PBS Frontline - Can You Afford to Retire?

[gindie]

Tonight, Tues. 5/16  Check your local listings

http://www.pbs.org/wgbh/pages/frontline/retirement/

 

[dory36]

Be afraid... be very afraid.

From their press release:
"To maintain their standard of living, experts say, Americans with pensions or 401(k) type plans need to accumulate at least six to ten times their annual pay before they reach retirement."

If these experts believe retirees will live on 25-35% of their preretirement annual income, this is about right, but I suspect that's not their message.

 

[kaneohe]

I 'm guessing that they are assuming reasonable SS in addition.

 

[HFWR]

Maybe if you're 65yo, have a pension w/health insurance, SS and Medicare...

 

[perinova]

if at 62 they have nothing much saved (like some data shows) they will live with that much. and only after 10 years of saving about 50% of their income.
I think I'll watch that show

 

[Cut-Throat]

Also from the press release:

"The nightmare I have," says pension expert Brooks Hamilton, "is the vision of people ... outliving their retirement income, and being down to Social Security, and inflation eats them up." Hamilton feels that the impact of this crisis may go beyond individuals' standard of living. "What holds up our economy ... is consumer spending. When retirees are 20 percent of the economy and run out of money, then `poof' there goes the economy."

Of course the one factor that will save this is immigration. Bush is now sending the national guard down to 'protect' our borders. You know how well his timing works! :

 

[youbet]

Maybe if you're 65yo, have a pension w/health insurance, SS and Medicare...

Agreed HFWR. On this board we, of course, think in terms of ER supported by personal resources. Government policy, for the wave of retirees coming over the next couple of decades, is to retire at 66, collect unreduced SS and use Medicare for health insurance.

Not exactly the way I'm going to do it but, apparently the way many will be doing it.

 

[youbet]

Of course the one factor that will save this is immigration.

Help me with the math here C-T......   Hamilton refers to a situation where retirees are 20% of the US population.  In order for this to be a smaller percentage, say 10%, does that mean we would have to double the US population via immigration for the math to work out?  If so, it seems like a lot of additional folks to arrive, establish residency with homes and infrastructure, put to work, etc., in such short order.

Perhaps additional fixes need to be implemented beyond open immigration?

 

[Cut-Throat]

Help me with the math here C-T...... Hamilton refers to a situation where retirees are 20% of the US population. In order for this to be a smaller percentage, say 10%, does that mean we would have to double the US population via immigration for the math to work out? If so, it seems like a lot of additional folks to arrive, establish residency with homes and infrastructure, put to work, etc., in such short order.

Perhaps additional fixes need to be implemented beyond open immigration?

Let me help you here - Yes, our economy is all about growth, without it markets stagnate as Japan, Germany and Italy has. Eventually our 'growth' will come to an end. - When exactly when would you like to see us stagnate? - I'm hoping after I'm dead.

So either you convince all of the young mothers in this country to start popping babies in the next year or two or you open up immigration.

What are your additional fixes?

 

[youbet]

Thanks C-T!  As always, your remarks can be counted on to be well thought out, free of emotional content and helpful to us all!

 

[Cool Dood]

Productivity gains are more important than population gains for growth, and will be even more so in the future. One smart guy who invents a machine that makes car parts instead of a hundred thousand humans making those car parts is doing more for the economy than thousands and thousands of immigrant laborers.

Heck, most unskilled immigrants probably cost the economy a bit more than they add to it, or barely break even.

 

[Cut-Throat]

Productivity gains are more important than population gains for growth, and will be even more so in the future. One smart guy who invents a machine that makes car parts instead of a hundred thousand humans making those car parts is doing more for the economy than thousands and thousands of immigrant laborers.

Heck, most unskilled immigrants probably cost the economy a bit more than they add to it, or barely break even.

But that one smart guy is not going to pay for your Social Security or buy all the car parts that are made!

