 | « Back to article | Print this article |
Should you expect 15 per cent return over a period of 20 years? Dwaipayan Bose has the answer
Illustration: Uttam Ghosh/Rediff.com
A couple of years back, the missus and I went to our bank branch for some work. As my wife was patiently waiting for her turn (while I was playing a car racing video game on the iPad), a young relationship manager approached her with a tax savings investment scheme. My wife asked him how much returns she can expect. The relationship manager said that, since the scheme was market linked, returns can be volatile, but if my wife remained invested for 5 years or more she can get very good returns.
I thought it was a good answer but the missus, being a banker by profession, thought that the answer was vague; she wanted a number. The young man said that, the scheme had given more 25 per cent annualised returns in the past; he then decided to be conservative and de-risked the forecast by 5 per cent.
He told my wife that, even if she got 20 per cent returns, over 5 to 10 year period, it can be a lot of money. Till then, I was listening to what the young man was saying with some interest, but as soon as he made his forecast based on the past returns, I lost interest and went back to my video game.
Historical returns are no indicators of future returns
If you have invested in mutual funds, hopefully, you would have read the disclaimer that, mutual funds are subject to market risks and past returns are not indicators of future returns.
At this stage, you may ask, why mutual fund literature, print or online, talk about past returns if they are irrelevant. It is a good question, which we will address towards the end of the post, but I can give you many examples, where mutual fund schemes were not able to sustain strong historical absolute returns (in percentage terms) in the future.
Note the emphasis on absolute returns, because at the end of the day, as investors, we are interested in absolute returns. By going to Advisorkhoj MF Research section, you can see for yourself that many mutual fund schemes were not able to sustain their absolute return performance over a very long period of time.
Go to Advisorkhoj Mutual Fund Selector tool. Type the scheme name of a fund (growth option) which is at least 20 years old in the search box and click submit.
In the page that opens, click on the NAV movement tab and select "All" in the chart. Hover your mouse over the chart and you will be able to see fund NAVs on different dates in its history. Note down the NAV of the fund on the three separate dates 30/12/1996, 1/1/2007 and 2/1/2017.
Calculate the NAV growth between 1997–2007 and 2007–2017 in percentage terms. Both are 10 year periods (sufficiently long term), but the NAV growths in percentage terms are likely to be very different in the two periods. You can calculate the NAV growth of multiple schemes between 1997–2007 and 2007–2017; in most cases, they are likely to be very different.
The reason why the NAV growths are different is fairly simple. Our capital market and economy is maturing. We cannot predict the future with any degree of accuracy based on the past.
But you need to have return expectations for financial planning.
In our blog, we have said a number of times that, investment should always be goal based. You know your goal but if you do not know how much returns to expect, then how will you determine how much to invest? While past returns are not a reliable indicator of future returns, we need to have a fairly robust, yet simple to understand, framework to estimate future returns for financial planning purposes.
A fundamental basis for forming return expectation
In this post, we will discuss a fundamental basis of forming returns expectation. Readers should note that, there can be multiple views on future return expectations from market commentators, fund managers and capital market bloggers. We will share with readers, how we approach forming return expectations, which, hopefully, will be fairly simple to understand and at the same time grounded in economic logic.
Readers should also note that, in this post we are discussing very long term (at least 10 years plus) return expectations from equities. Fundamental analysis, in our opinion, works only in the long term and when we are talking about returns from equity as an asset class and not specific stocks, our horizon should be very long term.
Finance theory teaches us that, equity returns are based on two factors: earnings (EPS) growth and P/E multiple. Suppose price of a stock is Rs 100. The EPS of the stock is Rs 10. The P/E ratio is 10.
If the EPS of the stock grows to 11 and the P/E ratio does not change then, the share price will be Rs 110 (EPS 11 X P/E 10) and the return will be 10 per cent.
If the EPS grows to 11 and the P/E ratio also grows to 11, then share price will be Rs 121 (EPS 11 X P/E 11) and the return will be 21 per cent. Therefore, in order to forecast returns we have to forecast EPS growth and P/E expansion.
Many years ago, a CIO of a well known asset management firm in the US taught me a simple formula for equity returns expectation, which I want to share with you:
Returns of a diversified fund = Real GDP Growth Rate + Inflation + P/E expansion delta + Market Segment risk premium + Fund Manager Alpha
The formula may look too long and complicated, but it is not. The last two items of the formula are fund specific and we will discuss them later, but the first three items can help you form market return expectations.
The first three items of the formula are grounded in the principles of fundamental analysis. The first two items, real GDP growth and inflation are the fundamental drivers of EPS growth of the overall market (the market, after all, represents the economy of the country) and combined with P/E expansion of the market (benchmark) determine the returns of the market.
US equity market example
Let us see if this formula worked in a mature market like the US. We are obviously more interested in the Indian market, but if the formula works in the US, then we can expect it to work in India as well in the long term. Since we are discussing market returns (and not fund returns), let us drop the last 2 items in the formula.
The market return formula is much simpler:
Market Returns = Real GDP Growth Rate + Inflation + P/E expansion delta
As per well known macro economics portal www.tradingeconomics.com, the real GDP growth rate of the US over the last 30 years was 2.6 per cent. As per www.inflationdata.com, the long term inflation rate in the US (over the last 30 years) was around 3.3 per cent.
Let us now discuss the P/E expansion percentage in the US stock market.
The S&P 500 is the most popular benchmark of the US stock market. Over the past 35 years or so, the P/E multiple of S&P 500 expanded at a CAGR of 3 per cent (the S&P 500 P/E on January 2017 was around 22, while it was 7.7 in January 1982).
