Variance of returns

[utrecht]

I just spent a few hours running various retirements withdrawal scenarios based on variance of returns and I got quite a shock. Some of you may already know this but I was surprised at how drastic the effect is.

Here are my actual 401k returns for the past 8 years:

2001.....-3.9%
2002.....-11%
2003......34.7%
2004......16.9%
2005......14.6%
2006......16.8%
2007......14.8%
2008......-5.9%

Total return is 8.2% not accounting for regular deposits.

Lets assume there are 3 of us retiring on the same day. We all have the same amount of money and we all will withdraw the same amounts over the next 8 years.

Person A will get these exact returns as listed
Person B gets a constant 8.2% return
Person C gets the listed returns in reverse order

The starting portfolio amount and withdraw amounts arent really important since they are all the same. The final results is whats interesting / scary.

Person A ends his 8 years with $640000
Person B ends his 8 years with $736000
Person C ends his 8 years with $865000

Person C has 35% more money than Person A and they both the exact same cumulative 8.2% return over 8 years.

Having seen this, I have alot more retirement planning to do. Its quite unsettling. I had no idea the variance was this high from just a few years with only one drastically high year and one pretty bad year. Both scenarios (standard order and reverse order) start and end with a down year but that second bad year at the beginning for Person A is a killer.

Comments?
Suggestions how to dampen this effect?

[Ed_The_Gypsy]

Bingo! You just found something that eluded even Peter Lynch, the legendary manager of Fidelity Magellan! This is why Scott Burns told Peter Lynch that withdrawing 5% of a portfolio annually adjusted for inflation did not guarantee that an investor would never run out of money. 3.5% to 4% is more like "The Number", but that assumes a certain portfolio composition.

A recent thread here contained an article that said that it was safer to sell your bonds first before selling equities, to give the pot a chance to grow more or recover from an early dip.

I favor the idea of taking 4% or so of a portfolio (somewhere between 75 and 100% equities) every year and funding a 4- or 5-year-deep CD or bond ladder to smooth out the ups and downs of the market. When the CDs or bonds mature, put the cash in a money market fund for your annual expenses. You ought to set up the ladder in advance of retirement.

There are variations on this idea, but I am sure any one is an improvement on simply riding the market up and down.

I learned that I had been building my hopes on unreasonably high returns on investment. We have to figure out how to live on what we can reasonably expect to get out of our investments, not what we would LIKE to get. Wishing doesn't make it so.

Best of luck!

[Gone4Good]

Quote:

Originally Posted by utrecht Person A will get these exact returns as listed Person B gets a constant 8.2% return Person C gets the listed returns in reverse order The starting portfolio amount and withdraw amounts arent really important since they are all the same. The final results is whats interesting / scary. Person A ends his 8 years with $640000 Person B ends his 8 years with $736000 Person C ends his 8 years with $865000 Person C has 35% more money than Person A and they both the exact same cumulative 8.2% return over 8 years.

Your general observation is valid, but you might want to double check your math with respect to magnitude. I tried to recreate your experiment using a starting balance of $500K and a 4% withdrawal adjusted annually for 3% inflation. My ending values came out as:

Person A ends his 8 years with $718,281
Person B ends his 8 years with $703,973
Person C ends his 8 years with $739,681

About a 5% difference between the best and worst case scenario.

As to your question "how do you dampen this effect", I'm not sure you can. I suspect rebalancing among different asset classes will have some beneficial effect, but generally, we're stuck with whatever the market gives us.

My advice is to keep your initial withdrawal rate low and flexible.

[brewer12345]

(donning flame suit)

Buy an annuity if you cannot stand the volatility and are willing to make the considerable trade-offs involved. TANSTAAFL.

(\flame suit)

[utrecht]

Quote:

Originally Posted by . . . Yrs to Go Your general observation is valid, but you might want to double check your math with respect to magnitude. I tried to recreate your experiment using a starting balance of $500K and a 4% withdrawal adjusted annually for 3% inflation. My ending values came out as: Person A ends his 8 years with $718,281 Person B ends his 8 years with $703,973 Person C ends his 8 years with $739,681 About a 5% difference between the best and worst case scenario. As to your question "how do you dampen this effect", I'm not sure you can. I suspect rebalancing among different asset classes will have some beneficial effect, but generally, we're stuck with whatever the market gives us. My advice is to keep your initial withdrawal rate low and flexible.

My original portfolio balance and my withdrawals are quite a bit higher than yours. My assumed withdrawals are quite a bit higher than 4% because of my personal situation which includes a pension and a "DROP" account which allows me to deposit my pension checks into an account that pays 8-10% interest. It makes sense for me to not touch my pension and let the DROP account build up as long as I can make the 401k last. I want to withdraw as much as possible from the 401k and still have it last until the point where when it runs out, I will have enough in my DROP account where the 8-10% interest plus the pension check i will now start spending will equal my previous 401K withdrawals. i dont know if that made snese but in short, it means my 401k withdrawls will be much higher than 4% knowing my money will run out in 10 years or so.

