Friday, October 10, 2014

The Halo Effect

After posting about the Collins book, Good to Great, some people mentioned the book The Halo Effect: . . . and the Eight Other Business Delusions That Deceive Managers by Phil Rosenzweig as a counter to it.

So of course, I got it and read it right away.  And it's a really good book.  I agree with a lot of what Rosenzweig says and this is one of the reasons why I generally don't read management "how-to" books.  If I do read them, I read them quickly and don't think much of them either way because, frankly, I have never really been all that interested in management.  I have never wanted to be a manager or CEO or anything like that at all.  Mostly, I read it to get corporate histories especially if they include case studies.  Ironically, I actually read them precisely for the "stories".  How to improve a business or build a great corporation, to me, is secondary and not all that important.  It's the stories that I want to read.  And it's this "story-telling" that Rosenzweig criticizes.  He says that many of these management books (like the Collins books) are not scientific studies but are just a bunch of great stories.  That's actually totally fine with me.

In any case, I enjoyed reading Good to Great, and think it was useful.

Amazon/Bezos Book
I also just finished reading The Everything Store: Jeff Bezos and the Age of Amazon.  This is a really good read and I think everyone should read it, even people who don't invest in internet companies.  The way Amazon is changing the world goes way beyond internet businesses so I think it's an important piece of the puzzle for anyone who wants to understand business and what's going on in the world now.

This is a really well-researched book and one of the better business profile books.  There are a lot of books that seem to have been thrown together over a single weekend with sources/information seemingly acquired in a bar nearby a company headquarters.  When Apple got hot, a few books came out that were like this; very little new or interesting information and anonymous sources spewing gossip (as if the author hung out in a bar for a few nights and talked to some lower level employees), but no real, substantive information.  (The Walter Isaacson book on Steve Jobs was really good, though).

This one is not like that at all.

One thing that is scary is that it seems like Bezos doesn't even want to raise prices at all.  Some people expected Amazon to lower prices and put everyone out of business and then later raise prices to make a ton of money.  But that seems not to be the case.  At one point in Amazon's history, they were actually contemplating raising prices.  But Bezos' model for Amazon was Sam Walton.  He really dug into his book, Sam Walton: Made in America.  Around the time Amazon was thinking about raising prices, Bezos met with Jim Sinegal of Costco and heard that they won't raise prices to make more money.   Apparently, after this meeting, he resolved never to raise prices for the sake of making money.    I did scratch my head at this part in the book thinking that Costco actually makes money and has a decent return on capital; why won't Amazon raise prices enough to at least make a little bit of money?

Bezos / Good to Great
Anyway, I went off on this tangent for a reason.  As I was reading the Bezos book, I was surprised that Bezos was deeply into Jim Collins and even had him come to Amazon to show them how Amazon can become a great company.

Apparently, a light went on at Amazon due to Collins' presentation (which was based on the then yet-to-be-released book, Good to Great).   The part that clicked with them was finding out what they were really good at and then developing it and get the flywheel moving.

Back to the book.

So what is the Halo Effect?

The Halo Effect
There is a good summary of the book on Wikipedia: Wiki: Halo Effect.
  1. The Halo Effect of the book's title refers to the cognitive bias in which the perception of one quality is contaminated by a more readily available quality (for example good-looking people being rated as more intelligent).[6] In the context of business, observers think they are making judgements of a company's customer-focus, quality of leadership or other virtues, but their judgement is contaminated by indicators of company performance such as share price or profitability. Correlations of, for example, customer-focus with business success then become meaningless, because success was the basis for the measure of customer focus.
There are other delusions (nine in total) that are summarized on the Wiki page.

I agree for the most part with what the book says about these cognitive biases and delusions, but I wouldn't go as far as to say that these biases make Collin's book and many business press articles worthless.  Rosenzweig doesn't go that far, but comes pretty close.

Any time someone says they want to investigate why some companies succeed and why others fail, I am going to be interested.  The methodology may not be perfect and there may be flaws in the assumptions.  But that doesn't really matter so much as just reading about what people did over the years; what worked and what hasn't can be interesting on it's own and you can make your own adjustments about the respective biases.  Rosenzweig uses Cisco as an example of how everyone thought it was such a wonderful company during the internet bubble (great customer service) and then the very same reporters talked about how lousy Cisco is (with horrible customer service) after the bubble popped  (I think most of us, though, recognized Cisco as a huge beneficiary of the internet bubble in a virtuous circle and had no doubt it would eventually reverse).

If you had access to some of the top CEO's in the country and you were able to have lunch or dinner with them, wouldn't you do it?  Of course you would.  And of course you would learn something from doing that a lot.

And of course you would be subject to all of the same biases discussed in Rosenzweig's book. You would be talking to people you see as successful, and they will tell you what they have done, but there is no way to tell what the cause and effect was (were employees happy because the company was successful and the stock prices (and therefore their 401-k's and stock options) were up a lot, or did happy employees create the success?

But you would still do it.   And you would still ask the same sort of questions.  And then you would make adjustments to what you hear  (and even those adjustments would reflect your own biases).

