May 18, 2011 New Update
Here is the latest look at the “Sweet Sixteen” Dow recoveries adjusted for inflation/deflation I’ve been illustrating over the past year. The charts below compare the current Dow recovery since the March 2009 low with fifteen other major recoveries dating from the origin of this legendary index in 1896. (See the footnote for my selection criteria.)
At this point the Dow is 554 market days beyond the 2009 low. The index has moved from 4th to 3rd place in our Sweet Sixteen competition since our last check, one month ago today.
Why is inflation adjustment useful for this overlay? Throughout history the cost of living has undergone some dramatic changes, as this chart illustrates. High inflation, such as during the 1974 recovery, gives an exaggerated sense of price growth. Deflation, which accompanied several of the earlier market cycles, makes recoveries appear weaker. By adjusting for the inflationary/deflationary cycles, we get a clearer sense of the real value of the index price across time.
Now let’s extend the time frame. Here is a set of charts with increasing numbers of market days: 500, 1000, 2000, 3000, 4000, and 5000. Depending on the historical period, the number of market days in a year varies slightly. But it rounds out to about 250 market days per year. So the time frames in this series are approximately 2, 4, 8, 12, 16, and 20 years. The series features the 500-day chart with and without the 1932 recovery, which was a quite an outlier. At 1000 market days, the 1932 recovery continues to lead the pack. But at 2000 days (about eight years), the recovery after the 1921 low has risen dramatically. Of course, with the benefit of hindsight, we know that this remarkable advance was the last stage of the Roaring Twenties stock bubble, as the 3000-day (12-year) overlay makes clear. At 4000 days (about 16 years), the recovery from the low in 1982 is approaching the final surge of the Tech Bubble. The 5000-day chart shows how the Tech Bubble played out for the Dow, topping out in January 2000 after a brief scare in 1998 triggered by the Long-Term Capital Management Crisis (that dip after the 4000-day mark). The chart below shows the 5000-day (approximately 20-year) overlay:
The overlay charts give visual evidence of the wide range of recovery patterns. The table helps quantify the magnitude of the difference. Two of the earlier recoveries, 1903 and 1914, and two of the later recoveries, 1962 and 1970, subsequently failed. Likewise the 1938 and 1974 rallies failed before being rescued by later recoveries. This last observation touches on an important aspect of the overlay charts. As the timeframe increases, the same recovery may appear in multiple data series.
Cyclical and Secular Markets
How will our current recovery fare during the coming months and years? History shows us that some recoveries are the beginnings of secular bull markets. Others turn out to be cyclical bear market rallies. If we look at the broader S&P 500, we see that the current market is overvalued by the indicators I routinely follow (here and here). The Dow recovery indeed has more in common with recoveries in the earlier decades of its existence. Of the first nine rallies in the table above, only one other (June 1962) occurred after the middle of the last century.
The recovery since March 2009 is the second in the first decade of the 21st century, and it started from a lower low. As we can see in the inflation-adjusted chart below, history has witnessed several other examples of multiple recoveries in relatively close succession with lower starting points. Will the current recovery be another such example? Only time will tell.
Footnote on Selecting the Sixteen: My initial plan was to overlay all the Dow rallies following a 30% or greater decline. Using the traditional 20% decline associated with bear markets would have made the chart too busy, and it occasionally runs against conventional wisdom. For example, the Tech Crash in the Dow consisted of 3 baby bears (if you round up the 19.91% decline in January-March 2000) separated by two rallies over 20%. I consider it a single bear market with a decline of 37.85% and thus included the rally that began in 2002. I also treated the Crash of 1929 as a single bear decline, even though the 20% rule would have divided it into six bear markets with five intervening rallies. Likewise, and more to the point for the overlay, I treated the rally after the 1932 low as a single rally, even though the 20% rule would see it as an oscillation between three bull and bear markets.
Another liberty I took in selecting recoveries for the overlay was to include two rallies after declines of less than 30%. In both cases, they marked the beginning of a new economic era. One is the recovery that began in 1949 after the 23.95% post-war decline. The 12-year, 355% advance that followed warranted inclusion. Likewise I added in the first 500 days of the 250% rally that started in 1982 after a 24.13% decline. The 1982 recovery brought an end to the decade of stagflation and launched the great Boomer bull market.