Managing complexity

Reading time: 10 – 16 minutes

This is the fifth article in a series about Post-Modern Security Analysis.

The challenge of complexity

In part three of this series, I wrote:

… the first step of Post-Modern Security Analysis is simply to identify issuers or instruments that are too complex to analyze and move on to more worthy objects of analysis.

At one time, securities were relatively simple ...

This raises the questions, “How do you know when a subject is too complex to analyze?” and “What is the nature of complexity in modern capital markets?”

These questions go to the heart of Post-Modern Security Analysis: that of recognizing the problem of complexity and seeing the analyst’s task as a process of first gathering, weeding out, and documenting relevant facts about an issue and only then, with the information carefully recorded, analyzing the points that have been selected and verified.

Under classic investment analysis, as described in Graham & Dodd’s 1934 edition of “Security Analysis”, scant attention was given to the first step: the gathering and documenting of relevant facts. Almost the entire book was focused on the analysis of facts, presumably easily extracted from published sources such as Standard Statistics.

In those years, the analyst’s complaint was more likely to be lack of information, rather than too much information.

In Post Modern Security Analysis we cannot assume simplicity or insufficient information. Instead, we must start with the expectation that the major challenge will be to wade through and manage a vast swamp of information, with the factual mixed in with the fanciful and irrelevant.

Note: In some developing markets with weak regulation and relatively simple, family- or group-owned companies, the assumptions of Graham & Dodd may still be useful.

Once we have dug out and recorded the relevant facts, there are an abundance of textbooks that can help in analyzing a specific type of security or aspect of the market.

For example: See: Investment analysis

The face of complexity

In the early 21st century, security analysts are presented with a daunting quantity of information, easily available on any laptop computer via the Internet.

Complexity has many hooks and snags ...

Unfortunately, most of this information is dispersed among thousands or hundreds of thousands of sources, and in each source, the critical data may represent only 1% of the text to be analyzed — the rest consisting of irrelevancies, opinion, promotional puffery, legal disclaimers, and boilerplate.

Unlike the time of Graham & Dodd, investment vehicles today are often extremely complicated — the largest corporations sometimes have thousands of subsidiaries and affiliates, hundreds of lines of business, and are subject to dozens of legal jurisdictions. The problem is less likely to be a lack of information, but rather too much information.

What were Lehman assets worth?

Sometimes, vast and complex organizations are considered to be “too big to fail” or even as presenting a self-protecting diversification across many fields of endeavor and international borders. Often, however, safety in size or diversity is an illusion. Somewhere in the intricacies of a hyper-complex organization there may be a “toxic situation” — a hidden bomb waiting to be detonated to bring down the edifice.

Lehman Brothers collapsed in 2008 because it was over-leveraged, but it could not be saved, even by government intervention, because of over-complexity — no one seemed able to quantify the value of its assets.

Investment always involves risks, but there is a difference between normal business risk (competition, poor management, the business cycle, and the numerous “insurable risks” like fire, flood, and other named hazards) and hidden structural risks that may exist in overly complex organizations. Uncontrolled complexity is perhaps the greatest risk of modern corporations.

Unexpected risk

If an analyst is able to identify a risk, it is then possible to weigh potential rewards against this risk. However, if the size and nature of a risk is unknown to an analyst, not because the information was unavailable, but rather because the analyst did not do the research and uncover facts that could be instantly revealed with the click of a mouse — the question is “did the analyst act professionally?”.

For example, how many investment advisors recommended Bernard Madoff to their clients, without checking out the firm’s auditors, “Friehling & Horowitz”? Facts about the auditor may seem like a trifling detail, but in this case, simply checking out this tiny storefront operation would have told a prudent analyst that the risks in the Madoff proposal outweighed any possible benefits. Many who masquerade as analysts (including officials of the SEC) did not bother to throughly verify the simplest details of the Madoff operation, to the immense disadvantage of their clients.

Capital markets have become so abstruse, with so many nooks and crannies to be investigated, that it is tempting to assume the absence of risk, although this is not the counsel of prudence.

For example, one would expect that securities properly deposited with Central Depository (Pte) Limited, Singapore, would be safe, since Singapore is a major financial center in Asia with an excellent reputation for honesty and zeal for investor protection.

Securities are real only in legal hyperspace ...

However, the “Terms and Conditions” of the CDP Electronic Services (March 29, 2004) contain disclaimers against loss due to “any failure, delays in transmission, interruption, errors, omission, or breakdown of the CDP system or any equipment, system, server software, or terminal of the Depository or of its agents.” The remaining clauses of this section on “disclaimer & limitation” fairly well absolve the Depository from responsibility for an investor’s loss, even when a reasonable investor would ordinarily expect the loss to be covered by the Depository.

