Anatomy of a Crash
If you happened to be in the equity markets during 1987, at least on the long side, there is little doubt that the stock market crash that October cost you some money. In the days that followed one economist after another popped out of their ivory towers to announce that North America, if not the entire world, was heading into a depression.
Since we can't recall the Great Depression of 1988 we suspect that the cause of the 'crash' was something other than what most suspected then and, most likely, what most understand even now.
Perhaps the most significant intermarket reason for the 1987 stock mar-ket crash was runaway weakness in the U.S. dollar. However, the best place to 'see' the build up is through a comparison of the U.S. 30-year T-Bond futures to the Dow Jones Industrial Index. At top right we show, once again, this comparative chart to highlight the fact that stock prices were rising even as interest rates moved inexorably higher.
But...to really understand how this evolved we need to step back in time even further. Why were bond prices rising in the first place? The answer has much to do with the impact on interest rates by energy prices.
The oil price collapse at the end of 1985 and through into 1986 is often called the 'Third Oil Shock'. With only a marginal increase in OPEC supply the price of West Texas Intermediate crude fell from $31.75 in November to $10- with some Persian Gulf cargoes selling for as little as $6. Daniel Yergin, in his Pulitzer Prize winning book, 'The Prize'' docu-ments the story of one Texas Exxon station that offered gasoline for free one day in April of 1986, attracting a line up that extended for six miles.
On an intermarket basis we can see the relationship between T-Bill yields (the T-Bill futures chart at right has been scaled upside down so that a falling line depicts lower short-term interest rates) and crude oil prices. Oddly enough, even though the oil industry was caught most flat-footed by the price collapse, the charts show that this had been building from the middle of 1984.
Our point here is that the stock market crash was a function of a series of events that extended back for years. Even if we start in 1984 we can see that short-term interest rates were forecasting a collapse in energy prices that would 'bubble' bond prices higher into 1986 and then push liquidity into the stock market through into 1987. As crude oil prices turned higher so did interest rates. As interest rates turned higher the stage was set for both the stock market's bubble and its eventual col-lapse. When the dust cleared...it was business as usual- expect for the poor souls who had been blindsided by the markets into the crash and by the economists' forecasts afterwards.
Backing up even further we show, at top right, one of our favorite charts to highlight the fact that the entire series of events that created the stock market's debacle in 1987 could actually be traced back as far as the bond market's bottom in 1981- more than 12 months before the equity markets turned higher in August of 1982.
As then Fed Chairman Paul Volcker squeezed the life out of the infla- tionary cycle that had built through the previous decade, short-term interest rates peaked in early 1981 and then began to recede as the U.S. economy headed back into recession. As interest rates fell...the stock prices of those companies that are most sensitive to changes in interest rates- names the banks and financials- began to outperform. The chart show the ratio of JP Morgan's stock price to that of copper producer Phelps Dodge and contrasts it with the trend for short-term yields (ris- ing T-Bill prices denote falling interest rates).
If nature abhors a vacuum, the financial markets seem to like to create divergences and then close them up- often with devastating results. The bubble created from 1981 into the end of 1986 pushed bank share values well above those of the commodity cyclicals and then, to the surprise of most, this divergence was wiped clean by the combination of rising interest rates- coming on the heels of the post-crash recovery in the energy markets- and escalating commodity prices.
What we find most fascinating about this chart, however, is the way it acted right through the stock market's crash. The JPM/PD ratio squeezed lower into the crash, through the crash, and for months after the crash. From that point of view one could argue that the 1000 point gain in the DJII into the summer of 1987 and the 1000 point decline during the autumn was more of a result than a cause.
In any event, if we were to look for something similar- even in a general sort of way- in the latter half of 2002, we would need to find a number of basic intermarket similarities to the 1981- 87 time period. Let's see what we can see.
First, is there or has there been a 'bubble' in the financial sector created by a prolonged decline in short-term interest rates? The chart at middle right of the JPM/PD ratio and U.S. T-Bill futures prices argues quite nicely that there has...and that it began in 1995.
