In the mania leading up to the final stock-market high in March 2000, investors exhibited “irrational exuberance.” This extreme, emotional frenzy ended badly for them, with the S&P 500 index dropping an extraordinary 50% and the NASDAQ composite index plunging a staggering 75% thereafter.

Now, as the current “echo mania” comes to a close, investors have been exhibiting “irrational complacency,” an unwillingness to act on the increasingly negative warning signs about the U.S. economy and stock market, which is an affliction that history and logic shows will end at least equally badly for them, and probably much worse. See our April/May 2006 commentary for our discussion of how and why the second downleg of a Supercycle bear market[1] is typically even more severe than the first downleg.

Revised Government Data Confirm Our Recession Calls

At the mania peak of the previous Supercycle bull market period[2], we issued our April 2000 Market Commentary, in which Bob Bronson, principal of Bronson Capital Markets Research, our investment strategist, issued a forecast of an imminent drop in the stock market, which would be followed six to nine months later by a recession. In our commentaries in late 2000, we confirmed our view that a recession was developing. And in our April 2001 commentary, we dated the beginning of the business-cycle contraction as October 2000.

In November 2001, the National Bureau of Economic Research (NBER), in hindsight, chose a later date, March 2001, as the start of the recession. In our February/March 2003 commentary, we reported that the NBER subsequently revisited that decision and discussed an earlier date for the start of the recession, but declined to change it mainly because they had never changed a date for any previous recession.

A succession of revised government data released by the Bureau of Economic Analysis (BEA) now decisively confirms that the business-cycle contraction started in the third quarter of 2000, earlier than the NBER determined in hindsight, but entirely consistent with the forecast Bob Bronson had made in early 2000. Further, the revised data show that instead of just one quarter of negative economic activity during the recession, in the third quarter of 2001, as the BEA initially reported (see the black line in the chart below), there were three negative quarters: the third quarter of 2000 and the first and third quarters of 2001, as seen in the blue line in the chart below of real (inflation-adjusted) gross domestic product (GDP). Overall economic activity as measured by GDP over a period of more than two years has been revised significantly lower, confirming a longer and deeper recession than originally reported by the BEA, as Bob Bronson had anticipated.

When the NBER first determined in November 2001 on a look-back basis that the recession had started eight months earlier in March 2001, the BEA data at their disposal did not show any quarters of negative real economic growth in 2000 or 2001. By the time the NBER determined in July 2003 that the recession had only been eight months long and had ended in November 2001, they were using revised BEA data that showed three consecutive quarters of negative real GDP during the recession: the first through the third quarters of 2001. Since then, the BEA has further successively revised the real GDP data. As of today, the BEA data indicates three, non-consecutive quarters of negative GDP, as shown in the chart on page 2.

However, using another data series, which the government does not directly compute, but which Bob Bronson derives from the government’s GDP data, even more periods of negative growth emerge. “Real private domestic final sales per capita” shows a total of five quarters of negative economic growth: the third quarter of 2000 and the first and third quarters of 2001, as in the BEA data, but with the addition of the first and fourth quarters of 2002 (see the light blue line in the chart below). Note that three of these quarters are before or after the dates designated by the NBER for the last recession, again confirming it was more severe and lasted longer, with an earlier start and later end, than the NBER’s “official” declarations.

As we discussed in our Summer 2006 commentary, real private domestic final sales per capita may well be the single most important indicator of the true business cycle that can be derived from GDP data, since it excludes components in the government’s GDP calculations that are not truly reflective of the underlying business cycle. These components are: 1) price inflation, which does not reflect the real quantity economy; 2) government spending, which often runs counter to the business cycle because it ramps up when policymakers are attempting, almost always too late, to mitigate an economic slowdown; 3) foreign trade, which may or may not coincide with the U.S. business cycle; 4) private inventories, since excessive inventories, which add to GDP, are bearish, and inventory shortages are not necessarily bullish; and 5) population growth, which is completely independent of the business cycle. While these adjustments may appear to be large, “real private domestic final sales” have constituted 83% of real GDP since 1990, so that the vast bulk of the GDP data is being used, while elements perverse to identifying the true business cycle are eliminated.

Note that the chart on page 3 shows that this important indicator of the business cycle has been rolling over for the past 10 quarters (30 months) since its peak in the third quarter of 2003, which can be seen in the computer-generated best-fit trend line (the blue dashed line). The BEA data for the second quarter of 2006, the most recent data available, show that the cyclical trend for this indicator is now accelerating to the downside.