Now, an analyst knowing of this clause, may determine whether or not the risk is significant in a particular situation. If the analyst is considering holding millions of dollars in book-entry securities of a Singapore corporation for an extended period with CDP, Singapore, it would certainly be well-advised to be familiar with the terms and conditions of the depository.

Too often, a security analyst may think his or her job ends with looking at financial statements, the terms and conditions of issues, and statistics on the line of business. But modern securities are far more complex than that and ignoring operational details, tax laws, and local regulations may put an investor in peril. If the security analyst doesn’t do a thorough job in evaluating an investment, who does?

Taming complexity: gathering the facts

The process of taming complexity starts by having a disciplined, systematic approach to “gathering the facts”.

Gather facts as in an encyclopedia ...

1. Break down your topic into a series of “articles” (as in an encyclopedia). For example, to study a closed-end fund, set up separate “articles” on the issuer, each type of security issued, the management company, the law that regulates funds, tax provisions related to funds, and so forth. There may be less than a dozen “articles”, or hundreds, depending upon the complexity of the issue. In this process, you might soon identify organizations that may be eliminated simply on the grounds of excessive complexity.
2. For each “encyclopedia article”, create an outline of subjects to be covered and questions to be answered.
3. Working from article to article, search the Internet and other sources for facts relevant to the questions that you’re trying to answer. Transcribe the facts that you find, noting the sources.
4. Document your findings in clear, concise terms as if you were writing an encyclopedia article.
5. Save your research as a permanent record that you can study and update later.

Tools of the trade

Documenting complexity in the 21st century ...

If you were doing this research in the 1930s, in the time of Benjamin Graham, your findings would probably be filed in a manila folder, with type- or hand-written notes describing your findings, supported by newspaper clippings, photocopies of documents, a Standard Statistics report, a columnar “spreadsheet”, and perhaps a price graph. You would probably have only one folder because issuers were simpler than today. To gather this information, you might have had to visit libraries and official archives and clip and paste newspaper articles.

Filing system in Ben Graham's time ...

By the 21st century, in the age of Internet, your “folders’ would be electronic and your references to sources would be Internet links and “bookmarks”. Because issuers are so complex and because there is so much information, you might use dozens of electronic folders to sort out the facts. Instead of photocopies of documents, you would have copies of digital documents. With a laptop computer and access to the Internet, you may be able to conduct extensive research without ever leaving your chair.

In either case, you will have an organized set of facts on which to base an analysis. Is time, additional facts will be gathered and added to your “filing system”.

Disciplined documentation

Security analysts can learn from computer programmers the importance of documentation.

In the days of Ben Graham, an analyst’s “research” might consist of a single manila folder with some hand-written notes and comments on copies of financial statements, spreadsheets, or price graphs. Opinions and facts would be mixed together, as there weren’t that many facts and it was easy to tell the difference between fact and opinion and the source of information was often obvious.

In the 21st century, information must be treated with much greater formality, because there is so much data and the analyst needs to take precautions to not confuse facts with opinion or to forget where information was gathered.

Here are some basic rules for documentation of security research:

1. Be careful not to include non-facts (i.e., opinions) in your research notes. First of all, you generally don’t want to rely on the opinions (conclusions) of others, but rather to focus on factual data from original sources. Second, don’t jump the gun, by noting your conclusions before you have the facts. When you get to the analysis stage of your research, write out you analysis in a separate document, rather than in the folders in which you’ve gathered the facts.
2. Record your research in brief, clear statements, with careful indication of the source of each fact.
3. Don’t burden your research recording irrelevant facts (which probably make up most of the raw source material.)

In a future article, I will describe the modern electronic tools that make documentation much easier.

Abandoning research in mid-course

In some cases, you may find that there is too much information available to be analyzed in a timely fashion. The investment opportunity may be just too complex.

An organization may have too many subsidiaries, affiliates, lines of business, diffuse matrix management, or other elements that are difficult to grasp. Or you may find that critical information is not available or is undisclosed. If you can’t understand a situation, move on.

Complexity is not a virtue ...

There is no need to continue an investigation of the facts if you run across information that make the opportunity unsuitable for your needs, or if you conclude that the situation is too complex to permit you to waste time on further research.

Complexity is not a virtue. Complexity may even be a vice.

Your time is valuable and limited

You will not be able to understand every situation, for some investments are made purposely complicated by unscrupulous operators to mislead investors. Securities regulators generally only require “full disclosure”, allowing truthful disclosure of excessively complex situations that investors are unlikely to be able to understand to pass unchallenged. (More often than not, regulators themselves cannot perceive a dangerous situation.)

However, by adopting a formal, disciplined approach to documentation of facts before reaching a conclusion, you’ll be able to eliminate overly complex situations and save time.

Next Lesson: Post Modern Security Analysis: Part Six (Collaborative Research)