question about withdrawal rates (50+, years, support, single)

[borninsac]

Just about all that I've read about withdrawal rates focuses on market risk and its impact in decreasing investment value causing one to run out of money before they die. The assumption seems to be that one's cost of living remains fixed and increasing due to an inflation assumption.

For those who have been self-employed, you know as I do how business income fluctuates from year-to-year and how that fluctuation impacts money going "home". This impacts personal spending. I personally have never had fluctuating business income cause any serious depletion of savings because of adjustments made to discretionary spending. A similar result can happen in retirement by decreasing discretionary spending to offset a bad investment income year due to market forces. I think in most cases there is enough wiggle room in overall spending to cancel any market downturns and ride the tide until the next market uptick.

Now if you don't have any discretionary budget amounts, then you do have a serious problem with market risk and its impact on withdrawal rates.

[hikernut]

Originally Posted by FREE866

What gives me hope is that initially I might be withdrawing 5% , but then once I start receiving social security my withdrawals will drop to 1-2%

You may well be fine, or even way more than fine. I would take a look at things, year by year, to see if you are okay.

I've built spreadsheets that model the income sources over time. There are two or three different retirement funding phases in my spreadsheets.

During phase 1 of retirement we are living entirely off financial assets. There is one long-term account invested in stocks/bonds with the usual assumed SWR of 4%, or bit less if retiring earlier. Then there is a second account that is put into something safe like CDs or Treasury notes. This account is assumed to be exhausted during phase 1.

During phase 2 of retirement we are living off Social Security plus the SWR from the long-term account.

Then there is an optional phase 3, where part of the long-term account is annuitized. If one or both of us lives long enough, is still relatively healthy, and the long-term account has not fared well in terms of investment returns, we may choose to implement this phase 3. On the other hand, if investment returns are good there will be no need to annuitize.

[uhuru]

My advice and I retired in my late forties is not to use any formula other than as a general guideline.

Quite frankly, lifestyles differ and what one person may deem adequate may be very different than what another person would.

Today, we withdraw just the IRS minimum required distribution (MRD) and we spend at the most about 40% of that amount because between social security and other income we just don't need any more of the MRD.

The only way to really know what your expenditure level is now or will be when you retire is to prepare a realistic budget and compare it to what resources you have such as SS and pensions (if any). I emphasis the word "realistic" because it is easy to come up with a budget and then exceed it. This holds true whether it is a budget pre-retirement or post retirement.

I don't dismiss the formula approach that is often advocated but ultimately it needs to be individualized to each person's specific circumstances. The only areas where we spend more than we used to are on medical expenses and on travel. The latter is discretionary since we can cut back if we need to do so. The former is a function of aging and may well increase as one gets older. Fortunately, we do have excellent health insurance but those deductibles and co-pays do add up!

[lieqiang]

Originally Posted by rjm1cc

If you are looking at 40 years and think the market is price on the high end today you might be in the 2% to 3% withdrawal range (remember worst market).

I don't understand a recommendation of a 2% withdrawal rate, regardless of current market conditions. Obviously past performance is no guarantee of future results, but it is my understanding that historically a 3% withdrawal rate has never run out of money in a 40 year retirement. That includes starting at times of high valuations, encountering The Great Depression early on, hitting the late 60s extended flatline, etc.

I don't recall the exact number for 40 year 95% success rate but I think it was around 3.6%, where minor rules based flexible withdrawal strategies (like not getting an annual raise when portfolio return for previous year was negative) bring that up to 100% as well.

I get erring on the side of caution, but that is the whole point of the 95%+ historical success rates. They have that built in.

