Victor Cook

Value, tech, banks, airlines
Victor Cook
Value, tech, banks, airlines
Contributor since: 2007
Company: Tulane University
Your point is well taken. Being sure about which companies belong in any set of competitors is fraught with questions. It's likely that in Q1-2004 Mr. Gates would have laughed out loud at the idea of comparing MSFT ($295b market cap) with AAPL ($9.6b market cap) SG&A cost per dollar revenue, unlike Mr. Gerstner 's IBM (the elephant in the room) who saw DELL's low SG&A cost per dollar revenue as a challenging benchmark. One can never be sure about what comparisons are reasonable. But it may be wise never to overlook David especially if you're Goliath! Thanks for your comment.
In smart-tech companies nothing is typical, especially the revenues from R&D spend.
Exactly. But not "of course." “Best in class” depends on the companies included in the analysis. In Gerstner's “Who Says Elephants Can’t Dance?” the best in class was DELL with a SG&A cost per dollar revenue of around 32 cents. That was in 1993. IBM's cost per dollar revenue was 43 cents. You're right, 11 cents does not seem like a lot until you multiply it by the company's $62.7b in revenues. That was reason for what Gerstner called IBM’s $7 billion expense problem. By 2000 he had cut IBM’s cost per dollar to 24 cents but still fell short of DELL's 17 cents per dollar revenue.
In my October 11, 2010 article “What’s Facebook Worth?” I wrote:
“At this point it’s not clear whether Facebook will be a money machine like Google (GOOG), with a market value eight times revenue, or suffer from accelerated decrepitude like America Online (AOL), with a market cap less than revenue.”
Based on the 243,000 firms that reported (non-zero) revenues and market cap in COMPUSTAT from January 1, 1950 through December 31, 2008 I found the expected value of the distribution was 1.0. Or, in the long run, market cap equals annual revenues. Applying this expectation to a 2009 sample of 51 companies that would occupy the internet service market space with Facebook should it go public in 2012 with assumed annual revenues of $7 billion I found:
“The chance that Facebook’s 2012 market value will equal sales revenue [$7 billion] is one in two. The chance its market cap will reach $50 billion is around one in twenty.”
This prediction was neither bullish nor bearish. It was just what the data said. So what’s the point? For one thing, knowledge that the long run value/revenue ratio is 1.0 is a useful benchmark in premarket valuation of an IPO. For another, my prediction brackets approximately FBs market caps since it went public. And it presents probabilities for all the points in between.
Thanks. I'm pleased you liked it.
PF, thank you for your understanding of my analysis. BTW. I taught at the Wharton School in 1968 before moving with the Marketing Science Institute when Bob Buzzell took over leadership at HBS. Before your time I know, but common ground in any event. Victor
Good question. There could be, but I didn't run it.
Dec. 9, 2010
A footnote on “What’s Facebook Worth?”
On October 11, 2010 I calculated the chance Facebook’s market cap would reach $50 billion January 2012 was around 1 in 20. The chance it would top $56b was about 1 in 33.
On May 18, 2012 (the first day of trading) FB peaked at $45.00 a share. With 2.14b shares outstanding its market cap was $96.3b. On June 1, 2012 FB closed at $27.72 a share with a corresponding market cap of $59.3b. That’s a loss of $37b after two weeks.
What’s next? Does anyone still see $104b in Facebook's future?
James, there are two surprises in your comment:
(1) My seminar on the history of economic thought in the 19th century did not include Claude-Frédéric Bastiat and
(2) He is the first economist I've seen quoted in support of my theory of the little guy advantage!
Thank you.
You sure have a right to quibble about the title of my article. On the one hand it doesn’t fit Merriam-Webster’s first definition physics as "... a science that deals with matter and energy and their interactions." On the other, it does fit neatly into the second definition as "... the physical processes and phenomena of a particular system."
I question your conclusion that LUV’s move into Tampa presents a “… public interest in preventing any one carrier to buy up gates the way it appears LUV did in Tampa.” In my opinion the public interest should not extend into micro-regulating strategic distribution decisions like buying up gates, or contracting for shelf space, or winning more patents and copyrights, or more buying more time for ads than any other company on the Superbowl. Especially if these decisions increase the level of public service offered.
