Risk Indices Don't Consistently Lead, or Lag, Before Corrections

The conventional wisdom is that "risk indices" lead. As long as RUT and NDX are making new intermediate-term highs, the market is considered healthy.

We should, therefore, expect them to peak before the large cap indices, SPX and DJIA. If large caps decline but RUT and NDX continue to climb, "risk is on" and the divergence is expected to resolve higher.

Is the conventional wisdom correct? Do RUT and NDX tend to peak before SPX and DJIA?

The short answer is no. The slightly longer answer is that there is absolutely no correlation between market peaks and index leadership: sometimes large caps peak first, sometimes they peak last. Even more telling, sometimes SPX and RUT will peak together and DJIA and NDX follow.

The first chart looks at greater than 20% declines over the past 25 years. The numbers indicate which index peaked first, second, third or fourth.

DJIA peaked first in 2000, SPX peaked last and the risk indices were in-between.

In 2007, RUT peaked first, NDX was last and the large cap indices were in-between.

Small Caps Close Below Its 200-dma

On Tuesday, RUT closed below its 200-dma for the first time in more than 360 trading days, one of the longest such streaks ever.

On Wednesday, RUT fell further, to a strong level of support extending back seven months, and then reversed.

An Update on May to October Seasonality

In a sign of how short investors' memories are, one of the newest memes is that equities are unlikely to be weak this summer because they already corrected during the first four months of the year.

Anyone remember 2010 or 2011? Both started weak and got weaker in the middle of the year.

How about 2000, 2001, 2002, 2004, 2005, 2007 or 2008? Weakness early in each of these years did not preclude a second period of weakness in summer.

Weekly Market Summary

The biggest catalyst coming into this week was seasonality. Normally, the end of April and beginning of May is strong. There were few other extremes, except the equity put/call ratio which had spiked enough on Friday to precipitate a bounce (here).

In the event, SPX and DJIA moved up about 1%, but in most ways the bounce was weak. 80% of the gain in SPY this week came from overnight gaps higher on Monday and Tuesday. Trading during cash hours added little, indicating poor investor conviction.

More to importantly, breadth was terrible. DJIA made a new all-time closing high but here's how its 30 constituents performed: one made a new all-time high, two made a new 1-year high, only four made a new 2-month high and just 8 were at even a new one-week high.

The other indices are similar. 75% of the Nasdaq is below its 50-dma. Only 42% of the SPX is above its 50-dma even though the index is less than 1% from its all time highs. This breadth divergence is similar to July 2011 and April 2012.

The 1996 Analogy

RUT has now closed above its 200-dma 361 days in a row. The all-time record is 364 days from 1996.

SPX has also closed above its 200-dma for longer than any time in the past 30 years, except 1996.

SPX has not touched its lower weekly Bollinger in 17 months, longer than any time, except 1996.

And SPX is up 5 quarters in a row. In the five previous times this has occurred in the past 40 years, it has only closed up for a 6th quarter once. The year: 1996.

Apparently the current period resembles 1996.

The current rally started in October 2012 and ran up some 40% to January of this year. It swooned for two weeks, then rallied 8% in February. Since then it has chopped in a range for 9 weeks and counting.