The Mutual Fund Show: Does Buy And Hold Always Work?
Buy and hold may not always be the best strategy for a mutual fund investor. Here's why.
Is buy and hold the best investment strategy for a mutual fund investor? Not always.
It may not work if an investor has to take money out periodically to meet expenses, according to Vijai Mantri, co-founder and chief investment strategist, JRL Money. He explained citing examples:
Assumptions:
Investor buys a Sensex-mirroring equity large-cap fund or an exchange-traded fund.
Starting corpus is Rs 1 crore.Annual withdrawal of 8% of starting corpus, beginning from the first year in equated monthly withdrawals of Rs 66,667.
Withdrawal increases by 6% a year to account for inflation.
Investment is aimed to last 20 years of retirement.
The reason why this does not work is simple, according to Mantri. Investments and withdrawals work are exactly the opposite of a systematic investment plan, he said.
For a fixed amount, the number of units withdrawn in the years the market corrects is fairly high, depleting the investable corpus, he said.
In the 40 years since the Sensex was launched, 26 years were great for buy and hold, Mantri said. But investors following this strategy would have faced challenges in 14 years or 35% of the time.
For pure equity fund investors withdrawing money annually, the corpus, the amount taken out periodically and the started point play an important part in the eventual outcome.
Watch the full interview here:
Here are the edited excerpts from the interview...
Vijay, what is this whole buy and hold strategy about? Where and which instances or periods have you found that it has not worked?
VIJAY MANTRI: It was a sort of conventional wisdom to assume that when you are investing in blue chip stocks or investing in index or investing in equity mutual fund should just buy, fill it and forget it. You should not even touch or look at the investment... I was also one of the advocates of saying buy and hold, just don't touch your portfolio, do nothing, and it will meet all your requirements. But when you look at hard data, the story is very different. So, I'm still a big advocate of buy and hold but not in all the situations and circumstances. So buy and hold works when you don't have to take your money out, because in the long period of time, equity outperformed every other asset category and particularly equity outperformed debt, fixed income and also does much better than inflation provided you don't have to take money out from the market at all. But suppose you take money out from the market for a specific goal, whether it is to meet some emergency or goal for child education or wedding, or most importantly, the goal for retirement because that in my opinion, the biggest call an individual could have—though there are many instances where the buy and hold actually has not worked. Everybody is not Warren Buffett that you live for 90 years, and you don't have to touch a corpus. Normal folks, when they look at investing, there are two legs to the transaction. One is investment, the second leg of transaction is disinvestment. So, until both the legs are done, the transaction is not completed. So, when you take money out from the market for a specific goal, then there many incidences and there are sizable incidences where the buy and hold strategy has not worked for the investor and we would try to examine these in next 15-20 minutes.
Before we get to the hard data, what you're effectively trying to say is that if people make an investment or invest money into funds with a purpose of using them at some point of time and not necessarily keeping it as an investment alone, then the buy and hold strategy, sometimes works but sometimes doesn't work as well. In that it may not give the best outcome.
VIJAY MANTRI: That's very true and our research shows that. We went back to data from 1980 till 2020, we looked at the BSE Sensex data for almost 40 years, and the results are very interesting. So, if you allow me, I can explain you how we have done it. For instance, what we assumed that let's look at the one goal, that goal is retirement. We assumed that when people retire, they at least have a requirement of 20 years given a typical Indian would retire at the age of 60 and then they would look at the corpus to last for at least for 20 years, then actually the corpus is meeting his requirement where the entire retirement expenses get funded through disinvestment from your corpus. So, what we have done here is a very simple exercise. We looked at the 1 crore corpus which the investor could have invested, or he could have created over a period of time. On the day of his retirement, let's assume he has a Rs 1 crore corpus and then from month one, the investor, in year one takes out 8%, so Rs 8 lakh, the investor takes out because the Sensex has delivered 16% CAGR over a 40-year period. So, let’s look at half of that. Assuming that the investor takes out Rs 8 lakh per annum in equal monthly instalments of Rs 66,667. So, from month one to month 12, the customer takes out Rs 66,667. From month 13 to month 24, this amount goes up by 6% and then 6% year after year because we have to account for inflation as well. We have done this exercise that you start with 1 crore corpus; year one you take out Rs 8 lakh, year two you take out 8 lakh plus 6%, year two plus 6%, year three plus 6%. So, we keep increasing your withdrawal by 6% per annum. We have done that number. Now, when we apply this number on BSE Sensex from 1980, let's say somebody started with a corpus on April 1, 1980, what would have happened to that person as on date or on April 1, 1981? Every year we have taken the first day of the financial year. So, there are 41 observations that we looked at and we could share what has happened in these 41 years where in two-third of the years, there were great years where not only the investor has taken out a lot of money but the corpus as on date is very sizeable too. There are some incidences where almost in one-third of the incidences, the corpus has not lasted 20 years. So, the buy and hold strategy, which has worked very well when you don't have to take money out, has not worked for you one-third of the times when you have to take money out to meet your retirement needs.