Also Japan has had the best cars in the World, and that has not helped them save their economy

 

[obryanjf]

Help me with the math;

using 10 times the salary (10x) and 4% withdrawls (using ER SWR rules) yields 40% of pre retirement salary which is what many here claim to be their ER reality.  Add in some SS and one gets to 50% or maybe 60%.  Not catastrophic to ER types, maybe  a pinch to non ER types.  Some emotional adjustment likely to be needed.
Using a non fixed withdrawl method may give up to 5% withdrawls (based on my playing around with the numbers, with some 'minimal' risk on purchasing power).  

Add in the fact that the 4% SWR number is good for what, a 30 or 40 year simulation and the number could reasonablely be increased to maybe 5% for a much shorter life after ER (no late retirement).  

Now the 6 times salary may be a real stretch, but the 10 times salary with SS, faith in govt and a decrease in retirement years might work.

job

 

[jdw_fire]

Help me with the math;

using 10 times the salary (10x) and 4% withdrawls (using ER SWR rules) yields 40% of pre retirement salary which is what many here claim to be their ER reality.  Add in some SS and one gets to 50% or maybe 60%.  Not catastrophic to ER types, maybe  a pinch to non ER types.  Some emotional adjustment likely to be needed.
Using a non fixed withdrawl method may give up to 5% withdrawls (based on my playing around with the numbers, with some 'minimal' risk on purchasing power).  

If you use an inflation protected immediate annuity you can push this to 6% W/D for a 66yo couple as long as both are alive (you will lose half when the first dies) and 6.37% W/D for a single 66yo male with "no" risk on purchasing power.

Add in the fact that the 4% SWR number is good for what, a 30 or 40 year simulation and the number could reasonablely be increased to maybe 5% for a much shorter life after ER (no late retirement).   

You better plan on living to 100yo so no help here

Now the 6 times salary may be a real stretch, but the 10 times salary with SS, faith in govt ... might work. 

Faith in the insurance companies that sold you the annuities might even make 6 times salary work

 

[perinova]

Using a non fixed withdrawl method may give up to 5% withdrawls (based on my playing around with the numbers, with some 'minimal' risk on purchasing power).

Daddy O
Can you explain the method? Are you thinking of adjusting withdrawal rate of the current portfolio value?

 

[obryanjf]

I'll try to be concise and clear.  I modeled 5 withdrawal methods; variable withdrawls (% of portfolio value), fixed inflation adjusted (like FIRECALC), Hybrid (1/2 fixed + 1/2 variable), ESRBob and Guyton. 

During bad withdrawl years, I artifically keep the inflation adjusted withdrawls at 75% of the initial withdrawl; without this, the inflation adjusted withdrawls can get pretty ugly in inflation adjusted terms. For a 40 year simulation, the non fixed models yield withdrawls of around 5% +/-.  For 30/50 year simulations everything shifts a little left/right. The failure graphs show each model over a range of withdrawl percentages and for a time frame, 30, 40 or 50 years.

I look at the failure rates for various timeframes, also I tend to look at the inflation adjusted withdrawls as a percentage of the initial withdrawls (WDi/WD0 %) and the withdrawl % (WD%) to see how they behave over time. 

The variable, Guyton and ESRBob methods are best to provide higher withdrawls but are highly volatile, and the one I like is the Hybrid due to much lower volatility and stable withdrawls and withdrawl %.

See the graphs at with links below (see if it works)

For the MinMax graphs; Let me explain, the WDi/WD0 % (left axis) is graphed for each of the 75 runs.  This represents the % of purchasing power of the withdrawls over the years.  It starts at 100% and gyrates all over the place as inflation and the protfolio change.   Note the minimum if around 75%.  The average over all 75 runs is the bold red line.  Now the var graph is noticeably more chaotic than the hybrid.  Notice the fixed stays at 4% WD% until some failures at later years.  Review the other graphs and draw your conclusions.