The problem is that with this much variance in returns, the money could run out too soon. With 4% withdrawls, you can be fairly certain it wont run out. In my plan, I know it will...but still need it to last 10 years or so.

[utrecht]

Quote:

Originally Posted by brewer12345 (donning flame suit) Buy an annuity if you cannot stand the volatility and are willing to make the considerable trade-offs involved. TANSTAAFL. (\flame suit)

I dont think I want an annuity since they are so expensive but I do need to start researching things that can give me more consistent returns without sacrificing too much potential growth. I guess I need to look at balanced funds? Or possibly the new Vanguard annuityesque funds?

[crazy connie]

"DROP" ACCOUNT with 8-10% interest?? What? Where?

[Gone4Good]

Quote:

Originally Posted by utrecht I want to withdraw as much as possible from the 401k and still have it last until the point where when it runs out

Yup, a high withdrawal rate exacerbates portfolio volatility and increases risk. The only way to fix that without lowering the withdrawal rate (not an option in your case) is to lower portfolio volatility, which means reducing your stock allocation in the 401(k) account. Depending on how quickly you want to amortize that portfolio down, you could move to a 100% laddered approach or some hybrid.

[utrecht]

Quote:

Originally Posted by crazy connie "DROP" ACCOUNT with 8-10% interest?? What? Where?

Its part of my pension. I can either collect my pension checks, or allow them to be deposited into the DROP account. The DROP account pays 8-10% interest based on the avg return for the previous 10 years. It cant change more 0.25% per year in either direction and currently stands at the max 10%.

I want to withdraw the max amount possible from my 401k and other investments that allows it to last 10-11 years. At that point I will start collecting the pension checks and drawing the 8-10% interest from the DROP account and leave the principal of the DROP account alone so my interest payments dont drop over time.

To withdraw the amount I want from my 401k and still have it last 10 years, I need 7% returns over those 10 years and will be withdrawing about 11% the 1st year and ever increasing percentages assuming my returns are around 7%. I didnt think that would be a problem untill I started looking at this variance of year to year returns.

Now, my goal is to find some scenario that provides me the closest thing to constant 7% returns instead of some good years and bad years that equal 7% over the 10-11 years because if the bad years come first, the 7% overall return wont be enough.

I havent done nearly enough research into withdrawal aspects as I have accumulation. Should I be looking at balanced funds, income funds, the new managed payout funds or what?

[Gone4Good]

Quote:

Originally Posted by utrecht Now, my goal is to find some scenario that provides me the closest thing to constant 7% returns instead of some good years and bad years that equal 7%

I don't think such an animal exists, at least not "off the shelf". You can certainly put together a portfolio of REITs, MLPs, preferred bonds, junk bonds etc. that will yield 7% but your principal balance will still be quite volatile. The same is true for the "Managed Payout" funds. If you need a 7% return, I think you're going to have to accept some principal volatility and the attendant uncertainty about how long your 401(k) balance will last.

The trade-offs you're facing are pretty typical . . . more certainty = lower returns and lower withdrawals . . . OR . . . less certainty = higher returns and potentially higher withdrawals.

[dex]

I came up with these #'s - not too different - did I make a mistake?
Start with 1M

A
-3.90% 961,000 -11.00% 855,290 34.70% 1,152,076 16.90% 1,346,776 14.60% 1,543,406 16.80% 1,802,698 14.80% 2,069,497 -5.90% 1,947,397
B -5.90% 941,000 14.80% 1,080,268 16.80% 1,261,753 14.60% 1,445,969 16.90% 1,690,338 34.70% 2,276,885 -11.00% 2,026,428 -3.90% 1,947,397
C 8.20% 1,082,000 8.20% 1,170,724 8.20% 1,266,723 8.20% 1,370,595 8.20% 1,482,983 8.20% 1,604,588 8.20% 1,736,164 8.20% 1,878,530

[Alan]

Your general assumptions are correct but I cannot match your math.

Using your returns I calculate an effective year on year return of 8.7% not 8.2? start with $1,000 and each year apply your gain/loss and you end up with $1,947. To get to that same figure using the same return every year you need a return of 8.7%

Am I missing something

[RockOn]

Quote:

Originally Posted by . . . Yrs to Go I don't think such an animal exists, at least not "off the shelf".

If you are mid 50's and buy an SPIA and live to about 90 you get close to a 7% IRR. Some argue, but it is true.

[RockOn]

Quote:

Originally Posted by utrecht Now, my goal is to find some scenario that provides me the closest thing to constant 7% returns instead of some good years and bad years that equal 7% over the 10-11 years because if the bad years come first, the 7% overall return wont be enough.

Exactly my goal. Please let me know when you find it. I'll pay you a generous finders fee .