Vijay, what do you mean by the corpus didn't last? Did people stop investing, and then 20 years post that date, how did you come about this one-third of the instances?
VIJAY MANTRI: We assumed that somebody started with a 1 crore corpus and suppose somebody started on April 1, 1980. As on this day, the person has taken out Rs 23 crore in the last 40 years, but the corpus is still more than Rs 100 crore. If somebody started from April 1, 1980. Suppose somebody started on any date with any other amount of the corpus and the amount taken out could differ, so I call it a ‘great year’, and there are some 25 great years. 1982 and 1990 those were years which if you'd invested and then taken money out then still the corpus would be anything between some 23-100-crore-plus. These were the great years. In between, there were two years, 1986 and 1991 where the corpus has finished by now, but the corpus has served you for 27-28 years. So, if anything, it lasted more than 20 years. That is great because you planned for 20 years. So, year 1986 and year 1991, were the periods where suppose you started at that time, then the corpus lasted 27-28 years, but suppose you started on April 1, 1992 or 1993 or 1994 or 1995 or 1996, or 1997, or 1998, or 1999, or year 2000 or year 2006, 2007 and 2008, then the corpus has not lasted 20 years. In some of the incidences the corpus has just lasted seven years. Suppose you started in on April 1, 1992, then your entire corpus got over in October 1999. So, it varied from seven years to 17 years—during that period from 1992 till 2000. Then again, you had a great period of 2001, 2002, 2003, 2004, 2005 and even to a certain extent 2006 but suppose you started in 2007 and 2008, then in those two years of a starting period, the corpus got over in around 12 years. So, it didn’t last 20 years. We're not talking about Harshad Mehta days of 1992, where the market ran crazy valuations, we are talking about the current market, just not 10-12 years back and even if you see data for 2010 and 2011, if you started a corpus in 2010, the total corpus available to you for 2010 or 2011 is around Rs 34-50 lakh. It means that you just covered 10 years of expenses. You have 10 more years to go. The Rs 8 lakh today, would have been Rs 11-12 lakh per annum right now. So perhaps the corpus may last four to five years. So, as recent as 2010-11, the corpus would not last more than 15-16 years. So, you could safely presume that out of 41 years, there are 14 years in which the corpus has not lasted 20 full years. It means that around 34% of the time, the buy and hold strategy has not worked when you have to fulfill your retirement needs.
What it implies is, when you start and when you take out money, what phase the market is at that time, matters a lot.
Vijay, because people and an average retail investor or maybe even a sophisticated investor is no one to be able to predict that things will definitely last 20 years or not, are there some tricks how one can make this happen?
VIJAY MANTRI: That's very true and you're absolutely right. It means that two-third times it worked. One-third is a big number in percentage terms. Aesthetically, we could say that 16% is CR Sensex delivered in last 40 years but that is statistics. When it comes to your number you could be in one-thirds. So, it's like when you go for an open-heart surgery, the doctor will say, the chances of you dying in the operation theatre is one out of a million but for you, it is 50%. So, the statistic and a large number is a number but when it comes to you, it could be tragedy, or it could be anything else. So, it is very important to look at these data very closely and figure a way out what can be done. What would happen in the future nobody can predict because it’s a random walk. It is not the question of market’s destination after five-years, 10 years or 20 years. Everybody knows that over a 10-year, 15-year or 20-year period, the market would be much higher from these levels. There's no question about it and it is the last 40 years of data in India or many years’ data in the U.S. market do suggest that the long-term equity market does very well. But what is most important is, what happens when you are taking money out? What phase is the market at that time, so you cannot be sure at all about what the journey of the market from here onwards for the next three 3, 5-10, or 20 years is going to be. Given that kind of a situation, what an investor needs to do, I think is a very important question to be asked. If the buy and hold situation, data clearly suggests has not worked and mind you, those were days, the last 40 years in 1992 to year 2000, the interest rate was 12%. So, if you didn't have money in equity, you could fund through debt. Today's situation is very different. Today, if that kind of situation were to happen, then even the debt portfolio, would not able to take care of your retirement needs. So yes, we need to create a mathematical formula based on culling the past data, looking at the past data, churning the past data, going deep into the past data and figure out what my best strategy is where whatever happens in the past, if suppose the same thing gets repeated, the one-third chances come down to 0%. Yes, there’s a possibility, this can be done, but perhaps I think we need to have a separate show because I have worked on that and there also, a lot of explaining is required but perhaps we'll take some time and then come back to the viewers saying that we need to fund for retirement, and we need to keep the buy and hold strategy. Then here is a mathematical formula that would satisfy your needs.
Vijay, were there any specific reasons why in some of the cases, the corpus lasted for only seven, eight years?
VIJAY MANTRI: It's very interesting, there's a common thread to the entire thing and suppose you look at 1992, the market collapsed 20% but the corpus lasted seven years but suppose you look at the same data for year 2000, the corpus lasted 12 years. So, even in 2000, the market was at peak, but the market recovered very fast from 2001 and then you had a great period from 2001 to 2005-06, but from 1992 till 2004, the Indian equity market was almost flat. Why this phenomena happened?
