Process versus Outcome in Investing

It is Columbus Day -- not the "observed" day where government agencies and bond traders take the day off, but the real one.  Who better to remind us than Art Cashin:

On this day in 1492, Christopher Columbus landed at the Bahamas, believing that he had landed in the islands off Japan. And so history is indebted to a man with a flawed theory who set out on a venture where he didn't know where he was going, suffered sharp losses on the way and didn't know where he was when he got there. And he did it all on borrowed money.

The world only cared about the results.  It reminds us of investors.

Why Process is Important

Looking only at investment outcomes leads many into a trap they do not understand -- getting Fooled by Randomness.  With thousands of investment managers, gurus, mutual funds, and ETF's there will always be a choice that represents a hot hand.  Does this really predict future investment success?

Investors chasing performance get returns of about 1/2 the market average.  Two prominent commentators describe the significance of process.

Bill Miller, the stellar manager of the Legg Mason Value Trust made the point clearly at an investor conference, quoting former Treasury Secretary Robert Rubin:

Any individual decision can be badly thought through, and yet be successful, or exceedingly well thought through but be unsuccessful because  the recognized postulate for failure in fact occurs. But over time more thoughtful decision-making will lead to better overall results, and more  thoughtful decision-making could be encouraged by evaluating decisions on how well they were made rather than on the outcome.

At the start of this year, Miller gave his opinion about the markets:  The stock market is still cheap.

Barry Ritholtz recently wrote a thoughtful comment with the same theme:

Consider the long list of folks who have been right in their analysis, but wrong in the timing of the market reaction to this; Then think about some of the weaker bullish arguments -- there have been an enormous run of absurd arguments, false theories, ridiculous analyses. Regardless, these get overlooked by many as the markets continued upwards. Right answer, wrong process.

Ritholtz is noted for his bearish perspective on the U.S. economy and markets.

Conclusion

Anyone can look at a table to check out recent results.  Few can evaluate the investment process of pundits and fund managers.  They all sound good.  As one investor told me, "I understand what you are saying about process, but I cannot use that.  I only have results to go by."  The irony is that this may be the worst possible method.

While we have been enjoying a very good year, we fully appreciate this investor dilemma.  It is difficult to persuade clients about process when you are in the middle of a "cold" streak.  It takes confidence in one's process to maintain discipline in the face of adversity.

Cashin notes that the bearish advisers of Ferdinand and Isabella warned that Columbus would run out of fresh water before reaching Japan or China.  They were correct.  He would have.  It was a stroke of fortune that he reached the New World and more water.

It is up to each person to evaluate reasoning and process as well as results.

Comments

[RB]

I found nothing particularly controversial with "Blink." Gladwell says that first impressions can be misleading, but experts who train their first impressions/intuition/unconscious can arrive at the solution faster while avoiding the traps of prejudices/predispositions. Intuition is not just inbuilt (such as recognizing emotions) but can also be learned through careful study. In uncertain situations, it can be about learning to filter out all but the most important factors. Anyone who has realized the importance of "Set the problem aside and take a fresh look" understands the importance of the unconscious mind in problem-solving.

[VennData]

A Black Swan aficionado, possibly NNT himself, might say that the floor of the Atlantic Ocean is littered with the failures that weren’t able to achieve the acclaim and glory given to Chris Columbus. You’ll never hear of a one.

This post is something anyone who has active responsibility for allocating invested capital should read, but turning any ‘ah-ha’ into actionable outcome will be difficult.

I believe analyzing the investor process takes systematic thinking since you are analyzing an approach, a system, not outcomes. I don’t believe there is any room for the poppycock in ‘Blink’ since – save for all the caveats Caldwell dishes out – since it’s un-measurable, just a bunch of anecdotes. An actionable, measurable systemic would make an excellent framework for judging a particular investor’s process. A Black Swan aficionado would say there are plenty of other antidotes at the bottom of the Atlantic (NNT’s frustration with narratives)

One process analytic rule I like seems to be simple but takes quite a bit of work sometimes, it is to identify if a prospective investor says something material that contradicts other material claims in their process. So for example, on interesting contradiction within the many “gold bug” investors is as follows:

Gold is the only currency because it’s safe from government intervention and control, yet the governments of the world conspire to keep the price down.

The ‘gold bugs’ sell you on their particular private bank, usually offshore, or bags of silver bits, etc etc that they say will protect you from the upcoming crash. When the likes of GLD and SLV came out they pooh-poohed them.