On a personal note, I’ve been flying from MSY to PBI for many years. And I noted in my article that before LUV moved into Tampa the other carriers charged too much and took too long to make the trip. Clearly, management's move into improved the level of service dramatically.
Semantics aside, thank you for your thoughtful comments.
I'll have to think about the answer to your question. That curve first appeared in a working paper I wrote at Chicago Booth in 1971. I've still got a copy of that paper in my office at school. I'll see if I can find the seed there.
Meanwhile, your conclusion "that the size of a company eventually
stands in the way of its growth" is brilliant. It suggests another take on the curve: the cost of monopoly is infinite.
Thank you!
" ... Richard A. Friedman, a longtime Goldman partner, decided the Facebook deal was not suitable for his clients, in part owing to the high valuation and to a mismatch with his investment criteria. The $450 million investment values the Web company at $50 billion. After Goldman’s deal, some industry experts cautioned that Facebook’s growth would need to accelerate rapidly over the next couple of years to justify such a steep price — a risk with many brand-name technology upstarts." See Dealbook at
Every so often I have a student, usually an undergraduate, who’s very smart, often not well prepared and quick to shoot from the hip. He tries to dominate class discussion by challenging his professor in the belief that this will impress his classmates. Because of this behavior his most frequent achievement is to divert more class time than is necessary to answering his questions. At least this last question, for its brevity and relevance, is a big improvement on your earlier ones.

First, I did not say that Goldman had “…only a 5% chance” of making money on its recent investment in Facebook. As you noted above I did estimate that “The chance its market cap will reach $50 billion is around one in twenty.” Let’s revisit this question after Facebook’s IPO.
Second, Goldman invested $450 million for a number of shares in the still private Facebook entity. From this base they reportedly will create a “special purpose vehicle” and sell off pieces to HNW clients for a total of $1.5 billion. In this way Goldman provides Facebook with cover concerning the 499 shareholder limit imposed on private companies and gives Facebook badly needed cash. Bottom line: in a matter of weeks Goldman likely will book a little over $3 for every $1 it invested in the deal. What do I think of this transaction? It’s brilliant … unless the SEC finds it violates the 499 rule and the congress supports this finding. See the “500-Investor Threshold Debated for Its 47-Year History” on Dealbook
Finally, the Goldman transaction is reported to value Facebook at $50 billion. But to my knowledge neither the number of shares Goldman bought nor the number outstanding are on the public record. This information is necessary to calculate the implied secondary market value associated of this transaction. Which, BTW is NOT its market cap.
Thanks for your comments. Here are my replies by number:
1) Z stats are calculated, not “observed” in the data set.
2) Since the VRR has a lower limit of zero, you cannot “get to …a negative market cap.”
3) The histogram of the VRR looks normal, though it’s truncated in this application.
4) The distribution of market cap looks like an extreme exponential. If I remove GOOG (value = $197b) and YHOO (value = $23.6b) the mean and sigma fall to $0.7b and $1.2b respectively. Deleting these outliers hides the extreme risk inherent in this market. Normalizing with 20 sampling distributions removes this inherent risk entirely and greatly narrows the range of Facebook’s possible valuations.
If you haven’t already done so you might want to read Nicholas Taleb’s “Fooled by Randomness:” … “Outside of textbooks and casinos, probability almost never presents itself as a mathematical problem or a brain teaser.”
Your conclusion that “…from the mean and standard deviation it is clear that the distro is not normal” is almost correct.
What is clear from a distribution with mean 2.7 and standard deviation 3.5 is that it is not standard normal. A standard normal distribution is a normal distribution with zero mean and unit variance. There are an infinite number of normal distributions, each based on as many combinations of means and standard deviations. The stanard normal is one of these.

Your comment that my “z-scores and p-values are all very wrong” is incorrect. I would not have used the Z statistic if my sample did not satisfy the 68-95-99.7% empirical rule.
Thank you for the opportunity to clear up a common misconception.
Old Hand, thanks for your comment. "... pretty much any number you want to pick -- that someone else would buy" is the rule at this stage of the game. With one extra point: the chance Facebook's value is over $42 bn is about 1 in 10.
Your comment that airline management should try “… to figure out how to be more customer friendly and attract more air travelers and revenue” cuts right to the core of the problem. If mergers don't work, what's an airline CEO to do?