Second, has there been a more recent energy-oriented price collapse that helped lift the bubble to its final crescendo? At bottom right we show natural gas futures and T-Bill futures (scaled upside down once again). Notice that the concurrent with the decline in short-term interest rates from over 6% to under 2%- something largely ascribed to the excesses created in the tech and telecom sectors- natural gas prices fell by some 80%. Given that oil prices declined by about 70% in early 1986, perhaps this is just about right.
So...we have a 7-year bubble in the financial sector vis-a-vis the com- modity cyclicals and a late-cycle collapse in energy prices that helped to drive short-term interest rates lower. The only thing that is missing if we are to truly replicate the 1987 scenario is a rebound in energy prices- perhaps around the beginning of the year- that creates upward pres- sure on interest rates.
In fact, the chart does suggest that we are already close to 4 months into this process so a literal interpretation would call for continued strength into the late summer with fire works through into the autumn. The problem is...this time might be just enough different from 1987 to catch us all unawares.
What exactly happened at the start of 1995 to justify not only the stock market's rapid rise but the protracted decline in short-term interest rates that led to the valuation differential between the financials and true cyclicals? The answer lies in the foreign exchange markets.
In early 1995 the U.S.dollar hit a low against not only the euro but also the Japanese yen. Japan's economy began to buckle under the weight of what was to become a move towards deflation and its currency began to trend lower. At top right we show the euro futures and the S&P 500 Index. As the dollar turned higher, so did the equity markets but underneath all of this was the ongoing disinflationary trend.
At present the U.S. dollar is starting to weaken off. At middle right we show the euro futures and the S&P 500 Index futures over a shorter time frame. One intermarket conclusion would be that the first problem won't necessarily be in the stock market- much like 1987's pressure was largely a function of what was going on in the bond market- but rather in the currency markets. In other words, while stocks chop back and forth we could easily witness a 'crash' in the dollar with the euro shooting right through parity and heading, perhaps, towards around 1.08.
Again, from an intermarket stand point, does this make any sense? Of course it does!
The energy price collapse along with the series of implosions in the tech and telecom sectors, set the stage for dramatic action by the U.S. Federal Reserve. What is less obvious is the impact that this created on the spread between short-term interest rates from one country to an-other. At bottom right we show the Japanese yen futures and the spread between short-term Japanese interest rates (euroyen futures) and short-term U.S. rates (T-Bill futures). The spread has fallen from close to 6% in 2000 to somewhere near 1.5% today. As the spread was tightening the yen fell against the dollar as capital moved from Japan into the U.S. Now that the spread has steadied out, the flow of capital has abated and the yen is starting to rebound- much like a beach ball pushed under the water and then released.
So, what does all this mean?
The charts suggest that the U.S. dollar is in trouble unless or until short-term U.S. interest rates rise back to a point where speculative inflows help to steady the currency. It is almost a given that the Fed will not only raise rates this year but will do so at a pace that virtually no one currently believes possible.
In the meantime, of course, the U.S. dollar's decline will impact other markets in other ways. Gold and other commodity prices, for instance, tend to rise when the dollar falls. The most likely impact of a falling dollar would therefore be upward pressure on short-term interest rates as well as upward pressure on commodity prices. Given that we still have a fairly major divergence between the financials and commodity cyclicals, there is some logic- and indeed some elegance- to this sce-nario.
One of the corner stone intermarket relationships that we use is the 2- year lag between the bond and commodity markets. In short, a rapid decline in interest rates today would tend to create an equally rapid increase in the major commodity indices two years from now. For the purposes of this analysis note that the drop in rates that coincided with the collapse in energy prices at the end of 1985 helped create the com-modity price rise through 1987 and 1988 that helped drive interest rates and stock prices into the stock market crash. In much the same way the bond market rally that began in early 2000 was both confirmed and accelerated- at least in the short end of the yield curve- by the combina- tion of the tech/telecom weakness and the collapse in energy prices through 2001. In other words, the intermarket process that will lead to a collapse by the financials relative to the commodity cyclicals has al- ready begun.
At top right we show the CRB Index futures and the U.S. T-Bond fu- tures from the end of 1986 through 1987. It wasn't just oil prices, of course, that created stress in 1987- the entire commodity sector turned higher as bond prices turned lower.