Further, we expect that the BEA will adjust recent GDP data progressively lower over time, similar to the data on the GDP chart on page 2, eventually revealing that the true business cycle is deteriorating at a faster rate than is currently being reported.

The additional quarters of negative GDP growth that occurred after the NBER-designated recession of March through November 2001, according to Bob Bronson’s work, are also supported by the widely recognized Conference Board’s composite coincident indicator. Ironically, this indicator is comprised of the same four economic components the NBER uses to determine the starting and ending dates of recessions. However, as seen in the chart on page 5, the Conference Board data confirm both an earlier start to the recession (September 2000) than the NBER designation, as well as a significant, second dip in economic activity after the NBER-designated end of the recession, supporting Bob Bronson’s forecasts and analysis during that entire time. Bob had used this indicator, along with many other factors, to support his initial forecast in March 2000 of a recession later that year.

The stock market also confirms this view of the economy. Stocks peaked in March 2000, two quarters ahead of the now-recognized start of negative GDP, which is the typical interval and just as we anticipated. The later, second dip into negative GDP in 2002-03, seen in the chart below, explains why the stock market made significant new lows at that time, a full 11 months after the NBER-declared end of the recession in November 2001.

Stock-market lows have only occurred after the end of a recession twice in the past 152 years. The other time was in 1921, when the stock market made its low in August, just one month after the July designated end of the recession. This time, however, stock market lows occurred an astonishing 11 months and 16 months after the NBER’s November 2001 designated end of the recession, as the market double- bottomed in October 2002 and March 2003. This further supports our view that the NBER erred in both their starting and ending dates for the recession.

Rather than relying just on quarterly GDP data, the NBER also takes monthly GDP data into consideration. Bob Bronson has adjusted that monthly data to remove the upward bias of annual population growth of about 1% per year in order to normalize economic growth, arriving at “monthly GDP per capita,” illustrated in the chart below. This is a standard procedure for economic historians and international economists.

The plot of monthly GDP per capita reveals a 35-month period of no growth in real GDP from June 2000 through May 2003 (see the red box in the chart on page 5). This explains the widespread feeling by Americans during that time that they were not participating in the economic “recovery” the government said was underway.

Employment data, one of the four areas the NBER considers in dating recessions, also supports our view that the recessionary conditions lasted longer than the NBER has thus far acknowledged. The term “jobless recovery” was a useful claim for political purposes, but never had any economic validity. In fact, the contraction in employment extended all the way into the first quarter of 2003, a full 14 months past the designated end of the recession. When employment finally bottomed, the stock market appropriately double-bottomed along with it.

Finally, there is even credible evidence that the last business-cycle contraction never fully ended, but rather has only paused, and that the incipient recession will be another major dip in one long, ongoing business-cycle contraction.[3] This would be consistent with the lengthy down-up-down, ABC pattern that constitutes a Supercycle bear market, which we discussed in some detail in our April/May 2006 commentary. At the very least, the chart below from the Federal Reserve Bank of St. Louis shows that the economy has consistently performed below its potential since 2001. Potential GDP is computed by using something called the Taylor Rule, which is a formula based on real GDP, inflation, and the fed funds[4] rate. This chart casts serious doubt on whether the U.S. was ever in a sustainable expansion at all.

Bob Bronson has been supplying his analysis of this additional data to members of the NBER Dating Committee in discussions with them, suggesting they reconsider both the starting and ending dates of the last recession.[5] The chairman of that committee has now personally assured Bob, in writing, that the committee will review their findings. The committee’s decisions concerning the last recession are now so glaringly at odds with widely available economic data, particularly the revised government data on GDP, that they may make revisions, and Bob has been supplying them with some of the more compelling reasons why they should.

With all the data revisions and other strong analytic support for a change in the dates of the last recession, it is obvious that the previous recession was worse in both duration and magnitude than previously acknowledged, just as Bob Bronson had forecasted. Current economic data, as well as history, also suggest that the incipient recession will be even longer and deeper than the last one since, among other recessionary factors, it will be associated with the more severe, second downleg of a Supercycle bear market. It certainly will be far more severe than the current consensus of economists, and far more severe than investors have yet priced into stocks.

New Highs In The Dow Are Not Bullish

With the financial media’s preoccupation with the Dow Jones Industrial Average (DJIA) making successive, new all-time highs and breaking the round number of 12,000 today, you would think that such events signify something important and something terribly bullish. They don’t.

