|
In the case of a statistical
model, we may have mistakenly excluded key variables…or assumed a
mathematical structure divorced from reality.
|
STOCKS FOR THE
LONG RUN, BLACK SWANS, OR ENCOURAGING
CONDITIONS
While
studying physics, we came across the value of Planck’s constant, the value of a
unit of energy in granular quantum mechanics. Its value is 6.626 x 10-27
erg-sec. We wondered why Planck’s constant, and a few other natural constants, were exactly that. The 9/4/10 Economist reports that there
is now a partial answer to this cosmological question. Planck’s constant is
part of the universe’s fine structure constant i.e. 1/137.036; and researchers
in Australia have found, “…the fine structure constant may not actually be
constant after all. Rather, it seems to vary (somewhat) from place to place
within the universe.”
The
question this article asks is whether stocks can be usefully valued according
to a constant criterion. By citing several cases of evidence, we think not
because society evolves. But there are some simple qualitative universals in
finance, to cite two main examples, 1) Finance requires the capital of social
trust. 2) It’s better to do things soundly, with a margin of safety, to avoid
structural collapses. We will discuss what the concept of a margin of safety
means for stocks when the metric varies with conditions and events.
Quantitative
Universal Stock Market Models
Universal
quantitative stock market models are the basis for quantitative portfolio
management and financial economics. These models are implemented as follows: 1)
Start with the model. 2) Fit it to an appropriate
historical data series and voilà – a market law of nature discovered. The problem is
these models have been precisely wrong:
1) The Mean Reverting Gaussian model.
The Gaussian model describes yearly stock market returns in terms of their mean
(the value you can most often expect) and standard deviation (the dispersion of
these returns around the mean, taken to be risk). This model is a convenient
basis for almost all of stochastic calculus, quantitative finance, and
portfolio optimization. It is also the basis for Professor Jeremy Siegel’s
book, “Stocks for the Long Run (1998)”.
If the mean reverting Gaussian model is invalid, so is the major conclusion of
the book,“This remarkable
stability of long-term real returns is a characteristic of mean reversion, a property of a variable to offset its short-term
fluctuations so as to produce far more stable long-term returns.” 1
Ibbotson and Brinson
(1987) graphed U.S. stock returns from 1790-1985 and showed that the data was
approximately normally distributed. 2 We took that description with
more than a grain of salt. If you take a large 30 year sample of these
supposedly statistically independent data, from 1956-1985, the graphed
distribution is not normal (but erroneously profitable). If this distribution
of stock returns were a constant law of nature, with measurement error, such
sampling should still preserve the distribution’s bell-shaped aspect but more
raggedly.
US
Stock Market Returns 1956-1985
Ibbotson and Brinson (1987) data
|
|
|
|
|
|
1983 |
|
|
|
|
|
|
|
1972 |
1982 |
|
|
|
|
|
1981 |
1984 |
1971 |
1979 |
|
|
|
|
|
1977 |
1978 |
1968 |
1976 |
|
|
|
|
|
1969 |
1970 |
1965 |
1967 |
1985 |
|
|
|
1973 |
1966 |
1960 |
1964 |
1963 |
1980 |
|
|
1974 |
1957 |
1962 |
1956 |
1959 |
1961 |
1975 |
1958 |
|
←30-21 |
20-11 |
10-1 |
0-10 |
11-20 |
21-30 |
31-40 |
41-50→ |
However, our article, “The Nature of Stock Market Equilibrium”
shows that a stock market econometric model derived according to the Gaussian
assumption empirically existed during the years (1968-1999); but justifying
this model against real financial data required Nobel Prize winning logic (not
ours). Analyzing by level, rather than by yearly change, the stock market error
corrected according to events, around the economic fundamentals due to the
Fed’s efficacious monetary policy. The present problem at this 2010 writing is
that the Fed’s monetary policy, with rates at the zero bound, can no longer
error correct because there are too many problems with the economic system. The
Gaussian model describes a fundamentally balanced economy (if such ever
existed); it definitely does not describe a very imbalanced one. Paul Krugman writes in the 9/2/09 NYT:
|
Unfortunately,
…(the perfectly rational and mathematical)…view of
the economy led most economists to ignore all the things that can go wrong.
They turned a blind eye to the limitations of human rationality that often
lead to bubbles and busts; to the problems of institutions that run amok; to
the imperfections of markets — especially financial markets — that can cause
the economy’s operating system to undergo sudden, unpredictable crashes; and
to the dangers created when regulators don’t believe in regulation. |
Our other objection to
the Gaussian model is methodological. Like a cobb salad, it chops up, randomly mixes, and
therefore obscures the impact of major events on stock market prices.
