Investing gap between theory and practice

Google the phrase "gap between theory and practice" and you'll end up with dozens of links to scholarly articles attempting to explain why things in practice don't always match an academic theoretical version.  I find it somewhat ironic that academics are attempting to explain why practitioners aren't following their theories with more theories.  I've been confronted with the gap between academic finance theory and the actual practice of investing recently.

A month or two back I purchased a book on bank valuation.  There are a lot of dauntingly large books that with excruciating verbosity explain one small slice of the banking industry.  While there are a lot of books focused on a banking specialty there are not many general books, so I was excited to find one that I thought would be more general.  I was further encouraged when in the introduction the author stated that they intended to lay out a simple comprehensive framework for bank valuation.  My enthusiasm ended when within the first few pages I was confronted with a mathematical proofs showing that a bank with a higher ROE should be worth more than one with a lower ROE.  The proofs didn't end there, the book is anything but simple, and is extremely academic in nature.

I've taken a break from the book for now when I came across the chapter detailing how the author believes bankers make decisions.  The author laid out a very complicated formula for determining if a given loan or deposit would create value for a bank.  It included modeling out a number of future values for everything from rates to risk over the duration of both asset and liability side.  The formula took up most of a page.  I can assure you that most bankers do not use this formula to determine whether an action should be taken or not.  If so the models are faulty at best because no one knows what the bank will be paying on deposits ten years in the future.

I appreciate the input of academics on issues, they look at a problem in a very different and systematic way.  But I've derived the most value from practitioners.  One of the greatest books about value investing, Security Analysis was written by an investing practitioner.  If one were to take a look at the 'best' recommended investment books almost all of them were written by professionals detailing their experiences.  This is valuable because what experience provides an insight to and math misses is the human element.

Why do stores sometimes sell products for below cost?  They do it because they know that people can't resist a good deal, and often enough a good deal gets people in the door to buy more higher priced items.  I'm sure there's a paper out there discussing loss leaders, but selling something for less than cost is not rational, and no academic paper would tell you to do so, unless the paper were explaining why something that works in the real world works on paper.

It seems to be the difference between the way academics see the world verses how practitioners see the world is one of perspective.  Practitioners are forward looking, the academic is working to explain the past.  They want to know why something happen whereas a practitioner wants to know the future.  An academic might apply research on a historical event and attempt to apply it to the future, but it's always the past that's directing their outlook.

The two approaches are so fundamentally different I'm not sure they can be reconciled.  What academic studies miss is the human element, you can't model behavior easily.

Many value investors consider technical analysis as the voodoo science of investing.  It's purported to be this mystical system of trading not grounded in anything but professes to make millionaires out of chart squiggle readers.  I will admit it seems somewhat spacey, but doesn't fundamental investing seem crazy as well?  There are these imaginary values, intrinsic values that fundamental investors claim a business is worth.  Yet companies never trade at these values, and by some mystical mechanism market value will eventually reach intrinsic value and investors will know to sell.  Further these intrinsic values are supposedly easy to calculate, but everyone except the buyer is wrong on what they are.

Often reality is different.  Investors, both value investors or traders purchase a stock because they believe it's either undervalued or some market action/reaction will take place.  In many cases investors with years of experience have a sense about these this and "know" that something will happen.  Sure enough the stock reacts, either with the market, or due to a one-off event, shares run up and the investor sells taking a profit.  All the while patting themselves on the back because they were right about what happened.

The problem is no one knows exactly why stocks appreciate or decline.  We tell ourselves narratives about why something happened, but many times no one truly knows.  There are so many factors and participants at work in a single company's shares that it's impossible to determine what or who caused a price change.

I was at an event recently and was talking to someone who manages a multi-billion dollar mutual fund.  He clearly represents the practitioner of this story.  He could construct a brilliant narrative about almost anything economic happening in the world.  Want to know why Japan's debt was increasing but they didn't have inflation?  He had an answer that sounded great.  He was very skilled at constructing narratives.  But what I found interesting is that didn't seem to drive how he invested.  We discussed the types of stocks I purchase, to which he conceded that they will most likely clobber the market, but without professional participation because they're too small.  He said at the end of the day he's simply a handicapper, not much different from someone at a casino.  He looks at a company and assesses if the likelihood of something good happening is higher than the likelihood of something bad happening.  He buys companies with stable balance sheets at low multiples of cash flow, with the theory that it's unlikely these companies will go out of business, but it's likely that eventually the market will give them a higher valuation.  It seems simple and it is, and he's paid very well to do this.

