Gold to CPI ratio
This is a continuation in a series on the value of gold. Feel free to check some of my other posts including the subsequent comments:
The Shadow Price of Gold
Gold as an Inflation Hedge
In The Golden Dilemma, the authors look at the Gold to CPI ratio. Of particular interest is Exhibit 5 which relates the Gold/CPI to 10 year real returns of the price of gold. The exhibit shows a clear downward trend. As the Gold/CPI ratio gets high, real returns for gold decline.
Currently the ratio is about 7. This would seem to indicate that returns may be lousy for gold over the next decade. In at least one sense, this makes sense. After all, the TIPS yield curve does not offer high returns. (See here). With the exception of 30Y, all the yields are in negative territory.
I decided to invert the ratio to look at CPI to Gold. Here's a look at the correlation between CPI/Gold versus 10Y returns:
There's a good correlation there. Looking at the model versus actual gold returns:
The model does fairly well until the mid-1980's. I was able to improve the model by factoring in a suggestion by fool ryanalexanderson. I included subsequent 10Y real GDP growth.
This model fits the data much better. The big question is whether or not it will do so in the future. It is much easier to find models that fit to past data. Finding models that fit to future data is not as simple.
One downside to this model is that it requires you to estimate 10 years of real GDP growth. I assumed real growth of 1.67% for the remainder of the model's prediction which is about where my actual data ended.
In any event, looking at the Gold to CPI ratio (or its inverse) seems to fit well with subsequent gold price returns. Right now there's indication that future real returns for gold will be negative. That, of course, should be coupled with the fact that TIPS yields are currently in negative territory.