I have written a few articles on risk at this point. The first was on the difference between the Japanese yen and VIX as measures of risk (The Yen Vs. The VIX: 2 Aspects Of Attitude Towards Risk), with the former security more closely related to risk appetite, and the latter more closely related with expectation of risk. I then expanded on this idea by simply taking the change in USD/JPY and dividing it by the change in DXY, thus taking into account changes in the overall exchange rate of the dollar, as a possible reason for fluctuations in the yen (A Simple Risk Appetite Index).
However, this index is, as the name suggests, a simple one. It is partially impacted by the Japanese economy itself. There is, however, a way to limit how much that impact will have on the index, and that is to average it with another measure of risk appetite. For that, we can use gold. Or, if the data for gold is not immediately available, use GLD or gold futures, which closely mirror the price movement of gold.
I am using /GC in place of gold prices, for this second version of the risk appetite index. This index, unlike the simple index, is cumulative over a given time period, but is still a relative index. The formula is as follows:
In this formula, y subscript i is the ith percent change in USD/JPY divided by the ith percent change in $DXY, while g subscript i is the ith percent change in $DXY divided by the ith percent change in gold.
The reason for using the square root of the squares was in order to take into account both the yen and gold as indicators. The square root of the sum of the squares is essentially the norm. The multiplication by 1/2 ensures that if both indicators are less than 1, the resulting norm will be as well, and if both indicators are greater than 1, the norm will be greater than 1 as well.
This norm does have its limitations and it will not be the final version of the indicator. The biggest issue is that the norm assumes that the risk appetite, as indicated by price shifts in the yen, is independent of the risk appetite as indicated by the price shifts in gold, or in other words, that there is no inherent correlation between the yen and gold, resulting from changes in risk appetite.
Below is the 20-year weekly price history of the S&P 500 along with the revised appetite indicator, created using ThinkorSwim.
The first part of the indicator chart is flat because /GC price history does not go back far enough. However, it is clear that as we got closer to 2008 - 2009, the indicator dropped sharply. It did not start to recover until October of 2012. Furthermore, it never recovered to its pre-2008 level, and has started to decline again.
Just as with the prior indicator, a decline in risk appetite does not mean that a drop in stock prices is right around the corner. It all depends on how much risk people expect there to be, and how risky the market actually becomes. Risk appetite can decrease, but if investors believe that there is very little risk ahead, as has been the case for the last few years, money can still flow into the markets. But the more risk appetite declines, the less likely it is that we will see significant gains, especially if perceived risk increases.
Disclosure: I am/we are long SH,VIXY,RWM.
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.
Additional disclosure: I hold a generally contrarian portfolio.