Worried that you will never be able to invest enough for retirement? Do the results of a retirement calculator scare you? Learning about the 'bucket strategy' might help cope with these issues.
How much do I need to retire? Most people (among those who bother to ask!) are troubled by the answer to this question. The corpus required and therefore the monthly investment required seems so high that some think inflation is an urban myth! A middle-aged person who has not paid attention to retirement earlier is likely to end up with a much lower corpus than required. How does one cope with such grim realities?
At the end of the day whatever the numbers show one can only invest what one can. Efficiently investing this amount and creating a nest egg is a hard enough task. The even harder task of ensuring that the corpus outlives the individual follows. Hard because unlike other financial goals (eg. child’s education or the car we always wanted), the retirement corpus does not get spent in one-shot.
So how does one efficiently invest the corpus and draw from it to meet expenses and ensure it lasts a lifetime? Obviously like most problems there is more than one solution. The ‘bucket strategy’ is one of them. Subra suggested I make a calculator based on this resulting in a simulator and this post.
Wait a minute. My retirement is decades away. Why should I worry about withdrawal strategies now? So is mine (I hope!). Learning about corpus withdrawal strategies and playing with the bucket simulator helped me understand corpus accumulation strategies better. It also gives me some comfort, if not confidence, that even if the corpus falls short for some reason I can (try to) use it efficiently. It would probably help you in the same way. So do read on.
First let us review the methods of investing a corpus post-retirement and drawing an income from it. In all cases it is assumed that investing has been done with a good idea of the corpus required estimated with a retirement calculator.
I. Systematic Withdrawal (fixed return). This is the simplest way to do it. Invest the entire corpus in one safe instrument. In the initial years of retirement the interest earned will take care of expenses. Down the line as expenses increase due to inflation you start drawing from the principle. Ideally the lifetime of the corpus matches your own. Of course there is no way to ensure this … financially that is! So the risk of you outliving the corpus is quite high.
II. Systematic Withdrawal (averaged return). In this case you invest in different instruments. For example 40% of your corpus in FDs, 20% in a debt fund and 20% in equities. If you need 1 lakh for your expenses in the second year of retirement you draw 40% from FDs, 20% from the debt fund and 20% from equities. This is the irrespective of the performance of individual instruments. Depending on the allocation % and performance of each instrument this strategy may or not outperform the fixed return method.
III. Defined Withdrawal or Bucket Strategy. Here you invest in different instruments. However the way you withdraw from them is quite different. Let’s say you choose to put 40% in FDs which will take care of your expenses for the first 8-10 years of retirement. You then put 20% in a balanced mutual fund, 20% in a large-cap fund and 20% in a small- and mid-cap fund (this is one way to do it. There are several others). During the first 8-10 years you allow the mutual fund investments to grow. If the returns are phenomenally high in a few years you take out some profits and put them in FDs. If the returns are low or negative, you do nothing and hold. Towards the end of 8-10 years you take out enough to cover your expenses for the next 4-6 (or less) years of retirement depending on the performance of the funds. Then again you allow the fund balances to grow. Again if there is a good year or two you convert some portion into FDs and continue doing this. You can take also a call regarding transferring from one mutual fund to another. This method of not selling your MFs during bad market phases as much as possible allows these investments to grow and if played right can significantly beat the performance of both the above strategies. In this case there is no formula involved and all the decisions are taken by the investor. In this case each investment instrument is called a bucket since you hold the balances for as long as you can. The buckets will depend on the investors risk appetite. For example a debt fund can replace the balanced fund in the above example. The second bucket can also be FDs or debt funds which mature after 8-10 years and can then be transferred to the first bucket. A similar approach can also be used with the 3^{rd} bucket. However there may not be significant differences bet. strategies II and III in this case.
Here are a couple of comparisons of the three strategies made with my simulator:
Here strategies II and III use a corpus 20% lower than I. Still you can see both of them outperform I more than comfortably. Strategy I was designed to last for 25 years. The bucket strategy (blue line) in this case lasts for 22 years more, which is incredible! Before we get carried away we must realise that such phenomenal success is not always possible and depending on: investor decisions, allocation choices, sequence of returns and inflation things can go awfully bad. Here is an example.
The inflation in both cases is 8%. However unlike the first case in the first few years of retirement, returns from the ‘growth’ buckets (mutual funds in the above example) were poor year after year. I made some wrong choices and year 13 the portfolio went on a free-fall. So a reasonable understanding of the markets and when to exit or transfer holdings is the key to success.
The Bucket Strategy Simulator
When Subra suggested that I make a calculator based on the bucket strategy, I looked around for what was available. I found a wonderful free bucket investment simulator offered by ISG Planning. I decided to create such a simulator from the ground-up but with a lot more flexibility and features than this free version. I believe this is necessary to combat high inflation economies such as ours.
