10 Years, 100 Analysts, and 2,000

Stocks: Learning from Experience

Pat Dorsey, CFA

Founder and Portfolio Manager,
Dorsey Asset Management
Introductions
o Pat Dorsey, CFA
o Founder, Dorsey Asset Management
o Former Director of Equity Research at Morningstar: Created
investment philosophy, built team from 20 to 100 analysts,
developed institutional research platform
o Author of The Five Rules for Successful Stock Investing,
and The Little Book that Builds Wealth

o Dorsey Asset Management

o Single strategy, separately managed accounts
o Long-only, all-cap, global mandate
o Concentrated in 10-15 stocks

2
Agenda
o Lessons
o 100 Analysts + 2,000 Stocks + 10 Years = many
successes, many mistakes and many lessons
o Consistent Framework
o Wide range of companies
o Wide range of market environments

3
Process Matters
o “We enjoy the process far more than the
proceeds.” – Warren Buffett
o Easy for him to say!
o Of course, better process  more proceeds

o Think kaizen (continuous improvement)

o Standardize a process, then measure the results
o Compare measurements
o Innovate to improve results
o Standardize new, improved process
o Rinse and repeat

4
Continuous Learning
o Great investors take time to reflect
o How can I get better?
o Can I refine something I do well?
o Can I improve an area in which I’m weak?

o A good investment process evolves

o Buffett B.C. (before Charlie) vs. Buffett now
o “Nothing has served me better in my long life than
continuous learning.” (Munger)

5
Seven Lessons
o Mr. Market has Myopia
o No Comps = Good Value
o Operating Leverage is Mispriced
o Numbers Can Lie
o Disruption Creates Value Traps, Not Values
o Use the “Too Hard Bucket” Early and Often
o Moats Are Often Mispriced

6
My Mental Picture of Mr. Market

7
#1: Mr. Market has Myopia
o The Street is abysmal at incorporating
longer-term information into market prices.
o Reasons are manifold
o Average mutual fund has ~ 120% turnover
o Annual performance incentives
o Increased influence of hedge funds on sell-side
research revenue streams
o Absent a paradigm shift in the structure of
money management, this won’t change.

8
#1: Mr. Market has Myopia
o A few examples
o KLA-Tencor
o Ford
o Tiffany
o Beckman Coulter

o Takeaway: Focusing on the next few years,

(vs. the next few quarters) pays off.
o Time-horizon arbitrage is not just a fancy phrase.
o A view on cash flow two years out is a variant view.

9
#1: Mr. Market has Myopia

10
#2: No Comps = Good Value
o The more unique a business is, the more
likely it is to be mispriced.
o Chicago Merc Exchange: 1st public exchange
o 13x trailing earnings @ time of IPO (11x fwd)
o Mastercard: 1st public credit card network
o 17x trailing earnings @ time of IPO (6x fwd)
o Google: 1st public…whatever it is.
o 75x trailing / 22x forward
o Moody’s: Only public bond-rating agency
o 23x trailing / 17x forward

*This is not a recommendation to buy or sell a security.

11 Past performance is not an indicative of future results.
I Love Lazy Analysts
o The sell-side is paid to think inside the box

12
#2: No Comps = Good Value
o Takeaway:Look for businesses that don’t
have good analogues.
o When you find them, have greater confidence in your
variant perception.
o Odds are that you are one of the few to have actually
bothered to look beneath the hood.
o Odds are also that the results will best expectations.

13
#3: Operating Leverage is Mispriced
o Anchoring
o Hard not to be biased by past margins

o Linear thinking
o We don’t naturally think in terms of step changes or
power laws

o Social proof
o What’s the payoff in showing your boss (or your clients)
projections that look “nuts”?

14
#3: Operating Leverage is Mispriced
o Company “A” went public in 2006. Pretty hard to
make an earnings / cash flow forecast based on 30%
operating margin when the company had never
printed better than 13%.

Pro-Forma
*This is not a recommendation to buy or sell a security.
15 Past performance is not an indicative of future results.
#3: Operating Leverage is Mispriced
o A few examples:
o SuperMedia / Dex / Newspapers
o Chico’s
o CME
o Ford
o Mastercard

16
#3: Operating Leverage is Mispriced
o Takeaways:

o Forecast in dollars, not percentages

o Removes some of the anchoring bias, keeps you
focused on what matters: Fixed vs. variable costs.

o Don’t fear projections that look “silly.”

o How many analysts forecasted Mastercard’s operating
margins increasing from 13% to 44% in three years?
Or forecasted the New York Times’ margins shrinking
from 16% to zero in two years?
*This is not a recommendation to buy or sell a security.
17 Past performance is not an indicative of future results.
#4: Numbers Can Lie
o The Excel Fallacy
o Just because you can chart something doesn’t mean
it’s necessarily useful for forecasting.

o Inflectionpoints often show up in

scuttlebutt before they show up in data.
o Housing Bubble
o Cost-cutting during the Great Recession: “We’ve done
five years of restructuring in five months.”
o Dell

18
#4: Numbers Can Lie
o Takeaway: Forward-looking anecdotal
evidence may trump backward-looking
quantitative evidence.
o When everyone points to history to justify a thesis,
trouble may be brewing
o “All of the data is in the past, but all of the value is in
the future.” (Bill Miller)
o Caveat:Sometimes “the math is the math
is the math.”

