The Concept of Risk
(October 30, 2005)
Dear Subscribers and Readers,
We switched from a 25% short position in our DJIA Timing System on the morning of October 21st at DJIA 10,265 - giving us a gain of 351 points from our DJIA short on July 14th. On a 25% basis, this equates to a gain of 87.75 points. For now, we are completely neutral and in cash. This author is still looking to initiate a 100% long position in our DJIA Timing System in short order - although the market still remains dangerous and uncertain at this point (even with the huge rally last Friday). Despite this short-term uncertainty, however, this author will have to concede that the bulls now have the upper-hand. Anyone that is still shorting the markets here has to be very careful.
In our last commentary, I discussed the recent appointment of Ben Bernanke as the new Federal Reserve Chairman - along with a discussion of his beliefs and the policies that he will likely follow in his term as Federal Reserve Chairman. I argued that he will be a very active Chairman - on curbing both inflation and deflation - and also making sure that the private sector has a very clear idea with the intentions of Fed policy going forward. Please keep in mind that the subjects of finance and risk are still relatively undeveloped. Recall that as early as the 17th century, the modern concept of probability and game theory was still non-existent. The popular thinking at the time was that the outcome of each dice game was solely dependent on the Gods. Modern finance, as stock market participants know it today, basically only emerged in 1952 - when Harry Markowitz first pioneered the idea of the "Modern Portfolio Theory".
If folks like Ben Bernanke, Robert Shiller (with his somewhat utopian view of the world in the 21st century where all kinds of real-world risks can be hedged), George Soros (with his idea of "reflexivity"), Adam Smith (with his book on "The Theory of Moral Sentiments"), and Fisher Black have their ways, however, then we are still in the early days of Modern Finance, as well as our concept of risk measurement as we know it today. The Ben Bernanke Grand Experiment is now upon us, and it is going to be a very interesting and exciting ride, indeed.
As stock market participants, it is imperative that we can assess risk in the proper way. It is said that on the trading floor of Enron North America, the two must-read works were "Reminiscences of a Stock Operator" and "When Genius Failed: The Rise and Fall of Long-Term Capital Management" - two works which this author would also highly recommend reading. The implications of blind risk-taking are highly obvious in both books (Jesse Livermore lost his entire fortune in four separate instances in the first book and the highly talented LTCM team managed to nearly destroy the modern financial system in the second book), and yet, research has shown that many of the most intelligent people on the Enron trading floor were nothing more than compulsive gamblers or too arrogant for their own good. For example, the Enron daily position reports show that Greg Whalley, a former army tank capital and the COO of the wholesale business (whom Jeff Skilling described as one of the smartest people he knows), was down about $30 million in his trading account during 2001 when Enron collapsed (he continued to trade despite occupying an executive position).
Another such figure was the 26-year old trader named John Arnold. In the book "The Smartest Guys in the Room," the authors described John Arnold in this way: "In late 2000, a 26-year-old superstar trader named John Arnold - who was revered for being able to do complex mathematical equations instantly in his head - hit a bad losing streak and went from being up $200 million to being down $200 million in the space of less than a month. On Wall Street, such a performance might well have gotten Arnold fired. But when Skilling heard what had happened, Whalley told him that everything was cool. And so Skilling came down to the thirty-second floor, where the traders worked, and put his arm around Arnold in a public show of support.
The ability to quantify risk is not totally dependent on one's IQ - even though there is somewhat of a correlation. In a complex world and in today's information age, it is the ability to simplify each potential investment and filter out all the unimportant information - as well as minimize the number of real-world variables that we need to guard against. Let me illustrate. Suppose one believes that copper prices have now topped out and are in the midst of heading down. One can do several things to take advantage of this "knowledge" - including shorting the shares of companies that produce copper, buying stocks of companies that use copper (such as homebuilders), or shorting the commodity itself on the New York Mercantile Exchange. The speculator who have only dabbled in equities would argue that the first two options are probably the safest options, as they are unfamiliar with the commodity markets as well as uncomfortable with the huge amounts of leveraged involved. This author, however, would argue otherwise.