For example, in the back of Peter Schiff’s new book he even says that you shouldn’t buy the precious metal ETFs because he just doesn’t trust the structure and anything can happen to them. While he has many interesting things to say, ‘anything’ can happen to any investment. The fact that you can buy GLD through a discount broker and sit on it for the next thirty years is going to make it difficult for him to sell you gold coins, bars, and the tax-free Canadian royalty trusts he so loves. By the way, the Canadian government, out of the blue - sacre blu – is revoking that tax free status. So I guess he’s right, anything can happen.

The process is key. Expanding on the details will be a challenge I look forward to future posts on the subject.

[Mike C]

"Imagine how it would be if someone five or ten years from now was trying to analyze the portfolio of an apparently successful manager who might have done the same thing by selling short dollar futures."

Good point. One thing to consider IMO is that some investment positions benefit from multiple variables/influences, so they may not be identical to just being long or short dollar futures.

For example, I've had a substantial allocation to a diversified commodity fund the past couple of years which has contributed to strong portfolio returns. Now, commodity prices certainly have benefited from a weakening dollar, but they have also benefited from a global economic boom, especially in China.

IMO, an interesting theme here might be that a good investment process doesn't rely on binary, 0 or 1, right or wrong bets (zero sum such as futures), but instead seeks to take investment positions that could benefit from multiple influences.

I think Ward has an excellent point about the role intuition and subjective judgement plays. It seems reasonable that not every element can be exactly quantified and analyzed.

[Ward]

All of these comments are fascinating. The role of intuition, ie. BLINK information processing in the money management process is very much under appreciated in my opinion. We all get too caught up in analysis and process and obviously no large investor will give you his money to manage if you say something like, "I just buy the stuff that feels right and sell it when it feels like time." and yet that may be the only part of the process that can explain the persistence of some in the upper quartiles.

[RB]

A very complex post. It appears that it takes an expert investor and did Julian Robertson not have enough of them? How much of a dilemma is process? Here's a corner case:

Edward Prescott is a Nobel winner. He argues that Irving Fisher was right in 1929.
http://ideas.repec.org/a/ier/iecrev/v45y2004i4p991-1009.html

The timeframe for process here would be measured in decades although one could argue that the market is more efficient today (notwithstanding Buffett's experience with unwinding General Re's derivatives).

I'm reading Blink (fascinating read) on the Prof's recommendation and it's hard not to come away without a notion of the existence of expertise even in highly complex situations. Again, it seems like all of this works with expert consumers who are aware of the expertise of the producers or who just have blind faith.

[Jeff]

Mike -

Thanks for your reply. "Ignore" was a poorly-chosen word. Sorry. We must all remember that people can look at the same evidence and reach different conclusions.

I'll develop the theme more in future posts, but let's start with correlated stocks. Right now on the international front returns from almost everywhere outside the U.S. are correlated with each other and with commodity stocks. The common factor is the dollar. It is difficult enough for people to see this (temporary) link right now. Imagine how it would be if someone five or ten years from now was trying to analyze the portfolio of an apparently successful manager who might have done the same thing by selling short dollar futures.

Analyzing old portfolios is tough.

Thanks again for generating ideas.

Jeff

[Mike C]

Jeff,

Thanks for the reply and the compliment.

I'm not sure if I am misunderstanding, misinterpreting, or if this comment was specifically addressed to me:

"Basically, you *ignore* these opinions when I am trying to stimulate you to read them.

To be clear, I don't ignore your opinions. If I thought your opinions/thoughts/posts should be ignored, I'd stop reading your blog and commenting (which I've done with a blog that you recommended which IMO is mostly worth ignoring).

I am trying to be *very aware* of confirmation bias so as I've stated before I consider your blog mandatory reading because of the intelligent, thoughtful, balanced analysis/commentary you do regardless of whether I ultimately agree with the conclusion.

Hopefully, there is the potential for the dialogue to be useful to both of us. My thought is that dialogue is where further learning/insight occurs, and my replies are not intended to be taken as me lecturing someone else, but just to stimulate further back and forth dialogue. Perhaps I'm reading something that isn't there, but keep in mind, that with some of your readers, you may not be "educating" those with less knowledge, experience, or credentials, but instead having a discussion with someone more or less on equal footing.

You mention you do not agree with my conclusions regarding relevant time frames. I would be interested in specifics as to what time frames you consider relevant and why? How long do we need to evaluate a day trader? Is 1 day sufficient? Are 20,000 days necessary? If someone states their investment strategy/process is to outperform over the very long-term including both bull and bear markets, over what time frame should we evaluate their results? I am very open to other opinions on this.