Consider this sequence; if you look at the cumulative return of large cap stocks
over the last decade (Ibbotson, 2010), you will see the following:
Return of a 2000 Stock
Investment in Percent per Year, Ending in
|
2000
2001 2002 2003
2004 2005 2006
2007 2008 2009 |
|
-9.1
-10.5 -14.6 -5.3
-2.3 -1.1 1.1 1.7 -3.6 -0.9 |
In 1930, the cumulative return over
the ten year period was -0.1 per cent per year. Disregarding the previous years
of 1929 and 1999, the cumulative ten year returns for all other periods
since 1930 were positive (range +1.8% to
+20.1%). The present crisis of the financial system is major. The problem
now boils down to excessive debt: on both the private and now public balance
sheets, that has to be gradually reduced. In the 9/20/10 Business Week, Bill Gross says, “(The crisis) was a 10-20 year
moving in, and it will probably take a long time moving out. Some have
suggested the Biblical seven years fat and seven years lean.”
2) The Paretian Distribution looks somewhat like a Gaussian
distribution but is more peaked with an increased tail risk at the extremes.
Very good events and very bad events are more probable than in the Gaussian
case, and there is no mean reversion. The Gaussian model makes two main
assumptions:
a)
Yearly stock
returns are independent of each other and identically distributed (i.i.d.).
b)
The
variance (dispersion) of yearly stock returns is finite.
If stock returns have variance, they have a mean.
A
trading strategy is not practical for most people. We assume that most of our
readers would like:
a) Investments
that serve as a store of lifetime wealth.
b) Investments
that offer some certainty of profit after a reasonable holding period.
Qualitative
Financial Principles, Character
is Destiny
But
can finance, particularly stock valuation, be viewed as a social science,
covering perhaps some more distance than the case-by-case method without
crashing? We reviewed a book by Geoffrey Moore called Dealing with Darwin (2005). He makes a key distinction between
businesses whose expertise is efficiency versus those whose core expertise is
the design of complex systems where,“each
customer situation is inherently unique and evolving, so rather than having
tight execution rules, what is required is tight alignment around key execution
principles (our emphasis).” 4
1) Finance is built on the capital of
social trust.
2) It
is better to do things soundly, which means designing in a margin of safety,
just in case.
Now
that a large part of the U.S. financial system remains mistrusted; for example
the private mortgage system, the structured finance system, and lending to
small businesses, we don’t think anyone can honestly argue against these
principles. We will discuss the second, when a stock has a margin of safety,
and then how its environment is relevant.
The
Margin of Safety Principle
In
his lapidary book, “Value Investing (2001)”,
Professor Bruce Greenwald of Columbia University writes:
|
(Benjamin) Graham did more than put investing on a
rational foundation. He also engraved on the minds of his students and
devotees a strong preference for what would now be called value investing. To present the most
succinct description, the value investor seeks to purchase a security at a
bargain price, the proverbial dollar for 50 cents…. (Inherent in the Graham-Dodd approach is a) broad
vision, now commonplace among the most sophisticated investors, that
assessment of the strategic position of a company is central to any useful
valuation… A further advantage of the value investor approach
– first the assets, then the current earnings power, and finally and rarely
the value of the potential growth – is that it gives most authority to the
elements of valuation that are most credible….Ultimately the future does
matter, but it is important to separate what we reliably know today from less
secure conjectures about tomorrow. 5 |
The
metrics for value investing vary with time, but fortunately slowly.
Benjamin
Graham developed his value methodology in the 1930s, looking for low P/E stocks
(based on historic P/E trading ranges) and stocks selling at price below a
company’s (cash, receivables, plus inventories - liabilities). Accord to
Professor Greenwald, “…in the contemporary investment world, net-nets are, with
only the rarest exception, a distant memory. Modern value investors have had to
develop new approaches to discovering and valuing assets that allow them to
move beyond cash, accounts receivables, and inventory while still making their
investment decisions on the basis of the value of the assets today, rather than
(projected) earnings, cash flow, or whatever in the future.” 6
For
Growth at a Reasonable Price (GARP) investors, the
table on p. 144 is very useful. In that table Professor Greenwald, updating
value investing to include Warren Buffet’s growth case, provides a rigorous
calculation of a company’s value based upon:
1) Its sustainable cash flow
(historically determined).
2) Required rate of investment return.
3) Its conservatively estimated
growth rate, based upon growth within its franchise.
4) Return on capital invested.
In
spite of the current financial crisis, the parameters of the analysis remain as
before, with one exception. Even after this financial crisis, we can still
assume a required investment rate of return of 8-10% (depending upon the
company’s risk), intrinsic company growth rates of 4-6%, and a return on
capital calculation specific to the company. The only change we have had to
make, for large cap growth stocks, is increasing the margin of safety from 33%
to around 50%.