First of all, there are many other real-world variables with shorting the shares of individual copper-producing companies. There are geopolitical concerns, as well as concerns about certain individual mines and about company management. For example, what if the company you are shorting suddenly finds a significant amount of copper deposits during the time that you are short the company? Moreover, mining companies usually produce other commodities besides copper. For example, Phelps Dodge has done very well over the last few years most probably because of the rise in molybdenum prices - not copper prices. Freeport-McMoRan, meanwhile, has also take advantage of rising gold prices as well as copper prices. Sure, one can lessen the impact of these other variables by shorting a basket of copper-mining stocks, but this does not diminish the fact that other commodities and real-world variables also have a bearing. To ensure a "pure" play on copper and to minimize the real-world risks, one should directly short the copper contract on the NYMEX - and not the shares of copper-mining companies. Until a company has developed an ETF for the commodity, then speculating via a futures contract is still the safest and most logical way to go.
Another way to minimize the number of variables in a potential investment is to gather as much information as possible on each of your potential investment. Of course, this is based on the beliefs that the market is not totally efficient - as this author has always believed. In this case, diversification or buying the S&P 500 blindly is not sound advice - just ask anyone who bought during March 2000. On the other hand, if one has done his or her research, then holding four or five stocks in one's portfolio is most probably enough, as long as one can stand the volatility and have a sound exit strategy (Warren Buffett's partner, Charlie Munger, as previously stated that holding three stocks is enough if one can withstand the volatility). Anything over five stocks is probably overkill (there would not be enough time to keep track of all your investments and to think of new ones) for the typical retail investor.
I will now illustrate with an example. During the Franco-Prussia War from 1870 to 1871, Junius Morgan (the father of John Pierpont Morgan) rose to prominence by leading the financing for the French Government. Quoting from "The House of Morgan":
Junius's big chance for a state financing came in 1870, when the Prussians crushed French troops at Sedan in September, seized the emperor, Napoleon III, and laid siege to Paris. After a republic was proclaimed, French officials retreated to Tours and set up a provisional government. Otto von Bismarck, the Prussian chancellor, tried to isolate the French diplomatically. When they approached London for financing, he conducted a propaganda campaign, blustering that a victorious Germany would make France repudiate its debt.
Barings had already floated bonds for the Prussians, and the Rothschilds dismissed the French cause as "hopeless." Moreover, Mexico and Venezuela had recently defaulted on their debts, and no-one in London was in a mood to be venturesome at this point. The Morgan-led syndicate floated an issue of 10 million pounds (the equivalent of US$50 million at that point) at 85, or 15 points below par. Quoting from "The House of Morgan":
This sharp discount was designed to coax a skittish public into buying. The French felt blackmailed by these degrading terms, which they thought suitable for a Peru or Turkey. Yet Junius hadn't exaggerated the risks. After Paris fell in January 1871, followed by the Paris Commune, the bonds dropped from 80 to 55, and Junius desperately bought them to prop up the price, nearly wiping himself out. This was all very strange for a man who had urged caution on Pierpont: he was betting the future of his firm on one roll of the dice.
And yet, when the war ended, the French Government repaid the loan in full, netting Morgan 1.5 million pounds, or the equivalent of US$7.5 million at that point in time. From then on, Morgan's name would be mentioned with the likes of the Rothschilds and the Barings. No doubt, the venture of buying the bonds to prop up the price was a risky venture, but was Junius Morgan being a total gambler? In this signature deal, Junius Morgan would take a page from the Rothchilds' book - by using a fleet of carrier pigeons to ensure that Morgan would receive the most-up-date and the best information. Some of these pigeons were shot down and consumed by starving Parisians, but some of them actually got through the Channel. Morgan also engaged in extensive research of the history of the French government - stating that in the history of 12 French governments since 1789, "not one of these governments had ever repudiated or questioned the validity of any financial obligation contracted by any other. The continuing financial solidarity of France was unbroken." It would not be the first time that a Morgan - armed with more and better-quality information - has gotten the upper hand of the politicians, and nor will it be the last.