Here is a brief description: You have 5 buckets to choose from. You let bucket 1 be a ‘fixed income’ source (like FDs) where returns are fixed. The other buckets can be anything of your choice. You must input an average return and extent by which an asset can deviation from the average return (more the deviation, more the risk). For example mid-caps will deviate more than large-caps. A little bit more on this can be found in the excel file. Once you allocate to each bucket, you are ready to play. You advance year by year (of retirement) and take decisions: hold or transfer some amt to bucket 1. As you advance you can see the performance in the above graph. The idea of course is to beat the other strategies. As you play more and more you get better.
Bucket 1 return is fixed and the other returns are varied randomly. The random returns when collected over many years (30 or more) will resemble a ‘bell’ shaped curve with the peak representing the average return and the width of the curve (half-way from the top) representing the deviation or risk. This is known as a ‘normal distribution’.
More details can found in the excel file. It has instructions and a few more examples and a detailed read-me section. Do give it a go and let me know what you think. Much more advancements can be made. I will make it if there is enough interest. The excel file uses macros. So you will need to enable them before using. I had a great time playing and learning. Hope you do too. Feedback Please!
Latest version --> Download the Retirement Bucket Strategy Simulator (bugs fixed). Credit: V. Muthu Krishnan
Download the Retirement 'Bucket Strategy' Simulator (bugs fixed)*
(.xls file made with Excel 2002. Enable Macros)
Download the Retirement 'Bucket Strategy' Simulator (bugs fixed)*
(.xlsm file made with Excel 2007 Enable Macros)
* UPDATE: Thanks to Vandhana Rajendran for extensively testing the simulator and pointing out bugs. These have now been fixed.
Register for the Hyderabad DIY Investor Workshop Nov 27th 2016
Check out the latest mutual fund returns listing
Buy our New Book!You Can Be Rich With Goal-based Investing A book by P V Subramanyam (subramoney.com) & M Pattabiraman. Hard bound. Price: Rs. 399/- (Rs. 359/- at Amazon.in). Read more about the book and pre-order now! |
HI pattu,
I regularly follow your blogs.It s really awesome.
I also got some interest in calculator creating.
the above calculator i tried downloading , got some problem. this is my system problem i guess. I will see.
One thing i also got an interest in creating calculators on seeing you.
Your blog is totally different than the normal finance blog( I follow many blogs)
Your calculators are awesome. recently saw the debt fund vs fd.
Need to see more whenever i get some time.
It s very good blog.
Do you have a calculator with the below requiremnts
1. Rented(rest of the money invested in SIP) vs buying a house(EMI Paying monthly and house is also appreciating from the start)
HI Vignesh, Thank you. I don't have any kind of loan calculators. Perhaps for the simple reason that I don't have any. There are many rent vs. buy calculators available. I will have a look at then and if I can anything useful I will try and make one.
Very nice article! I have saved it in pdf for reading it anytime!
Thanks Ayush.
Some learnings from the simulator:
1. If the initial years are bad, it hurts more than if the returns are bad later on - because your capital would already have eroded in the initial years.
2. No one strategy is always better - there is a large amount of luck as well on how long your corpus would last !
Thanks again for a great tool.
Did my earlier comment vanish or is it being moderated?
Dear Vandhana,
I am delighted to hear from you on the bucket simulator! I spent a lot of time making this and I was disappointed until now that no one gave me feedback. Thanks I feel better now! I think your first comment didn't make it through. It not in spam either.
Yes you are absolutely correct wrt your observations. In general holding out and not selling equity during bad years will be an effective strategy. However high inflation will make that difficult.
Thanks again for spending time with the calculator.
Ok, since my first comment didn't make through, here is what I said:
Great and addictive tool 🙂 And it was a great learning as this concept was something I didn't know about. Initially, I thought it was more complex than the other tools, but once I got the hang of it, I really liked it.
Some other things I would have loved to have (maybe in future versions):
1. An ability to move money away from bucket 1 - afterall, greed can make one do that, and it is better to learn how disastrous it can be through a simulator than with real money.
2. Can we have switching costs and tax implications built-in somewhere? Afterall, each of these have different tax implications right?
3. Is there some way to see how much money I have moved from each bucket to bucket 1? Or an indicator to denote when my initial principal is fully recoevered. (assuming I haven't been moveing between buckets and have moved only to bucket 1) Basically, although it is a convoluted thought, I can be more comfortable if I know that the initial 60 lakhs that I allocated to small caps, have been fully recovered and parked in FD, so that any future losses, I can try to stay strong and wait instead of panicking.
That is terrific input. Thanks.