While my article defines the problem, it doesn't offer a solution. For a road map on how to find a way out of this dilemma I recommend airline CEOs read J.C. Larreche's book "The Momentum Effect: How to Ignite Exceptional Growth." Here's what Sir Richard Branson, who knows a thing or two about the airline business, says on the cover:
"This book shows you how to build momentum and leave your competitors trailing in your wake."
What might airline CEOs learn from it? One example of the insight that jumps from the pages of Larreche's book is that the relationship most air carriers have with their passengers never develops beyond their online booking, ticket purchases, onboard experiences and baggage handling. As JC says on page 154:
"There is no emotional connection. To generate the momentum effect requires a much deeper and more committed relationship than that offered by passive customers who just don't complain. Companies should measure their success by the number of delighted customers they have--people so thrilled with a product or service that they can't help but tell others about it."
We often hear the idea that "less is more." But that concept always is expressed in terms of money. The following comment in JC's book (page 27) explains when less is more from the airline passenger's point of view:
"... less should mean that they get exactly what they need and nothing more, with no superfluous elements that create complexity and could destroy value."
Reading this sentence started me thinking: are there any airline services that create complexity and destroy value? Yes. Baggage handling. This conclusion led me to write a series of articles exploring ways to “… be more customer friendly and attract more air travelers and revenue.” In the last article in that series I proposed a radical solution to the baggage handling problem: remove baggage from the passenger air transport system and shift it to the cargo air transport system. If you’re interested in the pros and cons of this idea start with
Thanks for the opportunity to put this idea back on the table.
On Sep 28 06:32 AM Tack wrote:
> The fundamental problem with almost all airlines (here again, Southwest
> "gets it" the most) is that they so utterly misunderstand their own
> businesses and have developed near-suicidal marketing and pricing
> approaches that unless they wake up to new realities, most will fail,
> whether merged or unmerged.
> As an executive (now retired) in capitally-intensive, fixed-cost
> industries during my career, one thing I learned very quickly is
> that volume is everything. When fixed costs comprise a huge segment
> of a business, it's never possible to achieve a positive result by
> adopting any program that reduces volume and gross revenues. Whatever
> savings may be attached to associated variable costs will be outweighed
> by the fixed costs, now allocated over a smaller base.
> So, on the surface, a merger would seem to make sense, i.e., expand
> volume faster than the expansion of fixed costs. The problem remains,
> however, that the airlines have totally forgotten how they generate
> revenue and how to grow that revenue. Consequently, they are seeing
> fewer and fewer passengers who wish or need to partake of their "services"
> (I use that term euphemistically).
> Think about it for a minute. Is anybody attracted to being humiliated
> at airport check-ins? to being financially punished for not "planning"
> a trip in advance? to getting penalized for changing or canceling
> a trip? or how about paying absurd fees to bring along your luggage?
> or paying 2-3 times as much to go 500 miles as 5000 miles?
> This is a mentality associated with arrogance and would work more
> "perfectly" in mandated-as-compulsory products, like health or auto
> insurance, but, it's insanity in a business where people make discretionary
> decisions. And, that's what the airline business has increasingly
> become, a business where the need and/or desire to use the product
> has become optional.
> Now, businesses have access to worldwide, instantaneous Internet
> communications and teleconferencing. The absolute need to send somebody
> at exorbitant cost to a meeting is greatly reduced. The greater competition
> in almost all world business segments also dictates that companies
> manage their finances, not just pay whatever the airlines wish to
> charge, as if there were no alternatives.
> And, consumers are finding that they can enjoy life just fine without
> flying around the world, or even around the country. It used to be
> that a trip was something exciting to look forward to, a special
> event. Now, the very thought of getting one's pockets picked and
> treated like some sucker to be exploited makes travel unappealing,
> and that's before we even get to the hassle, disrespect and humiliation
> that passes for "airport security." (I think the shoe bomber did
> more to damage the American economy that any other terrorist act,
> but that's another thread.)
> Airline executives seem utterly oblivious to the changed rules of
> the game of air travel and its increasingly optional nature. At a
> time when they should be turning themselves into pretzels trying
> to figure out how to be more customer friendly and attract more air
> travelers and revenue (not by nickle-and-dime fees, either), they
> are engaged in a nihilist race with each other to see who can drive
> passengers away the fastest. It's insanity, no less.