I guess what Rosenzweig has a problem with is how Collins simplifies and distills all of the case studies into a simple formula, but I think most readers read these books to get ideas and inspiration more than any sure-fire, fool-proof method of eternal success.

Managements Don't Matter / Nothing is Forever 
Rosenzweig also presents a study that shows that CEO's don't really matter as much as we think.  This may be true, but taking that to the extreme is not so great either because I don't think anyone would agree that the CEO doesn't matter.  I do think there are good CEO's and bad ones.  But that is not to say that good CEO's can't have bad results and vice versa.   He also points out that it is impossible to build a great company to stay great forever.  You can't build something to last.  He shows that previous excellent companies, companies built to last and good to great companies have done worse than the S&P 500 index, and the less favorable comps did better.  All companies go through a life cycle.  I agree with that too.

I personally believe more in CEO's than corporate cultures.  I would rather invest in a CEO I truly believe in (Warren Buffett, Jamie Dimon etc.) than a company known to have a strong culture of success (even though Goldman Sachs is one that I believe in strongly).

As an investor, I am highly skeptical of corporate cultures in general, unless it has been proven over multiple generations (like Goldman Sachs).  The fact that firms like Starbucks, Uniqlo (Fast Retailing), Dell and others floundered after their founders left (and they had to come back) makes me wonder about the importance of cultures versus CEO's.


Efficient Markets
Anyway, all this talk of CEO's not mattering too much and nothing can be built to last reminds me of the efficient market theory.  In the pure form of efficient market theory, the markets are too efficient for anyone to outperform.

That would make books like Intelligent Investor and Securities Analysis meaningless.  And all those Superinvestors (of Graham and Doddsville), Money Masters, the New Money Masters and the managers/traders in the various Market Wizards books by Jack Schwager would all be monkeys that just happened to flip a few heads in a row.

I haven't checked, but I tend to think that most of the managers covered in these books have done well over time, even after being featured in them (OK, maybe I said that about Good to Great and Rosenzweig has proven otherwise).  Sure, Tiger, Steinhardt and others closed shop after a bad year or two, but they closed out their careers with incredible track records.  I think there might have been some blowups in the commodities trading world (Richard Dennis?), but most people in the above books have gone on with great records (although increased size have lowered returns for many of them; the more common lesson in outperforming managers is that size kills performance).

Markets are Efficient Because Mutual Funds Don't Outperform?
One thing people keep pointing out is that markets are efficient because most mutual funds don't outperform.  This is something that I've been thinking about on and off for years.  Why do some people outperform while others don't?

Most mutual fund companies are asset gatherers.  Their primary goal is to increase earnings of the management company.   And we all know that increasing asset size makes it harder to outperform the market.

So at the asset management company level, they are incentivized to increase AUM.   They are paid a percentage of AUM so this makes sense.  If you want to increase earnings, you have to increase AUM.  It doesn't matter if it gets harder to manage as they don't get paid on performance.

At the fund manager level, managers don't want to diverge too much from the index.   If they stay more or less within a reasonable range of the index, they will be fine and won't get fired.  The risk/return is asymmetric in this case if you want to take big risk to outperform (OK to fail conventionally etc.).

So the asset management industry is often not even trying to outperform.  They are trying to maximize earnings to the management company.

Missing the Trees for the Forest
Anyway, I brought up this efficient market argument because the "CEO's don't matter" argument sort of reminds me of it.  When you look at the data overall, you might find that CEO's actually don't matter, just as if you looked at the performance of mutual fund managers (or research analyst buy/sell recommendations), you would think the market is efficient and fund managers don't matter.

Oftentimes, the big figure can obscure the little facts.

For example, back in 2000 when the stock market was trading at 30x p/e (or whatever it was), people concluded that stocks were not good investments.  That's sort of missing the trees for the forest because BRK and many other value stocks were trading at attractive levels.

CALPERs recently decided not to invest in hedge funds.   This may be the right choice for them; it seems like they couldn't allocate a meaningful amount to hedge funds and the cost of maintaining such a small percentage of the portfolio just didn't make sense.  That's totally reasonable.

On the other hand, I always hear figures about some hedge fund index and how that hasn't done too well.  There are a lot of hedge funds out there and most of them are probably no good.  But there are really good ones.  The problem with hedge funds, usually, is that the good ones are closed.  So again, I wouldn't look at the hedge fund index to decide whether hedge funds are good or not.  I would just look at funds individually and if there is a good one, invest in that and otherwise don't.    A lot of information is lost when you look at the various indices.

Likewise, if a study shows that the CEO doesn't really matter too much, that doesn't mean that much to me.  If you look at a big enough group, then bad industries or bad cycles can easily offset good CEO's and the sum of everything might make it look like CEO's don't matter.

Conclusion
Well, I don't know that there is a conclusion to this post.  I think the Halo book is a good one and very interesting.  The arguments make sense. But on the other hand, there is a bit of an "efficient market" sort of thing going on there too.

I like "stories" and I don't think there is anything particularly wrong with reading them as long as we understand the context and limitations of those stories.  I know I will keep reading articles in the business press, business books (not "how-to" books, but books about businesses and biographies of business people).