The idea of handicapping stocks can be extended much further, and at a certain level is the essence of investing.  I prefer to buy extremely cheap stocks with strong balance sheets.  Many of these companies are smaller, but that's only because of what's available.  But what I'm doing is finding companies where the likelihood of something bad happening is small, whereas they are so cheap that I'm betting something better will eventually take place.  I'm not always right, but one doesn't have to be correct 100% of the time.

Investing for many is no different than trawling garage sales for collectables and antiques.  Some garage sale buyers purchase large quantities of small items that they unload at flea markets for a tiny profit.  These are the traders of the garage sale world.  Others will visit sale after sale looking for an extremely mis-priced antique.  When one finds a mis-priced antique they usually don't know what it's blue book value is, but they know that something rare shouldn't be selling for $35, or a similarly low value.  These antique buyers are the value investors of the garage sale world.  There are other garage salers who are hunting for a lost Monet that they plan on hanging in the living room and cherishing forever, these are the Buffett investors.  The academic of the garage sale world is the auctioneer who is trying to piece together the garage sale action from a glass tower in Boston.  They see the sales prices and see the inventory and are trying to construct a narrative about the market.  Even though the auctioneer has all of the market's data it's truly the participants who know what's going on.  They have the 'feel' of the market whereas the auctioneer only knows the results.

The problem is the motivation of a buyer (garage saler or investor) is never clear, it's a human nature problem.  Traders identify patterns in charts that signal a market move.  Technical analysis feeds off itself, if enough traders participate then a signal becomes 'true' because enough people act on it.  Trading rewards those with superior tools and information who can act quicker.  Value investing is for slow traders.  But like in trading value investing incorporates a heavy notion of human judgement.  I have purchase companies that I knew were cheap and I had a sense something might happen.  It was from how management wrote their letters, and how they spoke of the company.  My sense was developed with experience, but it isn't something that can be modeled.

Both parties can benefit from each other.  Academics need to recognize that the market is fueled by human behavior.  Sometimes investors do things that are completely non-rational in an effort to impress someone, to gain a promotion, or just because.  These things don't fit nicely into models.  But practitioners don't get off easily here either.  They need to recognize the limits of models and studies instead of blindly following them.  If one looks hard enough any relationship can be forged out of stats.  It's relationships built on reason that have the highest chance of being true.  There are some investors who are such slaves to the models that they've failed to recognize the world isn't populated by computers.

The one thing I know is that successful investing requires a practitioner mindset.  You will never find the next best investment studying what Buffett was buying in 1952.  Even Buffett himself isn't tied to the past like that.  He like most investors is handicapping the future.  Evaluating stocks and determining that it's more likely that something good will happen verses something bad.  Then buying and waiting.


  1. Marty Whitman's books/shareholder letters are always fun to read because of his complaints against efficient market theory. He calls academics technicians with PHDs. Some of his points include the following (paraphrased):

    Academic theories are sort of ridiculous if you deconstruct them. One such claim is that because mutual fund managers cannot routinely beat the market that the market is efficient. Mutual funds aren't exactly the best vehicles for beating the market. You may have uneducated investors that force the mutual fund to sell at exactly the wrong times; the fund may be constrained to investing in one industry etc. The theory is like saying that because a man in a blue shirt with one arm tied behind his back can't box that all men with blue shirts can't box. Routinely beating the markets is difficult but beating them on average over time has been done by Third Avenue and others.

    Another ridiculous thing about typical academic investment theory is that although efficient market theorists do not believe in technical analysis they use technical analysis in numerical form (beta). Whenever a theory contradicts itself it's probably not worth the effort of learning. Also when a theory (such as efficient market hypothesis) is later refuted by research of the creator of said theory it is probably not worthwhile. Yet it is still taught in universities all over the world.

    There are some academic-types that conduct worthwhile research (Piotroski, Altman + so on). But I find it much more interesting to buy books detailing the experiences of successful practitioners because their advice is more likely to be worthwhile than someone with limited investing experience. Or for that matter than from academics with terrible track records whose funds have become insolvent.

  2. Your blog contains an really astounding information to your readers and I really enjoyed this one. thanks for this post.