1. Advisorkhoj
  2. DSP BlackRock Mutual Fund
  3. articles
  4. Valuation Matters: How can you use valuations to make better investment decisions

Valuation Matters: How can you use valuations to make better investment decisions

Sep 22, 2016 by Dwaipayan Bose | 61 Downloaded
Picture courtesy - PIXABAY

Over the past month or two, we have received a number of queries asking whether it is wise to invest in equity mutual funds since “the market is near its peak”. The market has been at its peak many times in the past; peaks are meant to be conquered. Irrespective of the market level, equity, as an asset class, gives superior returns in the long term compared to all other asset classes. You will hear financial planners advising people to invest in equity mutual funds for their long term financial goals.

As an investor, however, you may want to understand, what “long term” means in fairly specific terms. Is it 3 years, 5 years, 10 years, 15 years or 20 years? For the benefit of our readers, I asked this question to a certified financial planner based in Kolkata. He told me that, as per him, long term is at least 10 years. In his or her lifetime, an investor usually has multiple financial goals; not all goals are 10 or 20 years away. You may have read or heard that, over shorter investment horizons, timing is important for lump sum equity investments. You may have also read or heard that, timing the market is very difficult for average retail investors. In this post, we will discuss if it is possible to time your lump sum investment and if yes, how.

What is timing and is it difficult?

Timing refers to when you make your investment and when you exit your investment. In stock markets, you will make money by following a very simple rule; buy low and sell high. However, it is easier said than done. In a bull market prices keep increasing, while in bear markets prices keep falling. By looking at prices, you will not be able to determine, what price is low and what price is high. For readers, who are interested in theoretical finance, the Random Walk Hypothesis supports the assertion that market timing is difficult, if not impossible. Noted American economist and Nobel laureate, Eugene Fama, in his 1965 article titled Random Walks in Stock Market Prices, theorized that, stock prices move completely at random and is, therefore, impossible to predict. If stock prices are impossible to predict, then as per the Random Walk Hypothesis, market timing is impossible.

While the Random Walk Hypothesis is one of the foundational theories of finance, another American Nobel laureate, Robert Shiller, challenged the efficacy of the famous hypothesis and showed that, there is a relationship between P/E ratios and future price levels of S&P 500 index. In other words, Shiller suggested that, one can make more profits by investing in the S&P 500 index based on the index P/E ratios. In this post, we will examine whether Shiller’s theory works in Indian stock market as well.

Intuitively Shiller’s theory makes a lot of sense because you can make more money if you are buying an asset at a cheap price, rather than at an expensive price. How do we know if we are buying an asset at a cheap price? The price you pay for an asset should be less than the future cash flows that you earn from the asset, discounted for the time value of money. When you are buying a share of a company, you are, essentially, buying the future earnings (EPS) of the company. The earnings can be distributed to you (the shareholder) as dividends or be re-invested in the business, to give higher future earnings which will result in higher future share price.

P/E ratio is a metric which tells you how much price you are paying for buying 1 of earning (EPS) in the company. If you are paying a lower price to buy the earnings of the company, you are buying it cheaper and if you are paying a higher price, you are buying it dearer (more expensive). Therefore, lower the P/E ratio of an asset (or stock) the higher is potential return on investment and vice versa. Remember as earnings increase over time, the stock price will also increase, but the stock price in itself will not tell you whether the stock is cheap or expensive. P/E ratio is a metric which tells you, whether a stock or index (like Sensex, Nifty etc) is cheap or expensive. We should now, therefore, tweak the rule of making money in stock markets; instead of “buy low and sell high”, our mantra should be “buy cheap and sell expensive”.

Is the rule buy cheap and sell expensive difficult to implement for retail investors? If you track P/E ratios and set P/E based investment rules, it is not difficult to implement. The question is at what P/E ratio should you invest in equities and what ratio, should you shift your asset allocation from equities? Hopefully, you will get some insight on this question in today’s post.