The most obvious relationship at the moment would be the inverse move between the U.S. dollar and commodity prices. We may or may not get a rapid decline in long-term bonds but the most likely outcome here is a rapid and unrelenting decline in the external value of the dollar- until the situation overwhelms the 'inflation isn't a problem' psychol- ogy that permeates both 'the street' and the U.S. Federal Reserve. In other words, if there is one market that we can be 'flat out, no holds barred' negative on it is the dollar.
The financial markets tend to raise so much dust on a day-to-day basis that it is often difficult to see the forest for the trees. One of the advan- tages of studying intermarket trends is the acquisition of perspective. Below we show the S&P 500 Index in terms of both gold and the Japa- nese yen. We can 'see' the long-term trend of the S&P 500 Index in terms of the yen as well as relative to the price of gold and note not only what happened when the yen turned down against the dollar but also its impact on the relative price of two financial assets- stocks and gold. The charts convey the message that these markets are now well on their way to moving back into balance.
At bottom right we wrap things up with a ratio chart of the Dow Jones Industrial Index and the S&P 500 Index. While there are divergences and stresses between the currency, commodity, fixed income, and eq- uity markets, there are also problems within the equity markets. The DJII has been pushed upwards vis-a-vis the broader S&P 500 Index as money flowed has flowed away from 'growth' and back towards 'safety' or 'value'. Previous peaks in this ratio have preceded the 1987 crash, the 1990 bear market, and the 1997 collapse of Hong Kong's Hang Seng Index.
Before going any further perhaps we should circle back and try to put this all into perspective. We are arguing that the markets are moving along an intermarket path that started when the dollar began to rise in 1995 while interest rates fell. From an intermarket perspective the perfor- mance by the financial sector was more a direct cause of this trend while the action in the techs and telecoms might be described as an unin- tended consequence. After all, the only difference between a trend and a ridiculous bubble is the amount of capital that is made available.
In any event.. telecom problems are helping to hold short-term interest rates down even as other intermarket forces work to push them higher. While the Fed mumbles about the risks to the economy in the future, the markets are working hard to turn a relatively benign situation into a crisis. We could argue that it wouldn't make any difference either way- that the market's destiny has already been set- but that would take all the fun out of this.
At top right we show the Nasdaq 100 Index/S&P Retail Index ratio and contrast it with the trend for short-term yields (T-Bill prices scaled up- side down). This chart argues that the tech and telecom sectors are going to have to bottom and start to recovery- at least relative to the Retail sector- before short-term interest rates will be free to rise.
In terms of gold, we are still trying to figure out of if a target near $350 is appropriate or if something closer to $400 is going to be more accurate. At middle right we show the S&P 500 Index (scaled upside down) and the gold futures. If 1995 was the pivot for both markets then perhaps something over $400 is indeed appropriate.
Note, however, that in our work the absolute level for gold is less impor- tant than some of the relative levels. We would like to see the ratio between the Philadelphia Gold and Silver Index (currently around 82) and gold move down to about 3:1 before suggesting that a top has been put in. This argues that there is both time and price gains ahead.
At bottom right is a chart of Inco (nickel producer and former DJII component) from Canada's Toronto Stock Exchange and the DJII (in red).
Frankly, we believe that this is one of the most powerful charts that we can show- not because of what it might say about the base metals and the DJII, but rather for what it tells us about the power of an underlying trend.
The chart covers the period from the end of 1986 through towards the end of 1988. We have shown that the decline in interest rates through into the end of 1986 should have created a subsequent rise in commod- ity prices through into the tail end of 1988- which is exactly what hap- pened. In the midst of all this, of course, was the stock market 'crash'. The stock price of Inco rose with both the broad equity markets and the commodity markets, crashed back to the level that it started off 1987 at, and then accelerated higher once again. By mid-1988 while the econo- mists were still trying to figure out when the depression would start, Inco's stock price was 35% HIGHER than it's pre-crash peak. If we have a similar sort of equity markets problem some time later this year, at least we can take some comfort from the notion that the commodity markets cycle is going to remain positive well into next year.