Adding the three items, real GDP growth rate, inflation rate and S&P P/E expansion, the long term S&P 500 returns should have been = 2.6 per cent + 3.3 per cent + 3 per cent = 9 per cent (approximately). Does this match with the actual long term returns of S&P 500?
The average annualised long term return of S&P 500 over the last 30 years was around 10 per cent. The formula prediction was fairly close to S&P 500 actual returns. I was actually quite happy that, the formula underestimated the returns slightly; I am very happy when my investments perform above my expectations.
Long term equity returns in India as per said formula
We saw that the formula worked quite well in back testing long term US market returns. The appeal of the formula, in my opinion, lies in its simplicity and its intuitive (common sense) basis. Let us recall the formula and try to form reasonable equity return expectations in India.
Equity Market Returns = Real GDP Growth Rate + Inflation Rate + P/E expansion delta
Our GDP grew by more than 7.5 per cent in FY 2016. Multilateral agencies like the IMF were forecasting the Indian economy to grow by 7.6 per cent in FY 2017. The recent forecasts have been lowered a bit because demonetization move of the Prime Minister is likely to impact economic activity in this fiscal, but most multilateral agencies like IMF, World Bank etc expect India's GDP is likely to grow at 7.5–8 per cent in the longer term.
Can we sustain 7.5–8 per cent GDP growth for a long period of time? If China could sustain average 9 per cent GDP growth for 25 years, why cannot we? We have discussed in our blog a number of times, why 7.5–8 per cent long term GDP growth rate is feasible. Given the stage of economic growth and our demographic profile, there are enough reasons to be optimistic about our growth prospects.
Let us now discuss inflation. The RBI set a long term inflation target of 4 per cent. While a 4 per cent inflation target is laudable, our view at Advisorkhoj is that, it may not be feasible because of two factors.
As a country we are dependent on imports to meet our energy (crude oil) needs. Over the last couple of years, we had a bonanza in form of low crude prices, but as long as we are dependent on imports, inflation is not within our control.
Also, our agriculture is monsoon dependent. While the rain gods were benevolent this year, our dependence on monsoon makes it difficult to target and control inflation.
High inflation in past was also due to certain structural factors, which the Government is trying to address through reforms. Combining all the factors, our estimate of medium to long term inflation rate is around 4.5-5 per cent (slightly higher than RBI target).
Let us now combine real GDP growth and inflation to see what part of our formula tells us.
Real GDP growth + Inflation = 7.5–8 per cent + 4.5 – 5 per cent = 12–13 per cent
To estimate equity market returns as per the said formula, the only missing piece is the expected P/E expansion. The P/E expansion CAGR in the US, as discussed earlier, was around 3 per cent over the last 35 years or so.
Valuation (P/E ratio is a valuation metric) is a very controversial topic. Over the past two years, I have heard conflicting opinions from market experts on whether our market is over-valued, under-valued or fairly valued. I have heard / read market experts say, "India is a great growth story but the market may be a little overvalued. You may see some correction in the short term".
I asked an expert, if the expected correction would be related to valuations or driven by liquidity and risk sentiments? The expert did not have an answer; I do not think that we should be concerned what the market will do in the short term.
Fair price of stock, as per finance theory, is the discounted value of its future cash flows. Valuation is therefore, simply a reflection of the market's perception of future growth potential of a stock.
If equity investors in India, which arguably are at a much younger stage of growth compared to the US, are as optimistic about prospects of Indian companies as US investors were about US companies over the last 35 years, then significant P/E expansion is possible in India as well in the long term.
At the same time, since the Nifty is already trading at a P/E of around 22, you are also justified in questioning if 3 per cent P/E expansion CAGR is sustainable over a very long period of time. If we think 3 per cent P/E expansion CAGR is too optimistic, let us assume that the P/E expansion CAGR will be only 1-2 per cent. 1–2 per cent P/E expansion CAGR will take the Nifty P/E from its current level to 24.3 to 26.8 in 10 years, which will still be lower than peak Nifty P/E in 2007 (when it was nearly 28 times earnings). Layering on P/E expansion, equity market returns estimate will be:
Equity Market Returns = Real GDP Growth Rate (7.5–8 per cent) + Inflation Rate (4.5–5 per cent) + P/E expansion delta (1-2 per cent) = 13–15 per cent
What can 15 per cent returns do for you over a long investment horizon?
If you get 15 per cent returns over a very long investment horizon, you can create wealth for your long term financial goals. If you invest Rs 5 lakh for 20 years and get 15 per cent annualised returns, you will be able to create a corpus of more than Rs 80 lakh.
If you invest Rs 5,000 monthly through SIP for 20 years and get 15 per cent on your investment, you will be able to create a corpus of nearly Rs 75 lakhs.
Conclusion
In this blog post, we have discussed an economic framework for determining reasonable rate of returns from Indian equities, over a long investment horizon. Please note that, our economic estimate was based on certain assumptions. If these assumptions change over a period of time, you should revise your estimates accordingly.
As an investor, you should also layer on this economic estimate, fund specific characteristics, e.g. if you invest in small- and mid-cap equity mutual funds, you can expect a few extra percentage points of returns, commensurate with the risk associated with small-/mid-cap stocks relative to large cap stocks.
Finally, you can also expect some alpha from the fund manager. Remember, at the start of the article, we said that, we will discuss why we have references to historical returns in mutual fund literature later in the post. Historical returns by itself have little relevance as far as future returns are concerned, but relative to market benchmark index returns, historical returns can give investors some sense of the alpha generated by the fund manager. Higher the alpha created by your fund manager, greater will be your wealth creation over a long investment horizon.