Comment 70: U. S. Stock Market(5). Valuable lessons from the 2000-2001 recession (July 20, 2009)

Dow Jones Industrial Average (abbreviated as DOW) has rallied strongly from its intraday low set on March 9, 2009. Many Wall Street experts are claiming that this rally is the first leg of a new bull market, and the March 9 intraday low will hold. In Comment 67 it has been pointed out that to justify this rally, the current recession needs to be a “V” shaped one, meaning that the economy must recover strongly once the recession ends. To see how a “V” shaped recession should look like, annualized real GDP values from the 4-th quarter of 2007 to the first quarter of 2009 are plotted as blue horizontal bars in the first graph at right. To make the recession an ideal “V” shaped one, two red bars are added for the second and the third quarters of 2009 respectively; real GDP data for those two quarters are not available yet at the time of this writing. The red numbers in the graph indicate that the real GDP of the second quarter of 2009 should grow at an annualized rate of +5.7%, and that of the third quarter of 2009 +6.6% in order to make this recession an ideal “V” shaped one. Such stellar growths of real GDP are not impossible but unlikely. What if the growth of real GDP is substantially slower than as suggested by the red bars in the first graph? Let us consider a case that the real GDP rises to the level of the first (the lower) red bar shown in the graph in the first quarter of 2010 instead of the second quarter of 2009, and to see what happens. Under this scenario the comparison between the first quarter of 2010 real GDP and the first quarter of 2009 value shows a +1.4% growth. This +1.4% growth in the year-to-year comparison is enough to declare the end of this recession according to the definition of recessions adopted in article 12. However, the definition of article 12 will set the time of the end of the recession at the third quarter of 2009, not in the first quarter of 2010 since a year-to-year comparison is looking at the average condition of two quarters ago. The scenario under consideration that puts the end of this recession at the third quarter of 2090 is in line with the original projection of The Federal Reserve Board, though their latest estimates seem to have delayed the end of the recession to the fourth quarter of 2009 and even to the first quarter of 2010. We also believe that the most rosy case is for this recession to end in the third quarter of 2009 as considered in the above scenario. Even under the most rosy scenario, this recession will not be a “V” shaped one, but is a broken “V” that we define as a “U” shaped recession. This raises the question of what is going to happen to the current rally, is it just a powerful bear market rally or not? Before we jump into any subjective conclusion based on our personal greed or fear, it is instructive to consult past “U” shaped recessions and see how the stock market has behaved around those recessions. If we can explain those movements of the stock market logically, we will improve our chance in projecting the fate of this rally.

In article 12 8 recessions since 1955, excluding the current one, are listed. Among those 8 recessions only the 2000-2001 recession is qualified as a “U” shaped recession; all the other seven are “V” shaped recessions. To study the 2000-2001 recession, annualized real GDP in quarterly interval, from 1997 to 2004, are plotted as the blue horizontal bars in the second graph at the right hand side. Real GDP rose strongly until the second quarter of 2000, turned into a flat period of six quarters long, changed to a moderately rising pace until the early 2003, and then resumed its robust growth again. Since the economic havoc of 2000-2001 is substantially milder than the current downturn, a sharp drop as in the first graph is not expected, but the six quarter long stagnation period shown in the second graph is certainly a good candidate for a recession. To decide whether it is indeed a recession, the “year-to-year comparison of real GDP then two quarter push back” method of article 12 is employed, and the result is plotted as the red curve in the graph. A green horizontal line is drawn at the level of +1%. According to the criteria used in article 12, when the red curve dips below the green line, a recession is deemed to have started, and when the red curve moves above the green line decisively, the recession is considered to have ended. The second graph indicates that a recession had started in the final quarter of 2000, reached its nadir in the second quarter of 2001, and ended in the fourth quarter of 2001. Real GDP grew in a sluggish pace throughout 2002, and picked up its pace of growth only from early part of 2003. Indeed the 2000-2001 recession qualifies as a “U” shaped recession due to its sluggish recovery after the recession was deemed to have ended.