Sensex Price and P/E ratio

The chart below shows the monthly closing prices of the Sensex prices from 2001 to date. The chart shows that, the Sensex has risen from around 4,300 to nearly 29,000 in the last 15 years or so; CAGR of around 13%. If you simply went by the price levels of the Sensex to find investment opportunities, you would not have found many (except 2008, 2011 and early 2016, shown by red circles on the chart) and lost out on lots of money making opportunities.

The monthly closing prices of the Sensex prices from 2001 to date

Source: Bombay Stock Exchange

Let us now look at P/E of the Sensex from 2001 to date. The P/E ratios of the Sensex have fluctuated from 12 to 27. The maximum P/E ratio of the Sensex was around 27 in late 2007 / early 2008 and the minimum P/E ratio was around 12 in the depth of the financial crisis towards the end of 2008. With the benefit of hindsight, we can say that, the maximum P/E ratio was a result of irrational exuberance, while the minimum P/E ratio was a result of panic. You should also observe that, in the last 15 years, while the Sensex rose from 4,300 to nearly 30,000, the P/E ratio has remained mostly in the 15 to 25 band. The average P/E ratio of the Sensex in the last 15 years was 18.3.

PE of the Sensex from 2001 to date

Source: Bombay Stock Exchange

Let us now see if P/E ratios could have been used as reliable predictor of future gains. In other words, if we were buying the Sensex when P/E ratio was low, could we have got superior returns? In the chart below, we are plotting 5 year rolling returns of the Sensex since 2001 till date. Rolling returns are the annualized returns over the defined investment period (5 years in this case) at every point of time. We will compare the rolling returns of the Sensex with the P/E ratios at the beginning of each rolling return period. The vertical axis (Y – axis) on the left plots the Sensex 5 year rolling returns, while the secondary vertical axis on the right plots the Sensex P/E ratios.

Plotting 5 year rolling returns of the Sensex since 2001 till date

Source: Bombay Stock Exchange

From the chart it is visually apparent that, 5 year forward rolling returns of the Sensex have an inverse relationship with the P/E ratios of the Sensex. We will put our visual observation to a more rigorous statistical test later, but let us first try to get our hands around probabilistic dispersion of P/E ratios. In other words, we will see the probabilities of Sensex P/E ratios for being within some ranges. Most variables in nature follow normal distribution or what is popularly known as the bell curve. Normal distribution or bell curve implies that, the values are most likely to cluster around the average and the probability tapers off, as the values deviate further and further away from the average (the shape of a bell). As per normal distribution theory, 68% of the P/E ratios will be +/- one standard deviation from the average. 95% of the P/E ratios will be +/- two standard deviations from the average.

Dispersion of P/E ratios and boundaries

We had discussed earlier that, the average P/E ratio of the Sensex in the last 15 years was 18.3. The standard deviation of P/E ratios of the Sensex in the last 15 years was 2.8. Therefore, as per normal distribution, 68% of the times the P/E ratio will be 15.5 to 21. Look at the chart above. You will see that, you could have got very good 5 year returns, if you invested when the P/E ratio was around or less than 15. But investors should know that, 15.5 is the lower limit (at 68% confidence level) of the P/E ratio.

If you wait till the Sensex P/E ratio is 15.5, you may not get enough investment opportunities. In fact, for the last 6 years, the Sensex P/E ratio did not dip below 15. So at what Sensex level should one invest? There is no sacrosanct answer to this question. Over the past 15 years, we have had a number of bull markets and bear markets. Some similarities notwithstanding, every bull market is different and so is every bear market. As I had said a number of times in my blog, stock market investing is as much an art as science. The art part of the investment is very difficult; it takes professional expertise and experience. The science part is much simpler. I think statistics can provide useful guidance to stock market investing, if you use it appropriately in the prevailing market context. We will examine the statistical relationship between Sensex returns and the P/E ratio, to see how much returns can be expected (within probabilistic boundaries), when you invest at particular P/E ratios.

