S&P BSE SENSEX Valuation: All is Fair in Love and Markets

The SENSEX has crossed the 35000 mark and investors are getting a little uneasy. They have voices inside their heads asking "Is the SENSEX running ahead of itself? Should I cash out or should I stay in?" While those are difficult questions to give a definitive answer to, I thought I'd try to come up with a ballpark answer based on market metrics.

I am always amazed at how Prof. Aswath Damodaran could come up with a way to Value all living things (Kidding, of course). A few months back, I was going through some old blog posts of his and happened up this peculiar one: A Sweep Spot for US Equities: Opportunities and Dangers. The post was about Valuing the S&P 500 Index (US Equities) using well-known Valuation tools such as Cash Flows, Growth Rates, Risk-free Rates and Market Risk Premiums. I realized how bloody brilliant it was think about a broad market index that way.

After months of idle admiration, I decided to build a similar model and test in out on our very own broad market index: the S&P BSE SENSEX.

---

As with the original post, I will conduct this Valuation exercise based on 4 market metrics:

The Market Metrics

1. Expected SENSEX Dividends and Buybacks

If you have followed my earlier Valuations, you'd know that the most logical way to find out the Value of a stock is to discount its expected future Free Cash Flows at an appropriate Discounting Rate. However, it's very time-consuming and unnecessary to estimate the FCF of every firm in the SENSEX in order to arrive at its Value. So instead, I choose an easy way out: estimate the expected Dividends and Buybacks based on historical records, which is more or less synonymous with Free Cash Flows. Here is how it looks from 1997-2018:

(Source: BSE Website and DamodaranOnline)

While getting the Dividend Yield data from BSE's website was a piece of cake, the data for Buybacks was almost non-existent. I had to dig up 2018's Buyback data from Prof. Aswath Damodaran's website and then assume Buyback Yield as a percentage of Dividend Yield. In any case, we know that Buybacks by the SENSEX companies have been quite rare and this harmless assumption shouldn't affect the final Value by much.

2. Expected Growth in SENSEX EPS

SENSEX's EPS can be found by dividing its P/E by its absolute index level. The index level and index P/E levels are also readily available on BSE's website, making my job easier.

(Source: BSE Website)

While taking the average historical growth rate might seem trivial, they range from 10-13%, which we know to be consistent with the long-term CAGR of the index itself. The expectation here is that, of course, as Earnings grow, Dividend and Buybacks will too.

3. Risk-free Rate (10-year Indian Government Bond Yield)

For all Valuation-related exercises, we require a base rate to function as our Opportunity Cost. In the absence of any other logical metric, we turn to the long-term Government Bond Yield in India.

(Source: Investing)

It's crazy to think that 20 years back, the Risk-free Rate in India was as high as 12%, meaning Bank FD rate were higher than that. You could have realistically beaten the market if you'd made a 20-year FD in 1997. Imagine that! But with India's recent changes in RBI policy towards inflation-targeting, we see the Yields flattening towards 'a little above' the Inflation Rate.

4. Indian Equity Market Risk Premium

You can think of Equity Market Risk Premiums as the market demanding incentives to invest in stocks, which are riskier assets compared to Government Bonds.

(Source: MarketPremia)

A higher ERP indicates a defensive market where in investors are very cautious about investing and would like to pay a lesser amount for the same set of cash flows. The overall drop in the Indian ERP indicates that there an increased number of investors participating the markets, bringing it closer to pricing efficiency.

On an off-note, can you guess why India's ERP sky-rocketed during 2000? It's because of the fall-out of the Dot Com Bubble. You will also notice a little bump for the 2008 crisis.

---

Rounding up the Assumptions

With the data in hand, we can drill down and end up with what we require for the Valuation:

But now comes a predicament: which of these numbers are to be used. If anything, I would say it should be based on real-world logic:

1. The easiest one to figure is that the Risk-free Rate and Market Risk Premiums make sense only when you look at the recent figures. It doesn't make sense to consider the history here, for example, the very high 10-12% Risk-free Rate we had 15-20 years back. It does not represent the current economic situation. The same goes for Market Risk Premiums (Since ERP is just the spread between Risk-free Rates and Market Returns).

2. For the other two inputs, Total Yield and EPS Growth, there could be a thousand debates about what could be the correct number. I would rather play it safe and consider the longest term average (20-year for Total Yield and 10-year for Expected EPS Growth).

---

Valuing the S&P BSE SENSEX

When all that is said and done, this is the final result:

The Assumptions

This is how our assumptions look like (Discussed above):

The Cash Flows

This is how the Expected Dividends and Buybacks look like, discounted at the Expected Market Return (Risk-free Rate + Equity Risk Premium):

The Value

All these inputs throw the final Value of the S&P BSE SENSEX:

According to my estimates, SENSEX should be trading at 32562 levels. It is trading at 35690 levels, which is indicative of a 8.76% overvaluation.

---

Footnote: I Could be Wrong and Why

Valuing a single company is one thing. Valuing an index consisting of 30 different companies is entirely something else. I will not pretend that I performed an accurate Valuation, primarily for 2 reasons:

1. There are too many moving parts. Nobody knows what would happen to a single company in the future, let alone 30 of them. I wouldn't even refer to my Valuation as an estimate, but an informed guesstimate.

2. The Equity Risk Premium plays a crucial role in this Valuation. In an under-developed market like India, the ERP figure is not trustworthy and keeps changing all the time. In stark comparison, the ERPs of the USA and the European Region are fairly stable (Ignoring the swings during a market crisis). So, the accurate usage of this figure can make or break the final Value. I provided the logic for using the near-term ERP, but if you go ahead and use the 20-year ERP, the model will show an almost 50% overvaluation, which sounds ridiculous (Because it is - you should not consider historical ERP, as it makes no sense in the current economy).

I can safely say that the model is useful to watch out for excess overvaluations or undervaluations. But if you'd like to, say, trade the SENSEX using this model, I would seriously advice against it. Once again I reinforce the fact that the model is only a guesstimate.

---

The Model

Here is a link to download the model and check out the calculations for yourself:

(Download Valuing the S&P BSE SENSEX)

Got feedback or constructive criticism? Feel free to comment down below or drop a mail to dineshssairam@gmail.com. I hope this post sparks some interesting conversations about the current SENSEX levels.