Ten key investment issues

Ten key investment issues

A list can never be complete or fully comprehensive and this treasury is no ex-ception. There are, of course, numerous investment ideas and in focusing on the thirty we have discussed, we have had to omit several that could easily have been included. So here are brief discussions of ten more issues in investment that we think might be worth exploring in greater depth for the future.

Valuing Intangibles
The so-called knowledge economy is making asset valuation an increasingly difficult process. Easily measured corporate assets, such as plant and equipment, are declining in importance while less tangible assets, such as brand names, technology, software, and the skills and commitment of the workforce, are becoming more important. This is making even general judgments about the intrinsic value of a business very difficult.
Traditional accounting measures the financial impact of what has happened. And it nearly always relies on money transfers to establish a value transfer. But increasingly, with customers, competitors, and alliances changing roles, money does not capture every important event. Often knowledge transfers, market access, product integration, and personnel assignments are more important than a notation on an income statement.
Today, we have no systematic way of measuring these transfers, which are critical in a knowledge economy. We now have to find ways of valuing intangibles. Techniques to do so are being prepared and we shall see common usage in years to come.
Ethical Investing
Proponents of ethical investing advocate that businesses should be socially and environmentally responsible. For example, John Elkington, a consultant on issues of sustainable development, argues that companies should now pursue a triple bottom line of economic prosperity, environmental protection, and social equity. Ethical funds avoid investing in industries such as tobacco, alcohol, drugs, and arms dealing, or companies that flout health and safety regulations, use animal testing, or operate in oppressive regimes.
Ethics used to be a cyclical phenomenon. After a business expansion and bull market followed by a decline, there is usually a resurgence of interest in ethics and principles, especially in business. And when there is easy money to be made, the principles are forgotten, and in the aftermath, we all say: “How could we not have thought of whatever it was that we should have been doing to take care of other people and not exploit the conditions?” But a strange thing happened over the course of the 1990s: Interest in ethics picked up during the expansion. Business schools relaunched ethics courses, universities made a big push to teach ethics to undergraduates, and professional associations developed ethical guidelines. But ethics is extremely simple: Just ask somebody what have they done, at their own sacrifice, to benefit the interests of others.

Insider Trading
Even some hard-core believers in market efficiency will concede that company
directors trading in their own shares tend to beat the market. This suggests that
some people are more plugged in than others and that the key to active invest-
ing might be to follow smart insiders or, better perhaps, to follow companies
with smart greedy insiders: companies whose directors own a lot of stock and are
buying more. Certainly, that is the implication of a book by Michigan finance
professor Nejat Seyhun on how to make use of investment intelligence from in-
sider trading.
The conventional regulatory response to insider trading has been to try to keep the market fair by preventing insiders from trading during corporate black-
out periods and making government officials use blind trusts. So while the United States considers itself the model of disclosure in financial affairs, infor-
mation known by insiders is kept out of the market by our archaic rules. Perhaps
we should instead be encouraging insiders to get their information priced into the market by trading whichever way they wish, revealing their insider status though perhaps not necessarily their identity.

Agency
Agency issues tend to be forgotten in bull markets. But there is an ever-present phenomenon of people engaged in generally high-priced services—like investment managers, brokers, consultants—acting in their own interests rather than those of the fiduciaries they should be serving. Usually, the mistakes or omissions are fairly minor—high expenses; a limousine when a subway would do; wasted meetings; soft dollars, where institutions tolerate higher commission charges than do the more tight-fisted retail customers using electronic brokerage for their own accounts; and slow innovation, because it is so much fun making money the old-fashioned way.
This is really the edge of ethics. It is not out-and-out dishonesty. Rather, it is the avoidance of putting somebody else’s interests ahead of your own. It is a field of academic study, which should be pushed more by the various industry groups that hold out their standards of high ethics.

Models
Models are developed on the basis of history and although they may be quite sophisticated, they frequently prove to be useless or wrong. The problem is that they depend on the assumptions that the past is the prolog to the future, that all information is captured by the model period, and that no simplifying assumptions predetermine the outcomes. Models will operate slavishly, continuously doing whatever is instructed by an embedded algorithm. They can be marvelously elegant, and tend to become more so with time as they incorporate yet more initial conditions. But they are dumb: They do not learn but depend on the learning of the model builder.
Instead, we should be using forward testing and simulations, understand-
ing how the assumptions selected for modeling will determine outcomes. Simu-
lations are often messy, fuzzy, and seemingly counterintuitive. Similar initial conditions can produce differing results. The interrelations are complex and chaotic. Instead of neat one-to-one correlations of the components, simulations act as a whole with every part moving at once.
But simulations more nearly act as humans do: learning from the past and getting smarter in the future. Combined with complexity engines that do not guarantee a specific outcome from every event, you can play through business possibilities. We can train ourselves and do forward testing rather than relying on the historical accidents of back testing.