Why did the corpus didn’t last? It is the reverse of SIP.
SIP delivered you the best performance when the market is in the bearish phase, because during that time, your monthly purchases allows you to accumulate more and more units in bad market timing, but when you are taking money out and suppose at that time the market is in the bearish phase, then the reverse happens where you take more and more units out, and when the market starts recovering perhaps you don't have too many units to meet your retirement needs. So, it is just the reverse of the market. What matters is that particularly, in the initial four or five years of a withdrawal, suppose the market is in a bearish phase, then definitely it would have an impact on your corpus.
There was this instance of how an active fund from the house of ICICI Prudential has expense ratios which are similar or fairly low and comparable to maybe even passive funds so to say, maybe not comparative, but pretty low and therefore it kind of diminishes the differentiating factor between an active and a passive fund, if you will. Now, a lot of people I know, go about and choose the active or the passive fund options based on the expense ratio naturally so. Is that line getting blurred and do you think more and more AMCs would probably go down this path? Why or why not?
VIJAY MANTRI: The index fund, the choice of the expenses is there with the market participant. If someone wants to run it much lower, someone wants to run it at 10-15 basis points. When it comes to the direct plan or the normal plan, the choice is not there. It is a cause-and-effect relationship. So, let me explain you how it happened. In a normal plan, SEBI has prescribed the formula. First find it crude, 2.25% next, 200 crore, then 300 crore and it goes like that. It goes to 1.05% and anything about 5,000 crore then it will keep dropping by 0.5%. So, if the fund is 10,000 crore, the effective expenses is 150 to 160 basis points, 1% or 1.6-1.7%. Now, how are the direct plan expenses are calculated in the same scheme? Suppose the scheme, A has 1.75% weighted average expenses we call it TR, the total expense ratio. Out of that, whatever on weighted average basis the fund house is paying to the distributor, in form of commission, in form of event, in form of taking them out for a picnic, whatever money you are spending on the distributor, that entire thing is taken as a distribution expense. So, minus that and the rest is the expenses of direct plans. For instance, a fund house says 1.75 basis points, TR in a normal plan and that fund house pays 1.6% distribution commission, then 1.75 minus 1.6, 15 basis points would be the expenses ratio for the direct plan. If next year the fund house expenses drops down because next year, the trail commission comes down, this could be near when they are offering higher trail commission because ICICI Flexi-Cap Fund has just launched, and they paid a high commission in year one or perhaps year two. In year one, the expenses could be very low in the direct plan. Suppose in the next year, the broker plan expenses come down or the brokerage expenses come down from 1.6% to 1%, immediately you will see that the direct plan expenses are also going up. That is the reason you keep getting a lot of emails from the fund house where they're saying, the TR of this fund has changed this much. So, it is a mathematical formula. It is typical of a particular fund but industry level, I do not suppose so. For ICICI Flexi-Cap Fund, I'm guaranteeing you after four to five years or after even two years the expense in the direct plan will not remain 0.17%.
Can you tell us why you said that this won't happen and are you saying that the expenses of the direct plan, the expense ratio of the direct plan will move up? Why so?
VIJAY MANTRI: Yes, because their total expense is on the fund level. Now there are two plans, broker plan and direct plan. In the broker plan you pay brokerage to the broker. The ICICI Flexi-Cap Fund was this fund which got launched so perhaps they paid more money to distribute in the year one commission. So, perhaps they paid 175 even, whatever and that was the year one trail commission. Year two, the trail commission came down to 75 basis points or 1%. It means the expenses, the brokerage you paid in year one, 1.6% came down to 0.75% or 0.1%. Then your direct plan expenses will go up because direct plan is a result of normal plan minus distribution expenses. In year one, the distribution expenses were higher because they lost the NFO, year two or year three the brokerage will come down. So, the moment the brokerage will come down, the direct plan expenses ratio will move up.
But that would mean that the regular plan expense ratio should come down. Why should it result in the direct and expense ratio moving up, because that would remain a constant, right?
VIJAY MANTRI: The direct plan expense ratio is resultant of total TR. So, the TR is based on a formula, it is based on how the assets are built up and you reduce the distribution expenses. It is a resultant component of that. It is a little technical. The viewers perhaps can tweet, and we can respond to them in greater detail.
A lot of people get those emails as you said that there is a change in the TR, but people don't pay heed to it because it's such a small part?
VIJAY MANTRI: It keeps increasing. What you will see, 35 becomes 37, then 37 becomes 40 and 40 becomes 42. So, what you see, okay 40 became 42 but forgot that one year back, it was 35 basis points now it is 60 basis points. Just a small thing to the viewers. I'm going to release a separate video on this in far greater detail and data because of the challenge to the medium I could not put so much data, so viewers can see maybe 7 or 10 days later, the year-wise data of how much they have taken money out, what is the corpus today so I think that would perhaps help them.