To summarize: Before one makes an investment or speculate, one should do as much research as possible, and since many things can go wrong in this world of ours - one should also minimize the number of real-world variables in the investment equation. Time is also a factor. The more long-term the investment horizon is, the riskier the investment. That is why having an exit strategy is as important (if not more important) than having a good "entry strategy." Finally, risks, in general, are not easily quantifiable - especially when it comes to investing in equities or in betting on certain outcomes. Given the concept of "Modern Portfolio Theory," the typical professor would have labeled Junius Morgan as a pure gambler, and the retail investor who bought the S&P 500 in March 2000 as an investor. This is the main reason why Long-Term Capital Management mainly focused on bond convergent trades in developed countries when it first began business - and it did a spectacular job at that as well, until the hedge fund started operating in the emerging markets and the equity markets as well. The valuation of a bond is more objective, as it is basically the net present value of its future payments - which is strictly defined, and further discounted by some sort of credit risk. With equities, however, anything goes (the valuation can be justified by many models). It also did not help that LTCM also engaged in directional trades towards the end of its life as well. The lesson of LTCM is that despite the many advances we have witnessed in the field of finance, risks are still not easily quantifiable - and hopefully, this author has argued a good case of my own concept and perception of risk - as well as how we could take advantage of them from an investment standpoint. The steps that I have illustrated have worked throughout financial history, and they will continue to work unless some person or hedge fund successfully develops a supercomputer which has the ability to model every economic decision made by every single person in this world - essentially an impossibility (see "reflexivity" and "chaos theory").
Let's now focus on the current stock and financial markets. From a risk standpoint, this author has previously discussed that one of my favorite trades is the mean-reversion trade. This is typically seen in the valuation of stocks. Buy when stocks are undervalued, or sell when they are overvalued (of course, this is easier said than done). However, mean-reversion can be witnessed in many indicators, including monetary indicators and the concept of monetary velocity (i.e. the "turnover" of money). In last week's commentary, we first introduced our "MarketThoughts Excess M" indicator (the "MEM indicator"). "Basically, it works like this: Whenever our MEM indicator declines below the zero line, there is a lack of "excess cash" in the domestic economy, and vice-versa. The "lack of excess cash" indicates that the turnover of money is high (also known as velocity) - indicating that investors are "risk seeking" as opposed to being risk adverse. This is important - as excess speculation in an economy makes it more prone to a slowdown or a recession, especially given that the bull cycles in commodities, real estate, and the stock market are now getting very mature." Please see our last week's commentary for a refresh on how our MEM Indicator works.
In a sense, our MEM Indicator (the 52-week growth rate of the adjusted monetary base minus the 52-growth rate of M-3 - all smoothed using their ten-week moving averages) gives us one of the purest measurements for financial velocity - even more so than the traditional velocity measurement of GDP/M-3 (or M-2). Why? Because when speculative activity in the financial world increases, this is directly reflected in the M-3 numbers. Monetary aggregates such as the adjusted monetary base and M-3 can be reasonably measured. The GDP number, meanwhile, is not a number that can be easily measured. In fact, the GDP number is only a guesstimate- and is subjected to government adjustments nearly all the time.
Last week, I stated that: "As one can see, our MEM indicator is currently negative and has been trending downward for some time. This means velocity has been increasing - and combined with a tightening Fed (see green line) and a tightening Bank of Japan (the Japanese monetary base is now hugging the flat line), I believe the markets (especially the commodity markets and real estate) are sitting on the edge right here . Of course, similar to the story of our MarketThoughts Global Diffusion Index, this does mean that the markets will fall tomorrow. What it does mean, however, is that until the Fed and the Bank of Japan "play loose" with the monetary base again, any further rallies in the commodity, real estate, and the stock markets will be extremely limited. On the contrary, the downside is potentially huge - even though the market is currently very oversold."