The next line of interest is the bold red line with yellow dots (right axis).  It is the Maximun WD% over all runs.   When it starts to spike you are generally starting down the path of running out of money for a particular year.  Also my premise is when retired, I would start to get nervous if my WD % is running hot, like 15 or 20%.  Again draw your conclusions.

BTW, the growth and mechanics of withdrawing are as described by Guyton in his paper.  Do not expect my results to perfectly match that of FIRECALC or ESRBob.  dory36 or ESRBob may or may not agree with how I implemented their strategy.

job 

this was a suggested method of doing large attachment.  We'll see if it works.

 

[samclem]

Daddy O,

Wow, that's some work. I'm still digesting it, it's too bad the chart/axis labels didn't txfr with all of them.

The attachments opened fine for me (Windows XP, Mozilla)

Basic question: What data did you use for the modelling of inflation/asset porformance, etc (actual historical, monte carlo, rollling years, etc).

Thanks again. I think when we are done chewing on this is will reinforce that there's no single best withdrawal strategy (e.g. folks who can absorb more volatility can get higher average returns by choosing the appropriate withdrawal method).

 

[justin]

I tried playing with a hybrid variable+fixed withdrawal strategy in FIREcalc. Suprisingly good results!

Very safe results using 3% fixed plus 2% variable and 2% fixed plus 4% variable.

5% or 6% SWR with some volatility from year to year seems to fit me better than 4% fixed each year.

I figure I'll have at least 3-4 decades of retirement and maybe 6 or more.

 

[youbet]

Justin,

So how do you take the variable withdrawal rates to your budget? What happens in the low withdrawal years?

 

[obryanjf]

Basic question: What data did you use for the modelling of inflation/asset porformance, etc (actual historical, monte carlo, rollling years, etc).

Thanks again. I think when we are done chewing on this is will reinforce that there's no single best withdrawal strategy (e.g. folks who can absorb more volatility can get higher average returns by choosing the appropriate withdrawal method).

I use Gummys data with 10 asset classes plus inflation starting in 1928. I run with a baseline 1MM portfolio spread 10% Tbills, 20% 5 yr treasuries, 40% S&P, 10% small value, 10% large value and 10% small growth. No real logic there, just what I picked weeks ago. Based on other posts, you likely can add a 10, 20 or 30 basis points to the WD% by diversifing better.

I use 75 rolling periods, not Monte Carlo, when the data runs out it rolls over to the beg of the data(1928) to make a full simulation. All the models were subject to the same back end mechanics.

If I can reasonably make my spreadsheet/macros easier to understand and use, I'll post them so others that are inclined can do their own analysis.

job

 

[Cb]

Justin,

So how do you take the variable withdrawal rates to your budget?  What happens in the low withdrawal years?

Youbet, here are some of my thoughts on the "real-life" budget impact of variable withdrawal rates (lifted from a piece I wrote for my wife a while ago):

"So far I've talked about a consistent, year after year "fixed" rate of withdrawal for our retirement income (the only exception being that each year we'd increase our withdrawal by the rate of inflation for the year past). So even in the event of severe market downturns we'd still withdraw the same amount as the year before, plus a bit more to account for inflation.  Similarly, even after a strong multi-year run-up in our net worth, we'd just take the same amount as the year prior, with only the minor inflation adjustment.

On the other hand, if we were to show a bit of flexibility in our year-to-year withdrawals based on prior-year market performance, adjusting our draw somewhat based on year-end portfolio value, we could either:

- draw a somewhat larger amount right from the outset,

or (my strong preference)

- take the same ~4% initial draw, but with a larger margin of safety.

We probably wouldn’t be happy if we allowed our annual draws to track our year-end portfolio value exactly – even if we started out with a 20% higher withdrawal rate, chopping our spending by 20% from one year to the next, or increasing it by 30% from one year to the next (both of which would likely happen 2 - 6 times in the next 55 years) is more variation than we’d like. The 20% down years would be more uncomfortable than we’d like, and spending 30% more in a given year would likely be somewhat wasteful. I think it’d be preferable to allow our annual draw to float only slightly, back and forth about $3000 or $4000 during those more extreme once-every-ten-year market swings.