In fact, recent all-time record highs in the Dow bear an uncanny resemblance to the extremely negative chain of events set in motion 33 years ago – that is, at a similar point in the last Supercycle bear market period. Bob Bronson, who already had seven years of investment research and management experience under his belt by that time, remembers it well. The media’s current drive to sensationalize the Dow’s performance is being fed by TV talking heads who do not know, do not recall, and/or do not want to tell you “the rest of the story.” Here it is.

The 30-stock DJIA had reached an all-time high of 1,000 on February 9, 1966. It then declined for a period of time, following which it rebounded (the “B” in the ABC pattern of the Supercycle bear market) in what would eventually come to be seen as a bull trap[6], crossing the 1,000-mark again on November 10, 1972, six years and nine months after it first reached that record high.

The media made much out of this “psychological threshold” being crossed, which encouraged momentum-driven, short-term traders and longer-term (but unsophisticated) investors to pile in. This is as opposed to value-oriented institutional investors, who more prudently buy low and sell high. The momentum investors also did not want to miss the six-month period of relative strength in stocks that typically starts in October, but ignored the historical fact and logic idea that such relative strength only starts from a stock-market low at the end of a significant sell-off, not from the fundamental overvaluation at a stock-market high. Essentially defensive investors adopted an irrational, “defensively aggressive” strategy, indiscriminately buying the relatively safer, more conservative blue-chip stocks found especially in the winning, “safe haven” DJIA, which pushed the average higher.

As a result, the Dow made a series of successive, marginal new highs for eight weeks until it finally peaked at 1,053 on January 11, 1973. This was a relatively modest 5% gain over the 1966 peak almost seven years earlier.

At the end of that eight-week, classic blue-chip blow-off, the Dow started to plunge in the devastating second downleg (the “C” in the ABC pattern) of that Super-cycle bear market, ultimately declining 46%. The overall stock market dropped 49%, the biggest decline since the catastrophic Supercycle bear market that triggered the Great Depression in the previous Supercycle bear market period. Many older Americans vividly remember that their investments were essentially wiped out during that 1973-74 decline. The whole irrational, new-highs affair proved to be no more than a relatively short-lived sucker’s rally at the end of a lengthy bull trap.

Predictably, the hoped-for six-month strong season for stocks failed. Rather than advancing to a typical May high, the stock market declined and broke its October 1972 lows by early May instead. The over-the-counter (OTC) market, predecessor to today’s high-growth, technology-laden NASDAQ index, was true to form in leading the rest of the stock market lower, breaking its previous October lows three months sooner, in early February 1973.

In many important ways, what is occurring at the present time is a near-perfect parallel with this earlier period. Just like some three decades ago, unsophisticated investors are piling into the defensive, blue-chip stocks found in the Dow as the average hits record highs, once again hoping that a six-month period of relative strength will somehow start from the fundamental overvaluation found at a stock-market high.

However, we expect a similar failure of the typically strong six-month season to occur much earlier this time because the incipient recession is already at least a year ahead when compared to the earlier period, in which a recession started in November 1973, 10 months after the final Dow high. In fact, since August, September and October 2006 have bucked their typical patterns of declines and posted individual and cumulative gains, we believe that any typical, seasonal strength in stocks is largely used up and, therefore, that this final sucker’s rally at the end of a bull trap will end before the end of the year.

The Dow Is A Defective And Misleading Measure Of The Stock Market

That the Dow Jones Industrial Average (DJIA) is making all-time new highs at all is primarily because of the particular selection of stocks in the average, as well as the odd way the index is constructed. Other stock-market indexes are weighted by their equity capitalization (that is, the number of outstanding shares of common stock multiplied by the price per share), so that the largest companies carry the most weight. The Dow Jones series of averages are weighted by price. The decision to weight by price – initially just a simple averaging of stock prices – was made in 1896 when the original 12-stock Dow Jones Industrial Average was created and has never been changed, even though it leads to some strange consequences.

For example, even though IBM and Intel have similar equity capitalizations, and Intel has daily trading dollar volume that is several times larger than IBM, IBM has been perpetually given about four times greater weight than Intel in the DJIA, simply because of the weighting based on price that each was given on the day it was added to the index.

Further, with the DJIA making all-time highs, you might think that many, if not most, of its 30 component stocks would also be making all-time highs. Not so. Like all stock indexes, no matter how they are weighted, an increasingly smaller number of stocks hitting new highs can cause the index to register a new high.

Only four Dow stocks have hit record highs since the blue-chip blow-off started on September 26. No more than two have made new highs on any of the 13 days the DJIA recently hit record highs. Only one Dow stock hit a new, record high today when the Dow broke through 12,000. This extremely narrow new-high leadership is very bearish, both shorter- and longer-term.