This author continues to stand by this statement. In fact, our MEM indicator has stretched further to the downside last week - from a reading of negative 3.44% to negative 3.49%. This means speculative activity is continuing to increase, despite the fact that the Federal Reserve has been tightening and despite another anticipate hike in the Fed Funds rate by 25 basis points this coming Tuesday. Following is a weekly chart showing the MEM (MarketThoughts Excess M) indicator vs. the St. Louis Adjusted Monetary Base vs. the Dow Industrials from February 1985 to the present:
From a mean-reversion standpoint, the continuing decline of both the monetary base and our MEM indicator does not look good for the bulls in the longer-run. This author believes that the downside risk is significantly more than the potential upside, and thus the stock market (as well as the commodity and real estate markets) does not make a good investment at this point in time.
Speaking from a risk assessment standpoint, this author would most probably not commit on the long side in a substantial way unless most of our technical indicators are lined up in our favor - that is, until most of our technical indicators become fully oversold. Over the last couple of weeks, we discussed overbought/oversold indicators such as the NYSE McClellan Summation Index, the NYSE ARMS Index, the equity put/call ratio, the Rydex Cash Flow Ratio, the total amount of margin debt outstanding, and so forth. This week, we will update our readers on the total amount of short interest outstanding on both the NYSE and the NASDAQ.
Following is a monthly chart showing the NYSE short interest vs. the Dow Jones Industrial Average from November 15, 2000 to October 14, 2005:
The headline story was that the short interest outstanding on the NYSE hit an all-time high on October 14, 2005 - increasing 85 million shares to 8.65 billion shares. This was interpreted in a bullish way by most commentators. However, what was not emphasized was that the recent increase in NYSE short interest was not that substantial - as the four-month increase in short interest is only 2.85%, compared to the record of 10.79% set on May 15, 2005. While the amount of short interest has no doubt increased over the last two weeks, it is most probably still a far cry from the record increase we saw earlier this year. Until this author gets a confirmation of a significant increase in short interest, we should still be playing it safe, for now.
Let's now discuss short interest date on the NASDAQ. Following is a monthly chart showing the NASDAQ short interest vs. the NASDAQ Composite from September 15, 1999 to October 14, 2005:
The short interest phenomenon in the NASDAQ is very similar to the story on the NYSE. For the month ending October 14, 2005, short interest made a record high of 5.94 billion shares. However, it is interesting to note that the increase in short interest on the NASDAQ has also not been too impressive - as the four-month growth rate is still relatively low at 1.87% (compared to over 10% for the four months ending April 15, 2005). Again, until this author gets confirmation of a further increase in the four-month growth rate, we should still be playing it safe, for now. That means no commitment on the long side in a substantial way until the market gets more oversold.
For the readers who are interested, I now want to discuss a little bit about U.S. government bonds. As most of you may know, the yield of the 30-year Treasury bond has gone up significantly since two months ago - from a yield of approximately 4.20% to today's 4.77%. Most traders have attributed the recent sell-off in bonds to a rising fear of inflation. This phenomenon has been witnessed in the yield of government bonds all throughout the world, with the Thailand government taking the top prize as its ten-year bond saw its yield rise form 4% to 6.2% since the middle of June. However, like I have mentioned before, higher-than-expected inflation is not a given going forward, primarily because of the following reasons:
- This author believes that the world's developed countries are still experiencing deflationary trends - as the world's excess labor pool continues to become more efficient and educated. The United States has recently witnessed another setback for the unions - with the bankruptcies of Delta Airlines, Northwest Airlines, Delphi, and the subsequent UAW negotiations with General Motors. Unions are inherently inflationary for wage growth, and unless all the workers of a particular trade around the world can organize under one union, there is good reason to believe that unions will no longer be relevant in the 21st century labor force going forward.
