Investor Sentiment: Some Context

In my most recent post on investor sentiment, I suggested selling into strength and I suggested that the optimal time to do this was during this week. An astute reader asked the question: what do I mean by strength?

This is a great question because all too often market commentators make a "call" without the context of time (i.e., when things you predict might occur) or without the context of draw down (i.e., how much pain in the form of losses you have to suffer while being wrong before you get it right).

So let's define the question. I am basing my analysis on the following two observations: 1) investor sentiment, as defined by the investor sentiment composite indicator or "dumb money", is neutral for 5 consecutive weeks; and 2) price is below the 40 week moving on the S&P500 for 5 consecutive weeks. So the question becomes: when these two conditions are met, how does the S&P500 perform over the next 13 week period?

So we "buy" the S&P500 when both the "dumb money" is neutral and the S&P500 is below its 40 week moving average for 5 consecutive weeks. Positions are "sold" after 13 weeks or when the "dumb money" becomes bearish (i.e., bull signal). Commissions and slippage are not considered in the analysis. This is analysis is based upon 18 years of data.

If you only traded this strategy, you would have generated 11 trades. Only 2 of these trades would have been winners, and you would have lost 550 S&P500 points with only about 10% market exposure. This is quite significant. The majority of trades occurred in the 2001 - 2002 bear market or the current bear market.

But to give a little more context to the analysis and to try and answer the question of "what do I mean by strength", let's look at the maximum favorable excursion (MFE) graph for this strategy. See figure 1.

Figure 1. MFE

What is MFE? MFE measures how much a trade runs up before it is closed out for a loss or a win. Look at the caret in figure 1 with the blue box around it. This caret represents one trade. This trade ran up 2.5% (x-axis) before being closed out for a loss of 9% (y-axis). We know the trade is a loser because the caret is red.

So what can we tell from the MFE graph if we bought the S&P500 when both the "dumb money" is neutral and price is under the 40 week moving average? In 8 out of 11 instances, prices ran up less 2.5%. These are the carets in figure 1 to the left of the vertical blue line. Or to put it another way: the ultimate intermediate term top was identified within 2.5% of the initial signal in 8 out of 11 instances. In the other 3 instances, the S&P500 moved between 5% to 7% higher before heading lower.

So what do I mean by selling into strength? Based upon the scenario that I have defined, one can expect a top about two thirds of the time within 2.5% of the signal. We are now entering that window (i.e, the 5th week) where we would expect the market to begin to stall.

Investor Sentiment: Sell Into Strength (Still)

This is the fourth week in a row where the "dumb money" is neutral and the "smart money" is bearish, and this is not a scenario that is generally supportive of higher prices especially with prices on the S&P500 under their 40 week moving average. The ideal situation for higher equity prices would be for the "smart money" to be bullish and the "dumb money" bearish (i.e., bull signal).

The "smart money" (see figure 1) refers to those investors and traders who make their living in the markets. Supposedly they are in the know, and we should follow their every move. The "smart money" indicator is a composite of the following data: 1) public to specialist short ratio; 2) specialist short to total short ratio; 3) SP100 option traders.

Figure 1. "Smart Money"

The "dumb money" or investment sentiment composite indicator (see figure 2, a weekly chart of the S&P500) looks for extremes in the data from 4 different groups of investors who historically have been wrong on the market: 1) Investor Intelligence; 2) Market Vane; 3) American Association of Individual Investors; and 4) the put call ratio.

Figure 2. "Dumb Money"

When combining the "dumb money" indicator with other metrics such as poor market internals and price under the 200 day moving average, we find that rallies tend to fail after 4 weeks of the "dumb money" indicator being neutral. We are now in that window where we expect the rally to stall.

Market participants remain hopeful to start the new year. The major indices have been driven higher on lackluster volume in anticipation of better times to come. Those hoping for a bottom can point to the market's persistence despite the bad economic news. They often make statements like: "the market is discounting the bad news and looking ahead" or "the market always bottoms before a recession ends". My response to such axiomatic dogma is two fold: 1) if the market is "all seeing", then what was the market thinking (or discounting) at the October, 2007 highs?; 2) the market bottomed almost 12 months after the 2001 recession ended so not all market bottoms occur at the trough of the recession.

Admittedly, the most bullish thing about the price action has been the price action. All the secondary indicators such as sentiment, volume or market internals have not mattered. Prices on the major indices have moved higher and key price levels have been met. The Russell 2000 and NASDAQ Composite have broken through resistance levels (albeit on sub par volume), and the S&P500 and Dow Jones Industrials have held at support and moved higher. This is all good - not only are prices headed higher but we know where are our points of failure (i.e., breaks of support).

