The CNN Greed and Fear Index is an indicator compiled by CNN whose objective is to measure investors' appetite for risk. The index can oscillate between 0 and 100. It rarely reaches extreme levels over 80 or above 20. Recently it has traded over 80, even hitting 90 Wednesday. Does this mean it's time to sell?
I tried to find an answer by looking at GFI data back to mid-2010, which was the furthest back I could find readily accessible information. I created a couple of buy-sell criteria to assess how useful the GFI is. The result was unequivocal: when greed is high, the only thing to sell is the GFI. Hold on to your stocks!
What is the Greed and Fear Index?
The GFI is a composite of seven different indicators. Five of the indicators are a snapshot of stock market internals. They include:
- The CBOE VIX level. The VIX reflects investor expectation for market volatility over the next 30-day period. If investors expect high volatility, the VIX rises; this indicates fear. VIX is currently neutral, just under 13.
- The CBOE put-call ratio. This is a ratio between trading volume in puts versus calls. A low ratio (more call volume than puts) represents investor bullishness. Currently call action is more than 40% above put action.
- The relationship between the current price for the S&P 500 and its 125-day moving average. If current prices are above the 125-day average as they are now (by close to 7%), it means that investors are "being greedy".
- The number of stocks hitting 52 week highs versus 52 week lows. If the ratio is greater than one, investors are, again, bullish. Currently new highs are about 10% above new lows.
- The McClellan volume summation index measures advancing versus declining volume on the NYSE. This one's showing greed. In the last month about 24% more of each trading day's volume has been in advancing issues than declining issues.
There are two bond related measures in the GFI, as well:
- "Safe haven demand" measures performance of stocks versus bonds for the last 20 trading days. For the latest period, stocks have outperformed bonds by over 4%, an unusually large number.
- Junk bond demand is the last GFI component. Greedy investors are willing to accept a lower premium for low-grade bonds over high-quality bonds when they are feeling optimistic. Right now, the yield difference between junk and investment grade debt is 2.3%- a low number.
It's worth noting that none of the indicators in the GFI is forward-looking as regards the market. A contrasting example is the Index of Leading Economic Indicators (the LEI), which attempts to forecast the economy, and hence may have value predicting future market performance, by looking at things like housing permits, new orders and unfilled durable goods. The GFI is simply a snapshot of how investors feel at a given moment.
Does the GFI predict future market performance?
Since the GFI rarely treads above 80, much less 90, one might reasonably conclude that it's a good time to sell when it hits such a high level. I tried to assess how a "sell on high greed" approach might work.
The problem is that it might be easy to know when to sell, but it's a lot less clear to know when to buy back. So I tried a couple of different approaches.
First, I looked at data going back to late June, 2010. This was the period during which the GFI hit an extreme low, below 20%. The data was difficult to find. I ended up using a couple of different sources. One was CNN itself, which provides three years worth of chart data on its website. Here's a screenshot of the GFI for the last three years. Source is here.
I was able to go back three years further by scouring the web and finding a couple of sites whose GFI graphs matched. I ended up taking volume from here. Below is a chart of the GFI from 2010 to 2013, taken from Kirk Lindstrom's blog, here, and in turn from CNN's website.
Because I didn't have precise data but instead could only use loosely annotated graphs, my buy-sell dates weren't as good as I would have liked. Still, you would be surprised at the accuracy you can get from screen shots combined with Photoshop ruler guides! In the end, I estimate that my buy-sell dates were within +- 5 days of the exact GFI index dates I should have used. Normally this would be cause for concern, but you'll see from the results that a high level of granularity wouldn't have made a big difference, because the difference in investment returns was so big.
I created a couple of models using the SPDR S&P 500 Trust ETF (NYSEARCA:SPY) as my investment vehicle. Both models had in common that you sell SPY when the GFI hits what CNN calls an "extreme" level- over 80. This has happened about 14 times since mid-2010.
The trick was when to buy back.
In the first case, I bought back when the GFI fell back below 60. Note that sometimes the GFI went back over 80 before it hit 60. For instance in December 2011, the GFI hit 80 and then fell, but rose back above 80 again in March, without ticking my buy-in level.
The result from this first model- sell at 80, buy back at 60- was that I would have been in cash for close to 270 days out of the entire 2,170 day period. That's 12% of the time. But my returns would have been just over 50% cumulatively, versus 96% for the S&P.
To get a really accurate number, one would need to do two more things. First, there should be a dividend adjustment, since the buy-and-hold investor would capture about 12% more dividends. Second, and much more important, there's a major tax issue, because a buy-and-hold investor would not yet have paid taxes, and, depending on how the investment was made, could pay a much lower long-term capital gains rate.
As another test, I tried a different model. This one was to sell at 80, and buy back at 50. Results were worse. In this case, total return was 44% versus 96%. The investor would have stayed out for about 440 days out of the total of 2,170- 20% of the time. Another way to put it is that nearly half of the total return of the S&P 500 came during that 20% of the time when market sentiment was strongest.
Does All This Make Sense?
I have to admit that the extreme results of this little study were surprising. But if you look at the data, you can see that a lot of the outperformance came as a result of the market working off its extreme greed level without correcting all the way back to neutral. To be more specific, look at the period from late December 2011 to mid-March 2012. The market hit an 80% extreme in December, but the GFI fell very slowly, so that an investor would have remained uninvested for about three months, during which time the S&P rose from about 1215 up to 1350. Similarly, from December 2012 through most of February 2013, one would have been out of the market, while the S&P rose from 1430 to 1520.
There were even some nice rises during a short period after the market hit extremes. One would have been out of the market from the start of June 2014 to early July, for instance, but during that period the S&P rose from 1920 to 1980.
So the numbers work, and they make sense, too. This is what happens in strong bull markets. Corrections are small and sharp. Markets are often overbought.
This, by the way, is exactly what is also happening now in precious metals, specifically for the SPDR Gold Trust ETF (NYSEARCA:GLD) and the iShares Silver Trust ETF (NYSEARCA:SLV). Too much attention is paid nowadays to COT data, which indicates speculative sentiment in the market. COT open interests have been at historical highs. But this means nothing. All one needs to do is look at the similarly high open interests during the big gold bull run from 2009 to 2011. Specifically, open interest hit new highs week after week during 2010- along with the price of gold. With the S&P and GLD, we are dealing with exactly the same sentiment profiles.
The main question to ask oneself is whether you believe we are fundamentally in a bullish or bearish environment. If you feel bullish, then my advice is to feel greedy, as well.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.