVEV / EBITEV / EBITDACompound NI GrowthCompound PTI GrowthP/EEPS
Ratio of avg share growth to avg NI growth

What jumped out at me here was that at $55, Petsmart was trading at a ratio of .67 x the average net income growth over the last 10 years, discounted EV/EBIT and EV/EBITDA multiples vs. the 10 year historical average, a trailing P/E of 13.68x 2013 eps, about 500 bps below the 10 year average P/E.  All as a consequence of two days of trading during the week of May 12, 2014.

You know the adage that you don't need to swing at every pitch?  Well here was a fat pitch with a window of almost two weeks to act post earnings.  And those participants who had done the preparatory valuation work swung at this fat pitch on the basis of theoretical margin of safety at prices between $55 and $60 (in my opinion).  No technical analysis required.  No subscription to buy/sell software needed.  No listening to talking heads on CNBC or ROBTV plugging "top picks".  Just plain vanilla math making it's case in the face of valuation.

Back to the subject of risk. Conventional market wisdom would infer that post earnings, because the stock plunged 15%, volatility jumped and resulting beta jumped, the stock was more risky as a result. This is absolute rubbish!  Looking beyond option pricing models (which don't work for any period of time beyond the very near short term), there was indisputably less risk in valuing the annuity business of selling pet supplies (notwithstanding the earnings miss) at an EV of $5.9B vs. an EV of $6.9B three days earlier. I don't think I grasped this concept fully at at time.

In retrospect, I was lucky, and my timing was good.  But my luck was not conducive or equivalent to a systematic evaluation methodology which I could use on an ongoing basis to consistently generate above average returns.  It was a glimpse into what could eventually be (and what I'm striving to achieve now consistently).

Fast forward to end of June 2014, Jana Partners announced a stake.  Fast forward to fall 2014.  Jana Partners offered to acquire Petsmart for $82 per share.  I'm not going to bother getting into a discussion of how ridiculous the concept of efficient markets is here.  Suffice it to say, about the only efficient aspect of the Petsmart pre and post earnings reaction was the efficiency of the market makers' collective order books routing stop loss orders the day after the earnings release.  If markets were truly efficient, how could a business that was so detested in May 2014 at $55 per share, be revalued up to $82 on the Jana deal less than six months later?  I leave this question open to interpretation.

What follows is an analysis of Petsmart's of 10 year historical financials using my current evaluation methodology for the purpose of outlining the extreme in valuation exemplified by the gut reaction to the May earnings release.

I'm also including my evaluation methodology to empirically demonstrate what my theoretical margin of safety would have been right after the earnings miss.  I don't think I'm introducing bias into this analysis, as I actually bought Petsmart post May earnings using my earlier "simple" evaluation methodology (albeit without a deeper understanding of why I was buying as noted above).  And, lest it be disputed, Petsmart actually languished between $55 and $60 for a # of weeks post earnings, providing all participants the same opportunity to buy the stock for some time.

EBIT %7.32%7.64%8.27%7.61%7.53%7.31%6.92%7.53%8.23%9.63%10.02%8.00%
Avg EBIT00000000000
Add special charges0000000000693
EBIT before special charges21925731132235237036942950365169312.21%
Special charges as % of sales0.00%0.00%0.00%0.00%0.00%0.00%0.00%0.00%0.00%0.00%10.02%0.91%
Deduct avg special charges-69-69-69-69-69-69-69-69-69-69-69
EBIT after avg special charges15018824225328330130036043458262415.34%
SG&A as % of sales22.75%23.22%23.01%23.29%23.22%22.23%21.55%21.53%21.28%20.88%20.57%22.14%
R&D as % of sales0.00%0.00%0.00%0.00%0.00%0.00%0.00%0.00%0.00%0.00%0.00%0.00%
Add back 25% of SG&A170195216247271282288307325353356
Add back 25% of R&D00000000000
Adjusted EBIT32038345849955458258766675993497911.84%
Tax rate38.00%38.00%38.00%38.00%38.00%38.00%38.00%38.00%38.00%38.00%38.00%
EBIT AT198237284310343361364413471579607
+ Dep'n95108140157196225237236237238235
Adj EBIT AT13420125822624534848852458767969517.86%
NI per F/S12515718218525919319824029039042012.88%
$ Diff9447641-14155290284297289275
% Diff6.97%22.06%29.44%17.96%-5.52%44.53%59.43%54.20%50.56%42.59%39.59%
Cost of capital rates:EPVAdj for Debt & Excess CashAdj EPVO/S sharesEPV / shareBVPS / shareExcess EPVCurrent pricePrice / EPVPrem/Discount vs EPV
Upper bound - VC15%4,635.06-2254,410.06104.342.2810.494.03551.3030.08%
Combined NI & Div growth11.07%6,283.30-2256,058.30104.358.0910.495.54550.95-5.31%
LT equity return; 2014 - 192810.08%6,898.28-2256,673.28104.363.9810.496.10550.86-14.04%
Arbitrary 2 <10% - 12 %>10.00%6,952.59-2256,727.59104.364.5010.496.15550.85-14.73%
Per GF8.46%8,218.19-2257,993.19104.376.6410.497.31550.72-28.23%
Arbitrary 17.50%9,270.12-2259,045.12104.386.7210.498.27550.63-36.58%
Adj BetaRfRisk Premium%D%EAT Int cost

In retrospect, there were a number of factors in this case which seemed to work together to support a viable investment thesis at the time:

  • Market earnings reaction which temporarily suppressed valuation
  • As a result of the above, there were meaningful discounts to average P/E, average EV/EBIT, and average EV/EBITDA at $55 to $60
  • Theoretical margin of safety using a matrix of possible WACC's between 7.5% and 10% of between 14% and 40%
  • And lastly, potential moat inherent in the business model 

At the eventual buyout price of $82, the terminal valuations ended up as follows (using ttm Jan 2015 #'s):

  • P/E of 19.25x (vs. 13.7x pre-Jana and post earnings)
  • EV/EBITDA of 9x (vs. 6.83x pre-Jana and post earnings)
  • EV/EBIT of 11.6x (vs.9.33x pre-Jana and post earnings)
  • Pretax income multiple of 12.55x (vs. 9.17x pre-Jana and post earnings)

What I hope to accomplish in this overall analysis is an evaluation of the robustness of my current evaluation methodology in terms of evaluating theoretical margin of safety.

The end game from here is to:

  • Develop a robust evaluation methodology/model incorporating all the different aspects of my successful experiences to date (and to continue to add legs to the model to improve my methodology)
  • Run my model across as many current companies as I can (i.e., build a valuation database now of worthwhile candidates) 
  • Identify target margin of safety buy ranges in advance
  • Be prepared for all opportunities when they arrive, and when they do, make my purchase decisions mechanical in nature.  No emotion involved
  • Don't swing at every pitch.  Recognize opportunities by already being rigorously prepared in advance.

And one last comment.  None of this easy to do in real time!  Buying anything against the grain is difficult to do.  Even more difficult is keeping perspective and emotion in check, as the more price falls, the more self-doubt creeps in, combined with a questioning of the original evaluation methodology.  Which is why the evaluation methodology/model and the screening process has to be robust enough to weed out the garbage (of which there is plenty) so as not to experience permanent loss of capital.

No comments:

Post a Comment