On average, the stocks in the DJIA are still 27% below their individual all-time highs, as seen in the table on page 10. The S&P 500 index is still 12% below its record high, and the NASDAQ composite index is still a whopping 61% below its record high, both set in March 2000. Other broad stock-market indexes are also still significantly below their previous record highs, which belies any bullish significance attached to the Dow’s recent record highs.

Just as the DJIA is not an accurate gauge of the performance of the individual stocks in it, it is also not an accurate gauge of the overall industrial sector of the U.S. economy. There are nine commonly accepted sectors in the stock market,[7] of which the industrial sector is one. As seen in the upper panel of the chart below, the industrial sector has not even broken above its previous May high, much less its March 2000 all-time high. This technical divergence between the DJIA and the industrial sector of the stock market it purports to represent acts as a negative divergence for the whole stock market.

As seen at all market tops, fewer and fewer stocks are accounting for all of the advance of the stock market. Notice in the two lower panels of the chart above that neither the advance-decline line (number of advancing to declining stocks) nor the all-important advance-decline volume line (trading volume of advancing to declining stocks) are likely to break above the their May peaks. This type of negative divergence can be found in almost every sector of the stock market and is very bearish for both the shorter- and longer-term.

Declining Oil Prices Are A Reliable Indicator Of A Recession

On a final note, we have heard much in the financial media about how declining oil prices are bullish for the economy, since dollars previously spent on energy costs could be spent on other things. However, history shows that oil prices had peaked and were declining by the start of every recession during at least the past 60 years. Rather than triggering demand, a decline in oil prices is a reflection of slowing demand.

Notwithstanding a bounce, or partial retracement rally, that we expect is just starting for the energy stocks, the chart below shows that the prices of oil and natural gas have already peaked (see the blue line in the upper panel) and declined through the end of September, consistent with an incipient recession. This is similar to the peak in oil prices in late 2000 (see the blue line in the lower panel), which preceded the business-cycle contraction that is now recognized as having begun in the third quarter of 2000.

October 18, 2006

Prepared for
Robert E. Bronson, III, Principal
Bronson Capital Markets Research

FOOTNOTES

[1] A Supercycle bear market is the biggest sequence of bear markets during a Supercycle bear market period. For example, during the previous Supercycle bear market period from 1965-82, the interim Supercycle bear market lasted for the six years from the high point in December 1968 to the low point in December 1974. The latest one started on October 7, 1997, when money market fund returns started to sustainably outperform stocks, and exactly when we issued our call for it.

[2] A Bronson Asset Allocation Cycles (BAAC) Supercycle bear market period is a 12- to 20-year period of underperformance during which bear markets, anticipating economic recessions, as well as the recessions themselves typically are at least twice as frequent and twice as severe in magnitude and duration as during Supercycle bull market periods. Such a period begins when the return on money market funds sustainably exceeds the total return on equities, especially when downside-volatility-risk is taken into account. Further details are found in our research paper, “Bronson Asset Allocation Cycles,” available on request.

[3] As two four-year Kitchin Cycles,it would constitute one eight-year Juglar Cycle and the downtrend in the Kuznets Cycle. See Bob’s SMECT model: http://www.financialsense.com/editorials/bronson/model.html/

[4] The fed funds rate is the interest rate, set by the Federal Reserve, at which banks lend to each other overnight.

[5] This is reminiscent of technical conversations Bob Bronson had six years ago with the senior staff at the Federal Reserve Bank responsible for compiling the household debt service burden ratio about adding auto leases and security margin debt to their computations to more accurately reflect the magnitude of consumers’ debt problems. The Fed subsequently added auto leases to their calculations, but not yet security margin debt.

[6] A bull trap is a period of modestly higher highs in a fundamentally overvalued market that lures investors back into the market, only to be followed by a new or resumed market decline. A bull trap is baited by investor emotion from hope, greed, or complacency, which leads investors to pile into the market despite all fundamental warning signs to the contrary. Bull traps occur at major market tops. The bull traps in 1972-73 and 1999-00, which ended in classic blue-chip blow-offs, were each followed by market declines of 50%, and even more in equally-weighted stock indexes and the average equity mutual fund.

[7] The nine commonly accepted sectors in the stock market are: consumer discretionary (consumer cyclical), consumer non-discretionary (consumer staples), energy, financial, health care, industrial, (basic) materials, technology, and utilities.

© 2006 Bob Bronson
Editorial Archive

Contact Information
Bob Bronson
Bronson Capital Markets Research
Email