Thursday, May 15, 2014

Up 75% In 31 Months Without A 10% Drawdown

According to Ned Davis Research:
SPX falls by 10% or more about once every 16 months. It has been 31 months since the last 10% pullback, in September 2011.
So is SPX due for a 10% pullback?

The dashed line in the chart below shows gains and losses in the SPX of greater that 10% since 1980. The current uninterrupted advance from September 2011 is highlighted in yellow to the far right.

There is a very wide spread around that 16 month average between 10% corrections. Most advances are much shorter than the current one.

But, there have been three big exceptions: advances starting in 1984, 1990 and 2003. Each is highlighted with a yellow or green box. These shaded boxes are equal to the current advance in both duration (time) and magnitude (gain).

The first conclusion is this: on its own, the 16 month duration isn't particularly actionable. Yes, this advance is exceptionally long but, even if you exclude the 1990s, there were two other advances that went 6 months (1984) to 30 months (2003) longer than the current one. In fact, there's been a long advance once every decade. We're in the fourth one right now and the other three all went much longer.

What is notable, however, is that the current rally isn't just exceptionally long, but exceptionally powerful. The next chart compares the magnitude of the gain during the first 31 months of the current rally to the others.

Although the rallies from 1990 and 2003 (green boxes) lasted much longer than the current advance, the rate of gain in both was much less. In fact, it took each more than 4 years to equal the gains the current rally has made in 31 months.

The second conclusion, therefore, is this: the returns over the next year are unlikely to be great. The current rally could go sideways until the end of 2015 and still be equal to the 1990 and 2003 rallies. The 1982 rally went higher for another 6 months but then lost 36% and was lower a year later.

Lastly, bear in mind the larger context of the current rally. The next chart adds larger blue boxes equal to the duration and magnitude of the bull market from the 2009 low.

At more than 5 years, the current bull market (defined as a gain uninterrupted by a drawdown of more than 20% on a closing basis) is both longer and more powerful (on an inflation-adjusted basis) than either the one from 1982-87 or 2002-07. It is, in fact, longer than every bull market in the past century except the ones ending in 1929 and 2000.

In other words, this exceptionally long advance without a 10% correction is occurring at the point where virtually every bull market has already ended.

The blue arrows in the chart mark the end of each economic recession since 1980. What is notable is that the current advance is like the one from 1984: both started about two years after the end of the recession. The 1984 advance marked the end of the 1982-87 bull market.

In contrast, the other two long advances (green boxes) started right when the recession ended. Neither marked the end of their respective bull market. This could be why those advances lasted longer.

What does all of this imply for the market going forward?
  1. There are three possible scenarios: 
    1. More likely than not, the market will have its 10% correction in the months ahead. That's typical of the May-October period and it's typical of the mid-term election cycle.
    2. If the market doesn't correct by 10%, it will probably 'mark time' by chopping sideways for far longer than anyone right now imagines. There will be drawdowns close to 10% along the way that will wear out most investors.
    3. If there is a strong advance from here, it would resemble the summer of 1987 that resulted in a much larger correction and the end of that bull market.
  2. Regardless, equity returns on a net basis are likely to be well below expectations over the next year. This view is well outside of the consensus. 
The best scenario is a correction over the next several months. This would washout excesses in sentiment and valuation and set up a rally into year end. The returns on a net basis from here may not be great, but they would be from 10% lower.