Due to positive reader response via email, I will publish a brief analysis of the “big money” every week on sunday using the COT Report and money flow data. For this week we saw institutional money moving out of equities in a significant way: from the COT Report using percentiles–entering the beginning of the week, large institutions were long in the 66th percentile relative to historical norms for this time of year. As of Friday their position was in the 40th percentile, indicating a move of 26 points which is significant. This was also verified by money flows on WSJ Online on Friday:
Selling on Strength(Roll over for charts and headlines)
Typically when the SPY is in the number 1 spot, it is bearish for the stock market over the next few days. Institutional selling in general is bearish for the market unless it is being offset by a divergence in the form of heavy commercial (central banks, goldman etc) buying which is done on inside information of future announcements. This week the commercials (the central banks, JP Morgan and Goldman Sachs) were adding to positions slightly , although it was not a significant divergence like the last dip in equity markets that began in June. During that period, prior to the correction, commericals were actually selling substantially, and after a week or so into the the dip, they accumulated the highest position (100th percentile) held in over a year.In that case there was massive commericial buying that more than offset the selling by institutions indicating a vote of faith in the direction of the market. It seems more likely that this is an attempt to support the markets rather than an aggressive position bought based on insider central bank information. Looking at the sector level, financials were the most heavily sold, losing 3.38% on the week and a net outflow of over $350 million. Given that financials have been arguable the leaders in this major rally, this is not a good sign that they are now rolling over.
Looking at gold, we see that both commercials (central banks and miners) sold slightly this week, but are in the 3rd percentile relative to historical norms. This is atypical going into September which is usually a very strong month for gold from a historical performance as well as from a fundamental perspective in terms of demand. Institutions in contrast are in the 95th percentile, having sold slightly during the week. Essentially the institutions are close to all in. The small trader was adding heavily this week and is in the 81st percentile. As I stated before, this kind of behavior is not typical of what launches a successful breakout. When you add that to the fact that the US dollar lost only .2% for the week while gold rose 4% and commodities lost -4%, you have a strange divergence. Drilling down into the COT report we see that there was major movement on the part of institutions into the US dollar index. The only explanation for this is an unwinding of a carry trade in dollars. When you factor in that bonds have been rising as well, you get the general sense that institutions have been allocating there money more defensively going into the end of the quarter to preserve performance. Indeed the COT report confirms that institutions are at the highest level in 2-year Treasuries that they have been for a long time (100th percentile). The big money is selling the winners (financials), buying bonds and already in gold in case of another meltdown.
Analysis and Recommendations
Given that a lot of risk was taken on via carry trades in US dollars, a substantial amount of possible risk exists in holding gold positions at this time—a deleveraging will eventually cause the dollar to rise from its lows and hurt gold where institutions have heavy positions. This is especially true if the market starts to sell off, and money managers start to scramble to protect their performance. Bonds are also vulnerable, and the threat of surprise hikes remain an increasing probability—however small— which would crush prices- especially preferred shares and corporates which have risen anticipating the end of the recession. Make no mistake, I think the long-term direction of gold is likely higher, however Ithink it is a crowded trade at this time. Natural gas in contrast is the most underloved asset in the world with commercials holding a position in the 94th percentile (very bullish) while institutions are in the 5th percentile. My analysis would suggest buying gas following a 30-50 day high using Donchian Channels which have worked exceptionally well for gas. The strategy is to buy into strength instead of the extreme weakness we are seeing right now as a result of the accumlation of inventories. The bottom is probably close but impossible to forecast, however the upside is probably 100% or more. Any new major high will likely be significant and an indication that the worst is over. Ideally this would occur on negative sentiment and poor fundamentals as was the case with the recent huge equity rally.