- The beginning of any global inflationary trend should first be observed in countries like China and India - both of whom are experiencing a huge boom in the consumption of many commodities. Since commodity prices have been skyrocketing, there is no doubt that China and India countries are the marginal countries in this regard. The fact that the Chinese government is pumping up the monetary base at 17% after a slowing down in the last 12 months suggests to me that inflation is not a concern in China at this point. Moreover, even if the Chinese are experiencing high inflation, there is no reason to believe that overall, this would lead to higher prices in Chinese imports - as price increases in certain Chinese goods will be offset by price decreases BECAUSE of certain new imports from China, such as the $5,000 car from Chery scheduled to show up in U.S. showrooms come 2007. Any new tariff legislations, however, will be highly inflationary - and ultimately, bad for the U.S. economy.
- Sure, the ECRI's Future Inflation Gauge hasn't been this high since June 2000. But it is interesting to note that inflationary pressures actually peaked at around June 2000, and that the Federal Reserve actually stopped hiking rates in the same month!
- Certain commodity prices have started to weaken, such as crude oil, steel, etc. If this upcoming winter is warmer-than-expected in the U.S., there is no doubt that natural gas prices will plunge below $10/MMBtu very quickly. Please bear in mind that natural gas supplies approximately 24% of all our energy needs - compared to 40% for crude oil. Any sustainable drop in the price of natural gas should have a significantly deflationary effect on the U.S. economy. As an aside, readers who are interested in daily update on the remaining amount of shut-in oil and gas production should check out the following EIA link.
- If one charts an advance/decline line of the components of the CPI, one would find that the A/D line of the CPI has actually been trending down.
- The newly appointed Federal Reserve Chairman, Ben Bernanke, has maintained that in terms of price stability, the United States and most of the developed world is currently in the "sweet spot" - a spot which took us more than 20 years to achieve. More important, he has claimed that only by staying at current levels will an explicit target rate for inflation be credible going forward. Hence, you can be sure that he will be very active in fighting both deflationary and inflationary trends going forward.
Moreover, the Rydex Bond Ratio (Rydex bearish assets on government bonds divided by Rydex bond assets) is at a point where bond yields have topped out in the past - with the exception of early 2004. This can be seen in the following chart:
Witness the recent topping out of the Rydex Bond Ratio in December 2005, March 2005, and August 2005 - and the subsequent decline in bond yields. For investors who are interested in buying long-dated government bonds here, however, please be careful as the Rydex Bond Ratio has been very volatile over the last few weeks (editor's note: please remember that this should not be interpreted as investment advice). That being said, this author believes that the recent rise in government bond yields may be coming to an end very soon.
Let's now discuss the most recent action in the stock market by first turning to the chart showing the most recent daily action of the Dow Industrials vs. the Dow Transports. In retrospect, the decision to cover our 25% short position in our DJIA Timing System the Friday before last was near perfect timing. Our next signal will most probably be a signal to go long, but with the market never getting into a fully oversold situation in the latest decline, it is difficult to see how sustainable any upcoming rally will be. For now, we will just be cautious and wait:
In retrospect, the persistent refusal of the Dow Transports over the last month or so to close below its September 20th closing low was a bullish signal - with the ball now in the bulls' court at least for now. Bears who are still short the market should now be very selective and careful. The theme over the last two years remains the same: The ability to pick good stocks to go long or go short (as opposed to trend-following in the general market) is still very paramount to out-sized performance for the foreseeable future.