A close inspection of the blue real GDP curve in the second graph reveals three dips during the period of stagnation, but none of any two dips are consecutive. According to the popular definition of recessions that requires a recession to encompass two consecutive negative growth quarters of real GDP, the 2000-2001 economic havoc is not a recession at all since there were no two consecutive negative growth quarters during the period of concern. Thus a sharp contradiction between the popular definition and the definition used in article 12 arises. To resolve this discrepancy we need to compare economic pains inflicted by the 2000-2001 economic havoc with the seven earlier economic havocs since 1955 that are deemed to be recessions by the definition of article 12. Among those earlier seven economic havocs only six are deemed to be recessions by the popular definitions of recessions. The first economic pain indicator used here is the number of job losses, calculated from the monthly nonfarm payroll numbers. The results are expressed as % loss from the peak prior or during the economic havoc, and are presented in the middle column of the following table. The situation of the current ongoing recession is also listed for comparison, with the June, 2009 nonfarm payroll number used as the low point in the calculation. The second indicator is the % loss of DOW from its peak prior to the economic havoc, and the results are listed in the rightmost column of the table. For the current recession March 9, 2009 intraday low is used as the low point in the calculation.


From the above table we observe that as the job loss is concerned the severity of the 2000-2001 economic havoc is in the middle of the group, but as DOW % drop is concerned, it is only below the severe 1973-1975 recession. Thus the 2000-2001 economic havoc is indeed a legitimate recession. The 1969-1970 economic havoc encounters the same problem as the 2000-2001 one, and according to the popular definition it is not a recession due to the lack of two consecutive negative growth quarters. The job loss during the 1969-1970 economic havoc is comparable to the 1990-1991 recession, and is above the 1979-1980 recession. However, DOW % loss of the 1969-1970 economic havoc is much larger than other five economic havocs that the popular definition deems to be recessions. Article 12 properly classified the 1969-1970 economic havoc as a recession. Apparently the popular definition of recessions is far from adequate. That is the reason why we prefer the definition of recessions adopted in article 12 over the popular definition.

With the status of the 2000-2001 recession settled, we will start to investigate the behavior of DOW around this “U” shaped recession. Weekly values of DOW, from July of 2000 to June of 2003 are plotted as black vertical bars in the third graph at the right hand side. The top of a weekly bar is the highest intraday high achieved during the week, and the bottom of a weekly bar is the lowest intraday low registered during the week. A short black bar at the right of a weekly bar denotes the weekly closing price of DOW. The green curve in the graph is the monthly tally of initial claims of unemployment benefit, and the red curve is the monthly figure of real personal consumption expenditure plotted in log-10 scale. Both the green and the red curves synchronize with the real economy well, and is painting a picture of the recession and its slow recovery. Three blue down arrows indicate the time periods of the start, the nadir and the end of the 2000-2001 recession respectively.

The pink slanting line displays a remarkable feature of passing through four lowest points of valleys of the weekly DOW diagram, labeled as “A”, “B”, “D” and “E” respectively. The vertical bar with its bottom labeled as “C” represents DOW at the week of Sept. 17, 2009, that was, the week right after the 911 tragedy. The vertical bar to the left of “C” represents the week started at Sept. 10, 2009, but this bar contains only the trading data of Sept. 10 since stock trading was closed for the remainder of the week. The bar next to the Sept. 10 bar is the bar for the week started at Sept. 3, 2009. This Sept. 3 bar was showing the ominous sign of the coming collapse already. Even if there had not been the 911 tragedy, DOW would still have collapsed, but probably would not have penetrated the pink line. From the peak “H” to “C” DOW had lost 3500 points, then DOW started to rally back from “C” for nearly six months to arrive at point “P”, a 2700 point advance. How to explain this gyration of DOW? When a recession hits, the mood in general, of course, is dark. Stock market analysts will also be affected by the dark mood, and their estimate of company earnings will be pessimistic naturally. As the nadir of the recession is passed, as the case of “C”, company earnings will turn out not to be as bad as the estimates of analysts. Investors are encouraged by the “green shoots” and DOW will starts to rally. As the end of the recession draws nearer, more “green shoots” will appear, and the rally will turn into a powerful one, riding on the hope that a new bull market is born. Analysts will be engulfed by the rosy mood, too and their earnings estimate will turn to optimistic. That was the cause of the rally from “C” to “P”. However, if the recession is a “U” shaped one and the recovery is sluggish, the actual company earnings will disappoint the anticipations of investors and analysts eventually. The market will panic and turn toward south. Depending on the length of the sluggish recovery, the earlier low like “C” will be exceeded, and the powerful rally becomes just a powerful bear market rally. In case of the 2000-2001 recession the bear market bottom, that is, point “E” in the graph, was not reached until near the end of 2002, fully one year after the recession had ended near the end of 2001.

The current rally of DOW from its March 9 low is similar to the rally from point “C” to “P” in the case of the 2000-2001 recession. If the coming recovery is indeed sluggish as many economists including this website anticipate, this rally will become another powerful bear market rally. Even this recession ends sometime in the second half of 2009 or in the early days of 2010, the bottom of this bear market may not be formed until late 2010 to 2011.