Our budget is about 50% fixed annual expenses - those that are not optional and really don’t vary, such as property taxes, groceries, health & other insurances, etc. The other 50% are discretionary expenses that are more variable in nature…walking around money, remodeling, dining out, music stuff, vacations, etc.  So on the surface it looks as though whatever differences we’d see in our annual expenditures would have to come out of the discretionary accounts alone. But there’s actually a bit more to it than that…

The good news is that I believe the $3000 - $4000 dollar swings mentioned above would be relatively painless, because many of our discretionary expense items are “long-cycle”…things such as car and motorcycle purchases, home remodeling/major repairs, and Hawaiian vacations.  We won’t incur those expenses every single year. If we simply made an effort to make those expenditures (whenever possible) following good years, we probably wouldn’t have to shave our entertainment, gift, dining out, or golf budgets much during bad years. A business tries to do the same thing…when profits are squeezed they tend to defer major capital equipment expenditures, but they still pay their taxes, etc."

 

[justin]

So how do you take the variable withdrawal rates to your budget? What happens in the low withdrawal years?

Let's say I retire with $1,000,000 and I plan on taking $40,000/yr fixed (w/ inflation adjustments). That's the default w/ the 4% SWR.

Instead, I might take 3% fixed ($30,000/yr) plus 2% variable (initially $20,000/yr). This would give me either a higher standard of living initially, or allow me to ER a little sooner. I would be taking $50,000/yr initially. If the portfolio takes a massive 50% drop, to $500,000, I'm still drawing $40,000/yr (plus a CPI COLA on 75% of that amount) under this 3%/2% hybrid plan.

For a 2% fixed/4% variable plan, I'd draw $60k initially. W/ a 50% drop in portfolio value, I'd still draw $40,000/yr (plus a CPI COLA on 50% of that amount).

Those are pretty awful scenarios to encounter shortly after retirement. I've got a lot of discretionary spending built into my ER budget for travel and other things (~30% of the budget). I can cut back in lean years and live large(r) in good years. Capital improvements or purchases can be made in good years (new roof, house, car, etc.).

I also have the opportunity for spouse and I to work a few hrs/wk and make enough money to support 30-40% of our nondiscretionary budget items.

Then there's the conservatism built into the budget itself. Some of the expenses can be reduced when I have time to focus on them instead of working.

 

[youbet]

For a 2% fixed/4% variable plan, I'd draw $60k initially.  W/ a 50% drop in portfolio value, I'd still draw $40,000/yr (plus a CPI COLA on 50% of that amount).

I'm really having trouble understanding.  Sorry.

If your portfolio drops to $500K and you withdraw $40K, doesn't that mean your withdrawal percentage of your portfolio for that year is a whopping 8%?     How does taking 8% help overcome the massive loss in portfolio value?

The Retire Early Homepage site has an excellent article on withdrawal rates in hard times.  Take a look at it.

Also, for folks still a number of years away from retirement, you may be underestimating the importance of having funds for discretionary spending.  I'm only a few weeks from beginning non-work life.  My budget includes a descretionary category for travel, entertainment, toys, etc.  I can tell you with candor that as I stand on the brink of retirement, removing this discretionary spending from my budget to accomodate a lower SWR would be a very, very unhappy thing!      In fact, if I thought that was the situation, I would stay at work a little longer. 

Therefore, I chose to go with the more conservative 4% SWR that I feel I can stay with even in hard times.  That meant working longer than I really wanted to, but I felt I needed to accomodate my own desires.  So, consider the fact that as you near retirement age, the thought of giving up descretionary spending and hunkering down under a very tight budget will likely become more distasteful.