> For a very long time, I've thought there would be a huge opportunity
> for an airline to come into being, whose operating plan included
> no fees or penalties of any kind and had a simple revenue plan: all
> seats would be priced on a base fee to cover fixed business costs,
> plus a mileage add-on to cover the distance traveled. Ergo, the price
> all of travel would be proportional to the distance one goes. <br/>
> Such a plan is too simple, of course, for the convoluted thinking
> of the airline executives, although Southwest probably comes the
> closest. But, for the others --merged or not-- who keep seeing travelers
> as cattle marching inexorably up the chutes of the slaughterhouse,
> their days are increasing numbered.
Fat Panda,
You're right. I am comparing apples and oranges; and different markets; and different levels of competition; and different fleets; and different fuel hedging bets; and different cultures; and so on. That’s the whole point.

I designed the risk-adjusted differential to standardize the outcomes of these and many other strategic choices in order to inform investors about managements’ overall value-revenue orientation. If these differences didn’t exist the risk-adjusted differential would be zero for all companies all the time!
Thanks for giving me the opportunity to clarify this important issue.
Your conclusion would hold water except for one thing: Southwest has held the same lofty position on the value creation scale since Q1 ’93. That makes it 37 years of value leadership, most of the time without the need for fuel hedges.
You can verify this result in the 18 minute audio slide show “Y’all Buckle That Seat Belt” based on Chapter 2 of my book “Competing for Customers and Capital. Here’s a link to that presentation:
Thanks for the opportunity to clear up this commonly held misconception.
Change is the Only Constant,
Better strategic alternatives do exist.
For example, see my Seeking Alpha article “General Motors’ Natural Share Level: Can GM Be Like IBM?” (December 15, 2008).

A company’s “natural share level” occurs when the incremental cost of the next revenue share point equals the incremental earnings from acquiring that share point. In my book -- “Competing for Customers and Capital” -- I ran these numbers for Southwest Airlines in a similar peer group for Q1’03 on page 126. LUV’s actual share level was 7.6%. That share level produced an actual EBITDA of $139 million. At that time LUV’s “natural share level” was 13.6% with a theoretical maximum EBITDA of $176 million.
By Q2’09 LUV’s actual share level had increased to 10.0%. I haven’t had reason to run the company’s natural share level or maximum EBITDA for the most recent quarter. I guess you’ve given me one!
If you’re interested in how “natural” or “maximum earnings” market share is determined check out my 14 minute audio slide show on “The Rule of Maximum Earnings” from chapter 5 of my book. Here’s a direct link:

Thanks for your thoughtful comment.
User 457816,
You overlooked the last sentence of my article:
“If you want to understand the details behind this analysis, review my 18 minute audio slide show Y’all Buckle That Seat Belt.”
Clicking on the hyperlink at the end of this sentence you will find these results are based on a set of objective rules using audited financial data. You also will find the results are basically unchanged since Q1’93 for a comparable peer group of U.S. airlines.
Thanks for the heads-up.
Thanks for the congrats. Wish I'd known Jenna was just a cardboard cut-out during my interview so I could of been in on the joke!
Best regards,
Yes, of course I remember your comment. Unfortunately, I did not follow up on ways to combine Wachovia’s pre-merger numbers with Wells Fargo’s. I wish I had done as you suggested.
It would have been a simple thing to combine the two if WB were included in my 32 quarter analysis of the 92 public US bank and thrift holding companies with the most deposits. But it wasn’t. That’s because WB was not on the list I got from American Banker. Apparently their data supplier (SNL) removed WB since the merger had gone through when they compiled the list on March 11, 2009. So, rather than redo the analysis I let it ride.
I just checked WB’s 9/30/08 financials. At that time its price had dropped to around $3.50 destroying its market cap, while the merger would have doubled WFC revenues. The combination would have pushed Wells Fargo’s AQ score back far below zero. Then I wouldn’t have been nailed by Connell in my interview! Sigh…
VP of Common Sense,
Now that you bring it up, I think it's likely a combination of bank transparency and investor perspicacity. The largest possible positive gap is created between share of market cap and share of bank revenues when investors are confident in their judgments and banks are forthcoming in their reports.
Good insight! Thank you.
Your comments highlight several relevant issues. I will try to address them one at a time.
“The distortion of value and the relatively small importance of market cap. as a representative of quality invalidate this type of comparison.”