19
#5: Disruption Creates Value Traps
o Ifa business is being disrupted, it’s a value
trap, not a value.
o “Disruption and commoditization are the long story arc
of technology.”

Dell Avid
Nokia Best Buy
Eastman Kodak Newspapers
Garmin Encyclopedias

20
#5: Disruption Creates Value Traps
o Current example: Research in Motion
RIMM
o 9x earnings and an 11% cash return…
o Whatta deal!
o Or maybe not…
o Consumer push has flopped (Storm? Torch?)
o Multiple OS platforms are a pain for developers. Why
develop for BB OS and QNX when you can just write for
iOS and have a bigger market?
o Shipped tablet without an email app.
o Company is trying to do too much, rather than focusing on
core strengths.

21
#5: Disruption Creates Value Traps
o Classic
“regression to the mean” investing
may not work if the mean is shifting.

o Technologicalchange + competitive
evolution create non-stationary data series

o Takeaway: Be careful when a business in a

changing industry is “statistically cheap”.

22
#6: Use the “Too Hard” Bucket
o You can’t reliably forecast commodity
prices.

*This is not a recommendation to buy or sell a security.

23 Past performance is not an indicative of future results.
#6: Use the “Too Hard” Bucket
o Political / regulatory risk isn’t much easier
o True, the initial nasty regulatory proposal is
often watered down by well-heeled lobbyists.
o But sometimes messianic zeal (or ignorance)
trumps the politico’s need for campaign cash.

o What’s your edge in reliably forecasting the

outcome of a regulatory / political
decision?

24
#6: Use the “Too Hard” Bucket
o Political risk is worse than legal risk.
Juries may be even less rational than
politicians, but at least their decisions can
be appealed.
o For-profit education
o Sallie Mae
o Tier-1 capital ratios
o Interchange fees for debit cards

25
#7: Moats are Mispriced
o Competitive advantage is not part of the
modal investor’s mental framework.
o Recency bias looms large.
o “Spreads” on time-horizon arbitrage are material.
o Inefficiencies abound, even in larger cap stocks.

o Fuller’s three sources of alpha

o Informational, analytical, and behavioral
o Thinking about moats gets you two out of the
three

26
#7: Moats are Mispriced
o Competitive advantage is not part of the
modal investor’s mental framework.
o Recency bias looms large.
o “Spreads” on time-horizon arbitrage are material.
o Inefficiencies abound, even in larger cap stocks.

o Fuller’s three sources of alpha

o Informational, analytical, and behavioral
o Thinking about moats gets you two out of the
three

27
#7: Moats are Mispriced
o Competitive advantage is not part of the
modal investor’s mental framework.
o Recency bias looms large.
o “Spreads” on time-horizon arbitrage are material.
o Inefficiencies abound, even in larger cap stocks.

o Fuller’s three sources of alpha

o Informational, analytical, and behavioral
o Thinking about moats gets you two out of the
three

28
An Idea
o Thesis centers on AMD’s “Bulldozer” server
chip, shipping in mid-2011
o More cores per server = more efficient usage of
physical space and power consumption, vs Intel
o Should be very attractive to virtualized
datacenters
o However, notebook and desktop chips have
potential as well
o ATI acquisition puts GPU and CPU on the same chip
o Much better metrics than Intel’s Atom

29
An Idea
o Advanced Micro Devices (AMD)
o Huh?
o No moat, but low expectations
o Less capital intensive after foundry spinoff
o New chip architecture gives opportunity to
capture market share from Intel, which would
drive sales and substantial margin expansion
o Intel has 95% share in servers – little left for AMD to
lose
o New AMD chips should have an advantage in
datacenters
o Gaining 10 percentage points of server share would
double AMD’s operating income @ 65% gross
margin
30
Thanks
o Questions?

o Contact info: [email protected]

o 312-233-2544

31
Disclosures
This presentation is furnished on a confidential basis to the recipient for informational purposes only and does not
constitute investment advice. This presentation does not constitute an offer to sell, or a solicitation of an offer to buy,
any interest in any investment vehicle. Any securities mentioned are provided as examples and are not
recommendations to buy or sell.

Dorsey Asset Management, LLC (“Dorsey” or the “firm”) does not accept any responsibility or liability arising from the
use of this presentation. No representation or warranty, express or implied, is being given or made that the information
presented herein is accurate, current or complete, and such information is at all times subject to change without
notice. This presentation may not be copied, reproduced or distributed without prior written consent of Dorsey. By
accepting this presentation, you acknowledge that all of the information contained in this presentation shall be kept
strictly confidential by you.

This document contains information about Dorsey’s strategy and investment philosophy. It includes statements that
are based upon current assumptions, beliefs and expectations of Dorsey. Forward-looking statements are speculative
in nature, and it can be expected that some or all of the assumptions or beliefs underlying the forward-looking
statements will not materialize or will vary significantly from actual results or outcomes.

Dorsey reserves the right to modify its current investment strategies and techniques based on changing market
dynamics or client needs. There is no assurance that any securities, sectors, or industries discussed herein will be
included or excluded from an account’s portfolio. Investing involves the risk of loss of principal.

Dorsey is a registered investment advisor. Registration does not imply a certain level of skills or training. More
information about the firm, including its investment strategies and objectives, can be found in our ADV Part 2, which is
available, without charge, upon request. Our Form ADV contains information regarding Dorsey’s business practices
and the backgrounds of our key personnel. DAM 16-18

32