 

[justin]

I'm really having trouble understanding. Sorry.

If your portfolio drops to $500K and you withdraw $40K, doesn't that mean your withdrawal percentage of your portfolio for that year is a whopping 8%? How does taking 8% help overcome the massive loss in portfolio value?

The 3% fixed/2% variable withdrawal strategy I'm talking about works like this:

The "fixed" component of withdrawal works just like a regular fixed SWR - I take 3% of the initial principal balance each year, with inflation adjustments. So in "real" terms (ignoring CPI's potential shortcomings), I'm taking that 3% SWR each and every year. The variable component does not affect the 3% fixed component. With my original $1 million nest egg example, I'm taking $30,000 fixed every year, with inflation adjustments.

Then on top of the fixed component, I also withdraw a "variable" component. In this case, I'm talking about a 2% variable withdrawal. Each year, I take 2% of the total portfolio value. If my portfolio falls 50% to $500,000, I only get $10,000 for the variable component. If my portfolio goes up 50% to $1,500,000, I take $30,000 for the variable component.

At a minimum, the fixed 3% SWR gives me $30k/yr w/ inflation adjustments. The variable 2% SWR will usually get me up to $40k+/yr and in most years will put me well above $40k/yr.

I plan on ERing in my 30's, and I hope to have 4+ decades of living. Since the SWR calculations are based on the worst case returns and not on the average returns, I'm expecting to have a rather large portfolio 20-30-40 years out. With the hybrid fixed/variable withdrawal strategy I get to enjoy some of that portfolio growth. I don't want to be 65 years old with a $3,00,000 inflation adjusted portfolio and still be taking $40,000/yr inflation adjusted dollars when I can be taking $90,000/yr inflation adjusted dollars under the hybrid 3%/2% strategy ($30,000 fixed withdrawal plus 2% of $3,000,000 = $60,000 variable withdrawal).

The trade off of having an initial 5% hybrid withdrawal rate (and higher standard of living) with a little uncertainty from year to year is better for me than a 4% fixed withdrawal rate forever.

I have to have a plan to spend my money decades in the future.

If 10-20% of my years are spent with a small degree of scrimping and saving, and the rest are filled with living the high life (or at least 25% higher life), then that is ok with me. In reality, I probably wouldn't spend all of the surplus during the "good years".

By the way, I plan on having more than $40k/yr income in retirement and a nest egg larger than $1,000,000. The example I used was nice and round for illustrative purposes only.

 

[astromeria]

I can tell you with candor that as I stand on the brink of retirement, removing this discretionary spending from my budget to accomodate a lower SWR would be a very, very unhappy thing! In fact, if I thought that was the situation, I would stay at work a little longer.

If I had to give up virtually all discretionary spending, it wouldn't be so bad--hmm, that's why I call it discretionary 8). Instead of 2-3* hotels, it would be camping and visiting friends & relatives. Been there, done that--it was fun. Driving to big bro's cottage near the Hamptons and our friends' cottage on Edisto would hardly be a sacrifice. Doing without cable/satellite/Netflix/Barnes & Noble means borrowing books & DVDs & CDs from other folks (and lending ours)--no problemo, I already do a bit of that. Local beach walks, kayaking, biking, and art gallery crawls are free (so is sex...and what could be more entertaining?!)--hokay. Eating at home every day...I'm cool with that, love to cook. Buying a new-to-me car or computer...not so bad really, done it before. We could drop down to one car easy if DH retires. We live walking distance from grocery shopping and mail order our presriptions. And how about combining that long-desired trip to Thailand with a little dental work/surgery I could get into that. Of course they'll have to pry the Internet from my cold, dead hands...but I can use it for free at the library or my brother's home or office.

Maybe it helps that I was working class in my 20s (with kids yet), so I know how to handle it. Also my grandparents lived in a tiny public-housing apartment, and they were happy. Knowledge is power!