Relationship between Sensex rolling returns and P/E ratios

Bivariate Regression is a predictive analysis technique used to examine relationship between two variables, in our case, Sensex returns and P/E ratios. We looked at Sensex 3 years, 4 years and 5 years rolling returns from 2001 to 2016 YTD. We also looked at Sensex P/E ratios over the same period. Though Sensex P/E data was available for a longer period (from the late nineties onwards), we ignored pre 2001 Sensex because the market was distorted by Ketan Parekh scam and the dotcom bubble in the late nineties and early 2000.

We found that, there is a statistically significant relationship between Sensex rolling returns and P/E ratios. The best fit was found, when we looked at 5 years rolling returns versus P/E ratios. For readers who are mathematically inclined, the R-squared of the regression analysis was more than 60% and the P-values were low. We back-tested the predictive power of our regression model with actual Sensex 5 year rolling returns. The chart below shows, the Sensex 5 year annualized returns for prevailing P/E ratios as predicted by our model (orange line) and the actual Sensex 5 year rolling returns (blue line).

The Sensex 5 year annualized returns for prevailing P/E ratios as predicted by our model (orange line) and the actual Sensex 5 year rolling returns (blue line)

Source: Advisorkhoj Research

You can see that, our model results were reasonably close to the actual rolling returns given by the Sensex. Sensex outperformed our model from 2001 to early 2008, but that is a happy situation for investors to be in, because they would have got more than what they expected.

So, at what P/E ratio does our predictive model suggest that investors should make their entry? With all the caveats, as is always related to predictive modelling, it seems that, the 18 – 19 Sensex P/E ratio is a good buying opportunity with a 5 year investment horizon, since our model predicts the Sensex to give 15 – 17% annualized returns when the investment is made at those Sensex valuations.

If you invest in the Sensex when the P/E ratio is below 18, you can get even higher returns, but chances of getting very attractive (low) valuations are lower and therefore, investors must form a pragmatic approach. If you are investing through good diversified equity mutual funds, you can expect your fund manager to deliver a few extra percentage points of alpha, over and above the benchmark returns. We ran the regression model with a 4 year rolling returns versus P/E ratios and found that, the results were statistically significant (R squared of more than 50% and low P-values).

The predictive model is similar to the model with 5 year rolling returns. Therefore, you can apply the same investing rules, even if you have a 4 year investment horizon. We also ran the model with 3 year rolling returns but we saw that, the predictive power of the model weakens considerably. Therefore, a 4 to 5 years investment horizon is more suitable, as per our analysis.


It is often said that, timing the market is beyond the capabilities of the average retail investors. It may be so over very short investment tenures. However, over longer investment timeframes, 5 years or so, as discussed in this post, if you base your lump sum investment decision on valuations (P/E ratio), you will be able to time the market and get higher returns. As suggested earlier, you can get P/E ratio data in the BSE or NSE website and base your investment decisions accordingly. Making lump sum investment decisions based on valuations is not rocket science. All you have to do is, to track the market valuations and look for suitable investment opportunities.

Mutual Fund Investments are subject to market risk, read all scheme related documents carefully.

Locate DSP Blackrock Mutual Fund Advisors in your city

Dwaipayan Bose

An alumnus of IIM Ahmedabad, Dwaipayan is a Finance and Consulting professional, with 13 years of management experience, mostly in MNCs like American Express and Ameriprise Financial, both in India and the US. In his last role, he was the Chief Financial Officer of American Express Global Business Services in India. His key interests are building best in class organizations, corporate governance and talent development

comments powered by Disqus

DSP BlackRock Investment Managers Pvt. Ltd. is the investment manager to DSP BlackRock Mutual Fund.

The philosophy of DSP BlackRock Investment Managers Pvt. Ltd. has been grounded in the belief that experienced investment professionals, using a disciplined process and sophisticated analytical tools, can consistently add value to client portfolios.

Post Your Query

You have not found the answer here? Do post your query to DSP BlackRock MF.

© 2016 Advisorkhoj - A Gamechanger Business Services (I) Pvt. Ltd. Brand - All Rights Reserved