Encouraging
Conditions
Are
we taking advantage of the bargains available in the stock market, increasing
our allocation to stocks? Not yet. Value investing also requires a catalyst, an
event that will increase demand for a stock and propel its price upwards. An
environmental variable, such as sector rotation, can be a catalyst. Since the
Financial Crisis of 2008 is a crisis of the financial system, events
signaling the beginning reform of that system should occur. The first place to
expect that is in Washington.
On
Wall Street, the CW believes political gridlock is desirable because, “That
government which governs best governs least.” (in
spite of the financial crisis). Such sentiments sparked the American Revolution
with the Tea Party in Boston harbor, a protest against British control over tea
taxation. That protest, however, had nothing to do with the subsequent role of
government in developing the United States. The controversy of what the nation
might become was substantially settled by Alexander Hamilton (1755-1804), the
nation’s first Secretary of the Treasury, who advocated government intervention
in favor of business and economic development. We think the constructive role
of government really matters because, besides business, three
government-influenced systems now need repair: the federal budget, international
trade rules, and education.
The
outcome of the midterm elections in November, 2010 is crucial; because if the
far right obtains a blocking majority in Congress, the United States will not
be able to deal with its fiscal problems. The U.S. wishes to remain the world’s
banker. Imagine depositing your savings in a bank whose
management is gridlocked and squabbling.
Portfolio
Policy
What
to do? Most investors, we assume, seek safety in principal and a reasonable
profit. In The General Theory, Keynes
(1953) wrote about ignorance and then allowed, “There are…certain important
factors which somewhat mitigate in practice the effects of our ignorance of the
future. Owing to the operation of compound interest…there are many individual
investments of which the prospective yield is legitimately dominated by the
returns of the comparatively near future.” 7
In
traditional investment practice, bonds are for income and stocks are for
growth. Since U.S. growth is likely to be low as the economy recovers from this
debt binge, we would continue to suggest high quality medium-term bonds
*, held in a laddered portfolio; and, provided you are comfortable with their
assets, publically traded limited partnerships in the energy area. A laddered
bond portfolio is not the same as a bond mutual fund, whose maturity is
generally held within a narrow range. A laddered bond portfolio contains
corporate bond issues that mature in successive years (generally two to eight
years). The way to think of it is as a single bond, with a starting maturity of
around four years. Since savings accounts earn virtually no interest, a
laddered bond portfolio returning 4-5% is an acceptable substitute because its
risk decreases with time. This is a very nice characteristic with the
future monetary policy uncertain. Putting together such a bond portfolio
probably requires some professional assistance.
The
second component of a traditional investment portfolio is income producing
stocks. As a matter of policy we tend to avoid them because their duration
(payback period) is long, and thus they have high interest rate sensitivities.
If the economy goes to extremes, the dividend will be cut or if inflation
soars, their prices (in the short-term) will crash along with bond prices.
For
most investors, we think an investment in large-cap international growth stocks
is probably the easiest and the most appropriate. But we would like, as a
necessary but insufficient condition, to see whether Washington avoids gridlock
before assuming the financial system is starting to head towards error
correction and dynamic stability.
The
Economic Position of the U.S. in the World
On the 9/21/10
Bloomberg, the Chief Globalization Officer of a large U.S. high-tech company
said they would seek expertise in:
1) China for manufacturing.
2) India for software.
3) The U.S. for designing the I Phone (manufactured from a supply chain extending
throughout Asia). We thought he had to mention the U.S. to be politic.
The U.S. ought to demand broad
reciprocity from its trading partners. Consider this
article by Thomas Friedman in the 9/21/10 NYT. The issues of political
economics that we discuss will have a large impact upon equity investing.
Judgment now requires asking when the level of system uncertainty is
acceptable.
|
“In
normal levels of uncertainty, we’re wallowing in it like happy pigs, because
you need that uncertainty to find bargains…(However,
for now there’s so much uncertainty)…it’s overwhelming.” David Pedowitz, Neuberger Berman, WSJ 9/24/10 |
(Due to the length of this article,
you may have to adjust the print settings on your browser to enable a printout
in 12pt. text. In Windows 7 Explorer: go to the adjustment next to the printer
icon, then to Page
Setup, and uncheck the Enable Shrink-to-Fit box. What could better illustrate the Bhidé
article we cited. Windows 7, by the way, was developed
in a more-or-less decentralized manner.)
* On 10/19/12, we added: Due to very low AA bond yields, we would not pursue this strategy at this time.