By “the distortion of value” I assume you mean transforming the dollar value of capitalization into the market share of capitalization. If so, this does not “distort” value it just creates a linear transform of it. The correlation between a bank’s dollar value and share of value over the 32 quarters is exactly one. Also, market cap and bank revenue have equal importance in this representation of asset quality.
“It would seem that problems of scale would give the very large banks, (overall), a skewed result.”
Scale has no effect on a bank’s AQ score for two reasons. First, the unadjusted AQ score is calculated by subtracting share of revenue from share of market cap. This eliminates the scale effect: 25 minus 23 returns the same unadjusted AQ as does 5 minus 3. Second, these differences are normalized through division by the standard deviation of each bank’s 32 quarterly differentials. A bank's 32 quarter series of AQ scores averages zero with a standard deviation of one. So, there is no skew.
“Also, due to the illogical volatility in the banking industry in the last six months I doubt that this ranking will be accurate as to how the real banks read on the real ‘stress test’ list.”
Whether the recent volatility in the banking industry is illogical or not has nothing to do with a bank’s AQ score. The combination of transforming the dollar data to market shares and normalizing the differences has the effect of leveling the playing field. Whether the AQ will be more or less accurate then the real stress test is an interesting question. I share your doubt on this issue. But remember, these scores are based on real data for 92 real institutions over 32 quarters. Management has very little wiggle room on two of the three data inputs (number of shares outstanding and revenues) and no influence on the third (closing stock price).
One of my objectives in designing the AQ index is to “wash out” the effects of politics and manipulation. I don’t know if the AQ will succeed in this regard. But I do wish I had bet on that 50 to 1 long-shot in the derby!
Thank you for the opportunity of address these important issues.
In your comment you concluded:
“The model relies on ‘efficient market’ theory, which makes it hopelessly unreliable. You are assuming that investors, collectively, know the true value of these banks' assets. If that were true, we wouldn't have a problem to begin with.”
I would like to clarify these issues.
First, the AQ score couples the risk-adjusted effects of investor decisions (on a bank’s market share of capitalization) with those of management decisions (on market share of revenues) over the long term. No where in efficient market theory will you find mention of either company revenue or market share. It is precisely because markets are inefficient that I developed this round-about model of assessing a bank’s asset quality.
Second, please review the previous post in which I track the AQ index over the 36 quarters ending in 2008 for Goldman Sachs, Morgan Stanley and JP Morgan. The risky asset problems sleeping in the balance sheets of these three firms were reliably revealed by their AQ index years before they burst into the headlines. Investors clearly didn’t know about this problem.
By the way I did not include GS and MS in this analysis of 92 firms because they were not then included as bank holding companies in the American Banker's Q3-2008 list ... and my previous post focused on them anyway.
Thank you for the opportunity to clarify these issues.
You said:
"If financial shares get drawn down by a general market downdraft, the banks measurements will decline, perhaps significantly, independent of any actual change in their operations."
Since the critical inputs to the AQ are a bank’s share of value and share of revenue, rather than dollar values, there will be a change in a bank’s AQ score as a result of market downdraft only under two conditions: either the bank’s price runs counter to the market or there is a significant change in the long-run volatility of the difference between its share of value and share of revenue.
Thank you for raising this important question. It’s one that may concern other readers.
There is an unexpected alternative to the stress tests. See my August 12, 2009 SA article "Banks: Forget the Stress Tests, Use the Asset Quality Index." Vic
Sorry, the SA commenting software seems to misinterpret the equal sign as I entered it above. The sum is $12.772b.
No, I did not include Wachovia in the analysis. Since you plan to run an analysis on that bank combined with the eight included in my article you should know I’m using S&P Compustat data standardized for direct comparability among companies.
Also, I cross-checked the S&P numbers with EdgarOnline I*Metrix data -- As Reported. In the case of banks both services define Total Revenue as Net Interest Income + Total Interest Expense + Total Non-interest Income. For example, WFC’s reported Total Revenue on 9/30/08 was $6.381+$2.393+$3.998=$... Note that Yahoo Finance reports revenue net of interest expense for some banks. In effect, Yahoo removes what amounts to a bank’s "cost of goods sold" from reported revenue. I’ll be interested to see what you come up with. If you have any questions please feel free to contact me by email any time. Thanks for your interest.