[Lsbcal]

Quote:

Originally Posted by utrecht ...Here are my actual 401k returns for the past 8 years: 2001.....-3.9% 2002.....-11% 2003......34.7% 2004......16.9% 2005......14.6% 2006......16.8% 2007......14.8% 2008......-5.9%

Here are returns for Wellesley fund with around 38/62 allocation:
2001.....7.4%
2002.....4.6%
2003......9.7%
2004......7.6%
2005......3.5%
2006......11.3%
2007......5.6%
2008......-3.3% YTD

and here are returns for Wellington fund with around 65/35 allocation:
2001.....4.2%
2002.....-6.9%
2003......20.8%
2004......11.2%
2005......6.8%
2006......15.0%
2007......8.3%
2008......-2.8% YTD

So Utrecht, not knowing your 401k allocation I am guessing you might want to adjust to more evenly balance stocks and bonds. My own allocation is 55/45 and is about -1.8% YTD.

[utrecht]

Quote:

Originally Posted by lsbcal Here are returns for Wellesley fund with around 38/62 allocation: 2001.....7.4% 2002.....4.6% 2003......9.7% 2004......7.6% 2005......3.5% 2006......11.3% 2007......5.6% 2008......-3.3% YTD and here are returns for Wellington fund with around 65/35 allocation: 2001.....4.2% 2002.....-6.9% 2003......20.8% 2004......11.2% 2005......6.8% 2006......15.0% 2007......8.3% 2008......-2.8% YTD So Utrecht, not knowing your 401k allocation I am guessing you might want to adjust to more evenly balance stocks and bonds. My own allocation is 55/45 and is about -1.8% YTD.

Why? Assuming a starting balance of $10000 and no regular deposits, my 401K allocation would have resulted in a current $19474 balance and an 8.2% return.

The 1st allocation you listed ends in a $15598 balance and 5.4% return and the 2nd one ends with $16848 and 6.4% return. Why would I want one of those allocations over my current one while Im still accummulating? I dont mind the volatility much right now and have 7 years to go.

[utrecht]

Quote:

Originally Posted by RockOn If you are mid 50's and buy an SPIA and live to about 90 you get close to a 7% IRR. Some argue, but it is true.

Whats an SPIA?

[utrecht]

Quote:

Originally Posted by Alan Your general assumptions are correct but I cannot match your math. Using your returns I calculate an effective year on year return of 8.7% not 8.2? start with $1,000 and each year apply your gain/loss and you end up with $1,947. To get to that same figure using the same return every year you need a return of 8.7% Am I missing something

Using the XIRR function on excel....

Starting balance $10000
Ending balance $19474

Starting date 12/31/99
Ending date 6/13/08

I get 8.2% return.

[utrecht]

Quote:

Originally Posted by dex I came up with these #'s - not too different - did I make a mistake? Start with 1M A -3.90% 961,000 -11.00% 855,290 34.70% 1,152,076 16.90% 1,346,776 14.60% 1,543,406 16.80% 1,802,698 14.80% 2,069,497 -5.90% 1,947,397 B -5.90% 941,000 14.80% 1,080,268 16.80% 1,261,753 14.60% 1,445,969 16.90% 1,690,338 34.70% 2,276,885 -11.00% 2,026,428 -3.90% 1,947,397 C 8.20% 1,082,000 8.20% 1,170,724 8.20% 1,266,723 8.20% 1,370,595 8.20% 1,482,983 8.20% 1,604,588 8.20% 1,736,164 8.20% 1,878,530

Did you make a mistake? Yes and no.

If all 3 scenarios have the same total return and there are no withdrawals, the final balances will be the same no matter what order the returns come in.

The reason "A" and "B" have the same final result is because the returns are the same just in different order. When you account for very sizeable monthly withdrawals, it makes a huge difference in the final balance.

The reason that "C" has a different final balance is because the year 2008 returns arent for a full year. They end on 6/13/08. If they were for the full year, it wouldnt be a total return of 8.2%. So thats throwing your "C" numbers off a bit.

If you play around with the XIRR function on excel, you'll see that when making withdrawals the number change dramatically.

[Lsbcal]

Quote:

Originally Posted by utrecht Why? Assuming a starting balance of $10000 and no regular deposits, my 401K allocation would have resulted in a current $19474 balance and an 8.2% return. The 1st allocation you listed ends in a $15598 balance and 5.4% return and the 2nd one ends with $16848 and 6.4% return. Why would I want one of those allocations over my current one while Im still accummulating? I dont mind the volatility much right now and have 7 years to go.

Your original post said you were using the returns to see how a retirement portfolio would turn out. And you were kind of surprised to see that if you have poor returns early on then things don't turn out so rosy. If I've interpreted you correctly, then you would want to smooth out those returns to insure portfolio survival perhaps at the expense of a somewhat lower long term growth rate if we have a good market.

Of course, if you are talking about the accumulation phase then go ahead and go for the growth. I'm not sure you can have it both ways though. If you had a high equity allocation and retired in 1929 you might be in trouble. But if you had a high equity allocation in 1929 and continued to fund it through the accumulation phase for many years, well that is another story that could end happily after all. Of course, if you could not fund the portfolio because you're accumulation phase was interrupted by unemployment in the 1930's then that would be a bit more tragic of a story.