But let's remember that this is still a bear market, and the winning strategy remains to sell hope and buy fear. We hope that the lackluster volume doesn't matter. We hope that increasing bullishness does not slow down the ascent of stock prices. We hope that breakouts don't turn into fake outs (which is a hallmark of bear markets). From this vantage point, the sentiment data I am presenting suggests that rallies are better selling opportunities.

Reconsidering Gold

Back in August, 2008, I wrote an article about gold essentially stating that the biggest threat to price appreciation in gold was price appreciation in the US Dollar Index. At that time, that was a good "call" as the the US Dollar Index is up about 10% from the July 31, 2008 close and gold is down about 5% from that time after having been down as much as 25%. I based that assessment of gold and the US Dollar Index on my "next big thing" indicator which looks for secular trend changes. Based upon this metric, the US Dollar Index looked better than gold than, and it still does so today.

So fast forward to today and we see that gold has closed the month above its simple 10 month moving average and above a down sloping trend line formed by two prior pivot high points. See figure 1, a monthly chart of gold. Typically and on a purely price action basis, I would consider this bullish. However, for gold, these technical milestones have failed to produce any meaningful edge especially when we apply other filters of the price action (such as the "next big thing" indicator). In other words, I don't believe that this represents the beginnings of a new secular up trend in gold.

Figure 1. Gold/ monthly

The "next big thing" indicator is shown in the lower panel of figure 1, and essentially, it is not in position where we would expect a new secular up trend in gold to develop.

As far as the US Dollar Index, I believe there is still some upside left in the current price move. Figure 2 is a monthly chart with the "next big thing" indicator in the lower panel. The indicator remains in a position (like back in August, 2008) that suggests a new up trend should develop. Furthermore, moves in the US Dollar Index typically end with a some sort of negative divergence between price and an oscillator. In other words, you don't see price thrust off the bottom (like it did in August, 2008) and then just fall apart. It would not surprise me to see the US Dollar Index test and fail at the December, 2005 highs. This is at least several months away.

Figure 2. US Dollar Index/ monthly

The obvious is that this analysis goes against the consensus opinion. The consensus opinion is that the Federal Reserve is printing money to do whatever it takes to cure the American economy. This is inflationary. And obvious to all observers.

However, the following isn't so obvious: despite all the creation of money, money is not being lent as lenders are reluctant to lend or borrowers are unable to borrow or money is going to pay off debt. I am grateful to one of our readers who sent me a chart of the M3 Money Supply (estimated) from John Williams' Shadow Government Statistics. Growth of M3 Money Supply and gold price appreciation should be roughly correlated, but what we find is that M3 Money Supply is decelerating despite the Fed's Herculean efforts. Essentially, money is not finding its way into circulation.

You can find the chart of the estimated M3 Money Supply by following this link.

Short Term Over Bought

Figure 1 is a daily chart of the S&P Deposit Receipts (symbol: SPY), and the composite indicator in the bottom panel is an oscillator constructed from the following data: 1) $VIX; 2) Put Call ratio; 3) NYSE advance decline; 4) NYSE volume.

Figure 1. SPY/ short term oscillator

The indicator value is in over bought territory. The vertical purple bars highlight the previous 9 over bought readings going back to April, 2008. As you can see, 8 out of these 9 over bought readings led to either an intermediate term top or sideways movement in prices. The lone exception is the vertical bar with an oval on it; this occurred in July, 2008. Of interest, this is the one over bought reading associated with a significant up thrust in price - in other words, the 5 bar moving average of price actually closed outside the upper trading band (see red arrows). This kind of price action is noteworthy and exceptional. But this was the only such occurrence in the past 8 months.

In general, I am not a big fan of oscillators as they tend to fail in trending markets (i.e., over bought becomes more over bought). It is my belief that a significant upside move should be accompanied by a significant up thrust in price. This has yet to occur with the SPY on this current rally off the November lows.

Figure 2 is a daily chart of the Power Shares QQQQ Trust (symbol: QQQQ); the indicator in the bottom panel is constructed from the following data: 1) $VXN; 2) put call ratio; 3) NASDAQ advance decline; 4) NASDAQ volume. Currently the indicator is only heading towards over bought, but past over bought readings over the past 8 months generally have not led to higher prices. A recent price thrust is noted with the oval.

Figure 2. QQQQ/ short term oscillator

The bottom line: we have a short term over bought market within the context of a bear market. The market has lifted over the past 2 days on non-existent volume. The sentiment picture has been discussed. In the absence of overwhelming price action and breadth, my over all impression is that rallies are better selling opportunities.

Lastly, I will continue to watch the key price levels as the various ETF proxies seem to be trading around these areas of buying and selling producing choppy trading conditions. It would be nice to have a move above (or below) the key support (resistance) areas that is also a significant up (or down) thrust.