Saturday, April 6, 2013

Weekly Market Summary

A month ago, SPX entered into the area of strong resistance from its 2000 and 2007 peaks (chart). Since then, SPX has had a weekly open/close range of less than 20 points, or about 1%. In the last 12 days, it has alternated direction (up/down) every day. The character of this market was encapsulated on Tuesday; SPX made a 5 year high yet the McClellan oscillator (a measure of breadth) closed well below zero (-23).

Pure trend followers have been right to say that the trend is up for the US indices. Trend is the single most important factor we follow and we have agreed with them and kept our trend assessment green throughout 2013.

This week, however, the US trend has started to weaken. Both the $RUT and $COMPQ broke their November uptrends and closed at 4-5 week lows, essentially reversing all the gains in March. Small caps had been outperforming SPX in 2013 until this week; now it's a net laggard. SPX has not broken its channel, but it normally follows.

The Euro 350, All World Ex-US, $DAX and 4 of the 6 cyclical sectors of the SPX also broke their uptrend this week, as have DJT and $SOX (click the links to see the charts). These sectors and indices should be leading, and they are instead falling (chart). Hong Kong and EEM are at December's levels. When the rest of the world's indices are near or below the open for the year, it is noteworthy.

The rest of the world's indices are following macro developments. The Eurozone as well as the G10 Economic Surprise Indices (CESI) both fell below zero this week. Before the poor NFP report on Friday, the US CESI was headed back towards zero as well. After a tide of very good data, the economic trend in the past several weeks has been weak. Worryingly, this is a seasonal trend that has developed every year since 2010.

It is notable that the bond market was way ahead of equities in expecting a weakening of economic data. 10- and 30-year treasuries exploded higher this week; we have noted before the importance of these divergences (post and current chart).

The relative low risk of treasuries have now outperformed SPX and $RUT for 4 weeks and have equal performance since late January; treasuries have beaten $COMPQ since early January.  That is an exceptionally poor risk-adjusted return, something pure trend following has missed (and, in our mind, a key short coming of that approach).

Breadth remains poor.

There is not much value in following micro changes in sentiment; investors have been very bullish since February and, to take one example, Investors Intelligence recorded the widest spread between bulls (many) and bears (very few) of 2013 this week. Moreover, Wall Street has upped its targets after its initial estimates were surpassed. But the real tell on sentiment was the mainstream media attention given to the 'Great Rotation' out of bonds, a trojan horse we warned against (post).

Seasonality remains strong in April but we are close to the weakest 6 months of the year. Over the past 50 years, the Dow has risen an average of 7.5% during November-April and fallen 0.1% during May-October. Seasonality becomes a headwind in the next few weeks.

To be completely clear, all of this is not to say that US indices are on the verge of a major plunge. The overall economic trend is improving, albeit slowly. The Fed is accommodative and, despite all the criticism, it's policies have been successful. Our expectation has been for a typical correction followed by further upside. Since 1980, the average annual intra-year correction has been about 15% (median 11%). Read further here. Volatility is in a low period that is reminiscent of the mid-1990s and mid-2000s.  Our stalking horse continues to be the 2011 market. Wait for it, a fat pitch on the long side is coming.

Short term, the US indices appear to be in a sideways pattern that has occurred also during the past few years (current chart). With this as a model, a second test of the 1575-90 resistance area would be normal, as would a pierce of the 50-d. Friday was a strong rebound and short term divergences are positive.