1. Transfers from bucket 1!! hmm.. easy enough to implement. Precisely why I don't want to 🙂
2. This is a complex exercise. Typically valid, as per current law, only to debt oriented buckets. I will keep in this mind when I make a second round of changes.
3. I have kept the initial or 0th year always visible for this However many years into retirement only inflation adjusted comparisons make sense. Will keep this in mind too,
Many thanks. Greatly appreciated.
Dear Sir,
I feel there are couple of errors (or maybe my lack of understanding) in the 2007 version which I used:
1. The formula in Bucket 3 (cell Q13 onwards) seems to be wrong. The if condition IF(trans2="bucket 3",X12) should be (trans2="bucket 3", V12).
2. Similar error as above in column R and S. Column S has error in the "trans 4 = ..." condition as well, I think.
3. The spelling of Bucket 4 in V11, X11 and AB 11 drop down is wrong. This prevents the transfer from taking place and my money is vanishing 🙂
4. In hidden cells as well Bucket 4 is spelled wrong - don't know the implications though.
Many thanks Vandhana. Let me have a look at them and get back to you.
Awesome! I have corrected all the errors and updated the links with credit to you. Thanks.
Wow ! What a marvelous piece of work !! It took me intial 5 mins to understand the calculator and once done, I am now addicted to it. Gives me a great insight !. Knew that compounding does magic but never ever thought how powerful it is till now. Thanks a lot for the calculator !
Many thanks Nitin.
Sir,
One suggestion, The large cap equity (bucket 4) and small cap (bucket 5) usually has a correlation in real life. If you can built the corr inbuilt in the excel, it will be more realistic in terms of return.
Thanks
Yes Nitin. Very true. Such correlation is built into the ISG planning calculator mentioned in the post. I am not sure how to model this correlation for Indian equity. If you have any good resource I will be happy to modify it
Please share your email Add. I will send the calculation with all the data. The correlation thus arrived should be good enough.
Sir,
Awesome calculators. Thanks a lot.
I would like to know if you have any calculator which will help to find how realistic the goal is. I am happy to use MC simulator. but it gives how long my corpus can last. Suppose at the end of 5 years, my investments have given me a return of 6% CAGR. But my expected return was 11%. This 5% CAGR difference will make a dent in my retirement corpus. How much the difference would be and how to plug the hole. I understand we take average return and at some point of time we will come to the median of return over longer period of time. instead of waiting for that to happen, I would like to make additional funding so that I achieve my goal with relative ease. For example, once in 3 years, i would like to check if I can achieve the target corpus with the current CAGR of my portfolio. If not, I would like to know how much I need to fund.
As we assume so many variables, i may feel little more confident if I know this, right? Please share your views sir.
Dear Sridevi,
Thank you. What you are saying is often in my mind too. When there is volatility involved this is tough to implement. one way of course is to make the client to rebalance to maintain desired asset allocation. If the 6% is because of equities doing badly then rebalancing will increase the investment in equities. Of course the client must be mature enough reg. this. It is impractical to expect fresh investments to make up for the cagr difference. By definition equity investing will result in a few years of poor returns. I will soon post a new version of my rebalancing simulator where the CAGR can be calculated after rebalancing. This will reduce volatility and increase returns.
Like you note in your website, "Investing is a marathon not a 100m dash".It is quite all right for the marathon runner to take a drinks break and resume 🙂
Thanks sir for your response. As there are more assumptions in the retirement calculation where no one has any control, one of my customer asked me this question. Fair enough, if we get average return or expected return in the long run, but, why not, " what if we do not achieve the said corpus?" This question made us to work on this front too. Please let me know when you release the calculator. Once again thanks a lot for all your efforts.
Thanks for your response. Yes it is a very important problem to ponder. If you look at my VIP vs SIP calculator there are certain scenarios where VIP would involve much greater investment, but for a lesser return than SIP all other things being equal. So more investment is not the answer. Also,tax and exit load will eat into returns if we tinker with the portfolio. In my view disciplined threshold rebalancing (rebalance only when one asset increases more than 5%) is a good way to ensure the fruits of compounding are preserved (you can see results of this feature in the rebalancing simulator available in the blog).
I shall of course let you know when I post the new version. You could also consider subscribing for the posts 🙂
chief, this is with regards to worksheet no 1 [input] of the calculator vis a vis the calculation for the annual expenditure for the first year of retirement. for instance, i entered 30,000 as my current monthly expenditure, 10 years as the time to retire, and an inflation rate as 8 per cent, the calculator says the annual expenditure for the first year of retiremnt would be 8,33,390. but, the following calculator http://www.1728.org/compint.htm gives the figure of 7,77,210. why this discrepancy?