Risk
The conventional view of risk is that the long term is less risky than the short term because there is regression to the mean: If we wait long enough, economic balance will return. There is also the notion of a built-in equity premium: After a fifty-year period of expansion, we take it for granted that equities will produce higher returns because they have in the past—and we think it is because they have higher degrees of risk.
But there is a strong academic view that says risk may compound in the long run, and that you do not necessarily have regression to the mean. Rather, you can have deviations from the mean over long periods of time and things may get worse. There are economics of increasing or decreasing returns; and momentum and expectations may create their own trends. Are we prepared for the time when risk produces lower returns for equities? Or that risk itself, on closer examination, is something other than volatility but rather risk of loss and risk of being knocked out of the game?


Time
We are raised to think of time as absolutely immutable, a steady object: Time
goes on. Economists and investment people, for example, know that time is the
x-axis, with equal units all the way across, and that time flows continuously. And
we wear wristwatches that click at a steady rate. But this assumption is wrong
and it builds errors into our system. Time comes in packets, quanta, bunches of
information, at irregular rates. Time means different things to different analysts.
Some market events on particularly significant days are more important than ten
years of market data.
We should look at time with a great degree of skepticism and search for new tools in dealing with it. High-frequency analysis is one way to take time out of the equation and look at it as a complete variable. We will make better forecasts if we treat time as discontinuous, not continuous. For example, we might guess that events will compress as impending events are discounted faster than before. But events outside the markets will probably not happen much faster than before. So if markets react faster, we should have a quick discounting response to the hints and then long periods of frustrating dullness—a pulsating, quantum burst market with long pauses.

Diversification
Conventional investment wisdom says that diversification is good. If risk (volatility) is equivalent to return and we know how to increase risk, all we have to do is mix assets that are uncorrelated to increase aggregate (portfolio) returns and reduce risk. Alphas, betas, and R-squared are all part of the analyst lexicon. Close to the means, these measures seem to be quite stationary over the time periods we have studied them.
But an alternative complexity-based view suggests that past correlation tells
us little about stress times. The factors that determine values are nonstationary
and combine in different ways. The evolutionary history of markets is unlikely
to combine in the same way each time. And when they part, the results may ap-
pear chaotic.
Causality
Investment people frequently confuse coincidence with causality. With any two sets of finite data, we can find correlations that will satisfy a statistical test. Once we find those correlations, especially if they are plausible, we tend to enshrine them as causality, cause, and effect. We assume that the very conditions that produced those circumstances, coincidences perhaps, in the past can be projected into the future where conditions can be entirely different. Even if they were the same, the result might not be the same in a complex adaptive system. And once we project into the future, we further attribute leads and lags that are quite predictable—or rather, we think they are predictable—and ascribe those to the system that we have produced It is a tissue of assumptions, of implausibility, and produces a result that is
highly prone to flaws, failures, and excuses. We even go on to be prescriptive in
the cause and effect, to say that if we do this, that will happen. For example, we
say if interest rates are increased, that will make the stock market go down. This
is not necessarily true if the market interprets that move in entirely the opposite
way, perhaps as an indicator of a very strong economy. So it is not a one-to-one
relationship. There are new tools that are very infrequently used and that do not
promise as much: fuzzy data, high-frequency data, adaptive systems, simula-
tions. We should look at those as being at least more honest than what we do
now.


“What’s the Market Doing?”
In trying to diagnose the health of markets, it is difficult to find conditions in past practices that fit today’s circumstances. Perhaps it is because we are looking for physical conditions when we should be looking for the mental as a precedent for today’s market conditions. Today’s markets in the United States are manic depressives, with violent swings of euphoria and depression without any con-
nection to a sense of reality. Manic depression is not a curable disease, except in the extreme, but it can be controlled with Prozac and other antidepressants.
Or perhaps the market is a nuclear reactor. A nuclear reactor is absolutely necessary for the economy around it but not the other way around. The economy does not drive the nuclear reactor, except over the very long run. And the nuclear reactor has to reach a very delicate, critical balance within a critical mass to run. But not so much that you generate, in fact, a runaway meltdown, as happened in Chernobyl.
In the final years of the last millennium, we ran into a particular hot spot in this nuclear reactor of the market within the area of technology stocks. Normally, you put down some safety rods to absorb some of the reaction. But the last time we did that in this reactor, in early 1998, it generated a global credit crunch and we pulled the safety rods out very quickly. As a consequence, we had no tools with which to dampen down the market, no safety device. We do not want to have the radioactivity distributed throughout the air as sometimes happens in the case of a nuclear accident, and indeed we do not want to have to encase this market in a coffin of glass.
Finally, we have a very odd paradox today between the theoretical condi-
tions of markets and our abilities to implement market strategies. On the one
hand, markets seem remarkably like those of the 1870s when there was over-
supply of goods and a noticeable satiation on the part of consumers as an after-
math to the U.S. Civil War. Prices went down, commodity prices declined, and
markets were very volatile, much more than seemed warranted by the economic
conditions.
On the other hand, today’s technical implementation faces the challenges
of the new millennium: Markets are global; they are instantaneous; they are ac-
cessible; they are cheap. The individual rocket power of the free market forces
vastly exceeds that which can be marshaled by central bankers and treasuries,
who are powerless to influence these forces and must simply look on to see what
happens when markets move, thereby affecting economies and trade in their af-
termath.