The correlation between the Nifty PE and long-term bond yield (10-year GOI bond) is discussed in this post. I had earlier considered the relevance of the Nifty PE for the long-term investor and misconceptions about the Nifty PE.
A few days ago, I had asked, what is a high index PE?. There were two interesting comments by Deep and K
This yield is indicative of prevailing interest rate scenario. RBI decides the overnight (policy) interest rate by considering the overall state of the economy. The short-term bond yields closely follow the policy rate. The long-term rates are governed by supply and demand but still depend on the policy rate.
In what follows, the Nifty PE at the time of investment is plotted against the return from Nifty after 1 year.
This is independent of bond yield. The green rectangle shows the returns for Nifty PE 20-24. The short dividing line inside the rectangle divides positive and negative returns. Notice that there are approximately equal no of points on either side of the divider.
This means, investments made at high PE regardless of bond yield would result in a +ve or -ve return after one year with approximately equal probability.
Now we consider what happens to this probability or the number of +ve returns and -ve returns inside that box when the bond yield is considered.
The same graph as above, but only investments made when the bond yield is greater than or equal to 7% are considered. Hard to spot changes between the two graphs, aside from some points in low-PE region (<16) are missing here.
At least as of now, I am inclined to call any bond yield up to 7% as a low yield.
When investment are made at yields >= 8%, there is a marked change. The number of +ve and -ve returns have decreased, but there are more -ve returns inside than rectangle than positive.
This means that if you invest when the yield is 8% or higher and with a PE >20, there is a reasonable chance that the return after one year would be negative. However, it still does not preclude the possibility of a very good return.
This is merely an observation made with limited data. Please do not treat this as investment advice. I certainly am not going to.
Now for the record, let me complete this study with higher yields.
Returns when the investment was made at yield at 9% or higher.
Returns when the investment was made at yield at 10% or higher.
Returns when the investment was made at yield at 11% or higher.
Not much to say anything here.
Please do not conclude from this that PE+yield based investing 'matters'. Sure it affects the probability of getting +ve returns over one-year. If I repeat this exercise over 10-years, there is nothing to shout about.
There is no significant change in PE 20-24 band aside from a handful of "low return" points which vanish from another branch. Since the use of a 10-year window significantly brings down the sample set, I do not wish to read too much into this.
We will have to constantly keep in mind that only 17.5 years of data is available to study. This limits the possibility of saying anything conclusive.
Finally, I would like to plot the PE and the product of PE and yield. Both are normalized with respect to their maximum value (since Jan. 1999).
Notice that the Nifty has fallen sharply at much lower PE levels (than 1, the max). And since 2008, it has not touched '1'.
When the PE x Yield product is considered, the recent highs are significantly lower. What we consider as 'high PE' and 'high yield' constantly changes with time. This makes it more difficult.