Let's now discuss our most popular sentiment indicators - starting with the Bulls-Bears% differential readings in the American Association of Individual Investors Survey vs. the Dow Industrials. The latest weekly reading decreased from 7% to a highly oversold weekly reading of negative 14% - the most oversold reading since early May. Meanwhile the ten-week moving average edged down further from 4.0% to 3.7%, although this author would like to see a more oversold reading (similar to the negative 9.8% reading in early May this year and negative 15.8% reading in March 2003) before substantially committing on the long side:
For now, let's just say this author is still very cautious. The huge decline last Thursday and subsequent rally last Friday whipsawed many investors (and market-timers) and so I would not be surprised if emotions again get the better of them at least early this week. For now, this author will continue to reevaluate each passing day - so don't be surprised if we implement a new signal as early as this week or next week.
The Bulls-Bears% Differential in the Investors Intelligence Survey, again got slightly more oversold this week - with the latest weekly reading declining from 15.8% to 15.6% - again, the most oversold reading since early May of this year. The four-week moving average declined from 20.2% to 17.4%:
As the above chart mentioned, the four-week moving average of the bulls-bears% differential in the Investors Intelligence Survey is now at 17.4%, which is at our "optimal oversold reading" of 20% or below. However, readers should note that many of our other technical indicators are still not confirming this oversold reading, and thus being the cautious investor that I am, we should continue to hold off here. Again, while I have mentioned in the last few weeks that we may go 100% long in our DJIA Timing System within a short amount of time, it is not certain that we will, especially given the unpredictable nature of the stock market. A potential variation of this may be a 50% long position within the next couple of weeks. We will just have to see.
As for the Market Vane's Bullish Consensus, the most recent weekly reading has bounced back up slightly from 57% to 60% - potentially signaling a reversal to the upside:
Please note that last week's reading was only one percentage point away from the oversold reading in mid-August 2004. Meanwhile, the latest four-week moving average also declined from 60.0% to 59.3%, the most oversold reading since late August 2004. Like I have mentioned before, the burden of proof is now resting on the bears' shoulders. However, this author is still somewhat hesitant about going long at this point, as this author does not like to chase rallies - especially given the fact that this rally, similar to the many rallies over the last two years - are also most probably not too sustainable, given:
- The lack of a fully oversold condition in the latest market decline;
- The recent, significant decrease in our MEM indicator - suggesting that speculative activity has gotten more intense despite the Fed's continuing tightening.
That being said, if the market resumes its decline here and if the upcoming decline renders the market more oversold than we have seen over the past two years, then this author would not hesitant going long at that point.
Conclusion: Given the 5000-year recorded history of humankind, it is amazing to note that our understanding of modern finance and our concept of risk is still in its infancy. Consider the fact that our modern perception of probability theory did not emerge until the 17th century, and that other theories such as the Modern Portfolio Theory did not begin to be studied until the early 1950s. Please also consider that the concept of securitization really did not become popular until the 1980s. As the 21st century progresses, the ability to hedge real-life risks and trade them will continue to increase, and new, more sophisticated tools will be used to value these instruments. This author has argued that today's most sophisticated tools are nowhere near adequate to measure the risks inherent in such financial instruments as equities, as many real-world risks are still impossible to quantify.
As stock market participants, it is imperative that we are able to put the risks in perspective in each of our investments. Like I have mentioned above, the ability to quantify risk is not totally dependent on one's IQ - even though there is somewhat of a correlation. In a complex world and in today's information age, it is the ability to simplify each potential investment and filter out all the unimportant information - as well as minimize the number of real-world variables that we need to guard against. Blind diversification by buying the S&P 500, for example, does not minimize your risks in any way if one has not done the necessary research.
Finally, our long-term monetary and velocity indicators are still calling for lower prices ahead - especially for the commodity and real estate markets, since they have received bulk of the attention (liquidity and speculation) in the last few years. For now, whether the market will continue its rally from last Friday is still not certain, but unless the market chooses to retest its lows from last week, this author would most likely not chase any rallies going forward. Again, this comment does not apply if one is a good individual stock picker. For now, we will continue to remain neutral here, but will definitely inform our readers via our "special alerts" emails once we change our minds.
Henry K. To, CFA