I have been revisiting my assumptions lately to a degree, and you pretty much get 100% credit for that (you've had a meaningful impact on my thinking!). Prior to reading some of your work, I was pretty much 99% sure that the current market environment was a "echo bubble" from the late 90s. Now, I'm not so sure. At the same time, there are some very, very, very intelligent people with alot of experience and credentials who essentially hold this view. IMO, one must be careful when dismissing their thesis out of hand. Confirmation bias cuts both directions.

Thankfully, it is not necessary to hold the correct market outlook to make money. I (and my very small number of clients presently, it is hard building a RIA business when you are young) have done very well the past several years despite me being cautious.

As far as identifying highly correlated strategies, yes. Owning 15 different energy stocks would be a highly correlated strategy. Owning FXI, EEM, and Latin America would be highly correlated as they are all exposed to the emerging market theme to some degree. Could the average individual investor recognize this? I don't know, but they probably could understand it if I pointed it out and explained it to them.

Anyways, as always I think your blog is enlightening and thought-provoking. There are few if any bloggers that can stimulate me to rethink any single thesis I have (as I do a tremendous amount of work, studying, thinking to develop any given thesis). You do that. I look forward to your continuing thoughts.

[oldprof]

Mike -

I very much appreciate your willingness to make thoughtful comments. In many ways, investors like you are the target audience for my work.

Concerning this comment, it reminds me of my days in the classroom where I had a presentation that did not seem to work -- at least for some people.

In this post I tried to show that the very first thing an intelligent investor would do -- looking at a chart of past performance -- was wrong. To show this I offered advice from Barry Ritholtz, who has the most acclaimed financial blog, Nassim Taleb,who treated the subject in a widely-acclaimed book, and the advice of the leading mutual fund manager. To this I added my own recommendation -- more modest in influence, of course.

Basically, you ignore these opinions when I am trying to stimulate you to read them. You are confident of your ability to interpret results from various investment strategies.

I congratulate you on your study and your efforts, but I do not agree with your conclusion. You have implicit assumptions about what time frames are relevant and what periods of market history are "typical." You need to challenge yourself on these points. In particular, the market history for the last several years is wildly atypical, throwing off many of the pundits.

I have written about this in the past, but perhaps I need an update showing the cumulative effect of the build up to Iraq, the business emphasis on outsourcing and consultants leading to massive earnings and strong balance sheets, the skepticism introduced by the 2004 election campaign, the breakdown of the "presidential election cycle", the Fed rate increases from an abnormally low level -- and other factors.

None of these things have been present at any prior point in history.

It has also been a sweet spot for value investing, a method you endorse. Will you know when to switch?

Your analysis of results is better -- much better -- than most would do. Let us suppose that a manager did not have a single big winner. Can you tell that he did not have a bunch of other highly correlated strategies?

In our work, we analyze approaches like this all of the time. It is difficult and well beyond what an individual investor could do, even if making a major commitment of time and effort -- a high hurdle.

So I need to rethink how I present this.

Thanks again for your many thoughtful comments. You are obviously a very intelligent investor.

Jeff

[Mike C]

"Few can evaluate the investment process of pundits and fund managers. They all sound good. As one investor told me, "I understand what you are saying about process, but I cannot use that. I only have results to go by." The irony is that this may be the worst possible method."

I think at the end of the day though, results are really the only thing that investors can rely on. The absolutely massive caveat being the results have to be analyzed correctly in 2 major ways:

1. Composition of results. Did the manager do good because he bought one risky bet (like loading up on FXI, AAPL, CROX or GOOG at the beginning of the year) or because of numerous bets that worked. The former is likely luck while the latter is likely a good process. This also means you have to look at overall portfolio performance, and not just a single investment or trade. A good process can lead to a bad outcome on any single trade.

2. Time frame the results are measured over. I think this one is absolutely key. And the time frame has to be tailored to the strategy/process the manager is employing. It is probably fair to evaluate a day-trader over a 90 day period. That should be long enough to evaluate if his day-trading process works. On other other hand, a long-term value investor may need years for his process to show solid results especially in a bizarre market environment like say 1998-2000.

There are so many competing investment philosphies and processes, and many "work" in that they will show good results over a long enough time frame. Buffett picks stocks different from Lynch who picks stocks different from Heebner, yet they have all produced great results over multi-year time frames.

I recently presented my investment philosophy and process to my girlfriend's parents. They told my girlfriend they were extremely impressed with my presentation and knowledge. Truth be told, I could probably deliver a completely different presentation with a radically different philosophy and process and come across just as impressive.

The average person truly has little chance of distinguishing a solid process from a nonsensical one unless they want to embark on alot of studying themselves in which case they could probably manage their own portfolio.

So the amateur/novice or prospective client really has nothing else but results and perhaps personal trust in you to go on.