The discrepancy is deliberate. I usually calculate for n+1 years. . For n=8, it is 777210 and for n=9 it is 8,33,390. This extra year is a safeguard and a cushion.
ok. but, after posting the question, i realised that it was only a simulator. i had been playing around with it yesterday. it was quite a gud learning process. tnx
Thank you. I use this for all my retirement calculators. Just assume the investment starts from Jan of next year. The calculation is then exact.
Great Tool. Thanks Pattu
Many thanks.
Hi Pattu, you have made my retirement planning easier. Great tool, appreciate your hard-work, cant thank enough.
Many thanks. Good to know.
Did use the simulator with various combinations and it did scare me, couple of times the CAGR return from equity (Bucket 4 & 5) was negative and the corpus did not last desired period, the returns were less than fixed return systematic withdrawals. This tool does indicate that one would end with begging bowl in old age, very scary indeed.Thank you very much for your time and effort in educating us and that too free.
Agree that it can be scary. However if you book profits often and have more secure buckets it is possible to manage and put off buying an annuity for as long as possible.
pattu, could you look into following (am using excel 2007 version):
1. H19 of input sheet is not taking care of B27, B29 and B30. withdrawal from bucket 1 will be after spending B27 and B29 right? If so, bucket1 will last more than what is shown here.
2. at year 7, B19 is 2.98 and O19 is 5. When I move to next year, it says that insufficient funds in bucket 1 though bucket1 would be sufficient for next year.
7 2.98 9.0% 5.5% -0.4% 15.9% 78.4% 5 15 6 8 11 7
3. If the corpus in B12 fetches constant rate of B11, then all withdrawals would be met and there is no need to worry right? However we have a lower corpus as entered in B20 and hence we need to allocate buckets and try to get a higher rate. It might be beneficial to display systematic withdrawals using B20 instead of B12 as the person has only B20 amount. This would also provide a comparison as to how faster fixed returns vanish compared to the volatility of equities. Just a thought. Also the systematic withdrawals would need to take into account the extra income earned as well.
4. The number of years payment can be made from annuity-calculator-April-2014.xlsm shows more years for the same corpus, rate of interest and inflation. For example, 33L with 9% interest and 8% inflation gives 14.63 for 2.4L as first year amount. The same values entered in B12, B11, B9 and 20000 in B5 shows 12.5 years in the graph for systematic withdrawals. Is it something to do with montly withdrawal versus withdrawing at start of the year?
I can say I am educated, illiterate in financial planning, today i have invested Rs 100000 in CD in axis bank having gone through the difference between FD and Mutual fund I feel i have done mistake could you enligh to me... in near future I will be receiving Rs 20 lakhs as retirement proceeds I am in dilemma as to what and how safely I have to deposit its is a big question Mark, My requirement annually is rs 3.95 lakhs what to do...the facts in retirement bucket is helping me
Excellent, I look forward to learning from this blog.
Thank you.
One word, excellent work. How come I miss this one for long time? Yes as everyone said it is very good tool to estimate how long the retirement corpus last.
Something I noticed. Not able to enter decimal % like 6.5% for fixed income. Also, I don't see any place to enter tax % for each bucket. I assume that one has to enter the expected return after tax.
I played with couple of different rates. Surprising to note that higher the standard variation , the amount last longer than the lower standard deviation. But I may be lucky in that simulator, it may be other way around as well.
It is good ( at least in the simulator), I could achieve the target with less than the expected retirement corpus provided in the first page.
Once again thank you for providing this tool.
Thank you. You can enter the decimals. It will just not show but will be used. Yes post-tax. There are several possibilites. Sometimes it is just luck.
Great tool, something I have looking for since typical Monte Carlo simulators don't allow drawing from different "buckets". Also addictive, I have run over 50 iterations with different returns and STDEVs.
I didn't believe my funds were as volatile as the simulation until I went back to look at year by year returns. Its a good lesson in harvesting returns from higher to lower volatility funds, also in not overspending in early years.
Thanks!
Respected Pattu Sir,
Gr8.Excellent.This is very useful for those retirees who could not create a sufficient retirement corpus unfortunately.Best of luck to such retirees.
Pattu sir,
I have not worked on this but thought it as an input. After retirement say if a person takes a early retirement at 40. The first 10 years(41-50) cash flow should be managed to fixed returns and then for the rest of the 30 years currently he should have the corpus say assuming a return of 12% the rest of the corpus grows in equity. Then the strategy should be sliding window syndrome, if on a year the equity returns is more than 12% take the money that is needed for the 10th year and put it in fixed return bucket, so that you always have a cashflow for 10 years. Now say for 2 years the returns is -ve or < 12% you will have a cash flow only for 8 years and if there is a good year and again fill the bucket so that cash flow is maintained for 10 years and you are taking from equity only when the returns is more than your expected assumption.