[mathjak107]

you are correct , 3% draws with at least 25% equity has never run out of money over 40 years .


here is a chart based on sbbi data going out to 40 years . sbbi is ibbotson / morningstar .

depending on who's data set and what the allocations are results vary a bit . i wouldn't get to concerned about things like the differences between 3.6 or 3.5 or 3.70 since your exact data set will be different .

in fact if you enter expenses in to firecalc , firecalc treats certain expenses like healthcare and long term care by its default differently than fidelity does .

the firecalc default is to bump them by the regular rate of inflation , fidelity bumps them by 5.50% which is now down from the previous 7.50% they were using on healthcare .

fidelity uses monte carlo simulations to find even worse combo's than historical , firecalc uses historical . firecalc's claim to fame is it uses actual historical , although it does have a monte carlo mode . so trying to compare 90% success rates between the two is not going to be apples to apples , but they will be close enough within a 1/2% or so .


the original trinity study got slightly worse results than bill bengan because the trinity used longer term corporate bonds and bill bengan used 5 year gov't bonds .

[mathjak107]

as far as what is an acceptable rate ? that is something you have to decide .

i know if i was building a house today here in nyc , i want that house to survive the likes of sandy .

i don't care if we had a thousand storms to date and none were comparable in intensity , i just lived through sandy . . all i care about is i want it to survive the worst to date with some extra slack built in . .

i don't want to rule out another storm like sandy , i want to allow for it and plan for it .

so that being the case i want to survive all worst case time frames to date at a 100% . i rather plan lower and take raises all long if things are better than planned for .'

taking pay cuts later on can be a lot rougher once a lifestyle is built around more money .

how much discretionary income you have should be a big factor in your decision . if you have quite a bit than go for a lower success rate , you can always cut back . but if everything is a need and you have little to cut back , you may want to plan around a lower success rate .

a very important factor is life expectancy .

as i mentioned a single male only has a 47% chance of seeing 85 but a couple has a 74% chance .

those figures can bolster success rates since if you planned to 90 odds are statistically you will not live that long as a single male . a 90% success rates becomes close to 100% in that case .

but life is filled with all kinds of emergency spending and big expenses that can consume a major part of a yearly budget so i prefer to have all the cushioning i can and just take raises if things go better than planned for .

[rjm1cc]

Originally Posted by lieqiang

I don't understand a recommendation of a 2% withdrawal rate, regardless of current market conditions. Obviously past performance is no guarantee of future results, but it is my understanding that historically a 3% withdrawal rate has never run out of money in a 40 year retirement. That includes starting at times of high valuations, encountering The Great Depression early on, hitting the late 60s extended flatline, etc.

I don't recall the exact number for 40 year 95% success rate but I think it was around 3.6%, where minor rules based flexible withdrawal strategies (like not getting an annual raise when portfolio return for previous year was negative) bring that up to 100% as well.

I get erring on the side of caution, but that is the whole point of the 95%+ historical success rates. They have that built in.

You are correct as I was just rounding for a range. But remember the history used and the design of the calculators do not include the low interest rate environment we have, the possibility of just printing money to pay off debt and investments declining in value as wages and taxes go up.

Look at what you could have safely earned on a million dollars in 2000 and what you can earn now. Try a 30 year treasury bond or something similar. You will be getting under 2% but of course you can spend down the principal.

Life expectancy is also a problem. Not an expert but I think the tables lag a couple of years behind actual results. It takes a while to get new tables approved for the insurance industry. I think that about every 10 to 20 years life expectancy increase for the people being born. Thus the table may not really take into consideration that 40 year old may have another year or two of life compared to the table. Also personnel factors such as health habits and the work you do can change life expectancy. In short 40 years might be 50 or even more.

[mathjak107]

we have been adding one year of life now every 4 years since 2000 . more folks are living to even ages like 90 than ever before . for a couple today the odds of a 65 year old couple having one of them see 90 is about the odds of a coin toss . that is pretty high .

but there is still a lot of padding in those calculators since they are based on pretty poor conditions already especially at 100% success rates . .

life expectancy influence success rates and makes them higher as well as in practice humans spend dynamically spending less when markets are down .as we age we tend not to need inflation adjusting yearly as well . so all these things tend to improve on the success rate .

if you have a high success rate all these things tend to make it even better with a bigger cushion .

i have yet to see anyone except pfau and kitces combine life expectancy statistics in with success rates .

the fact that almost 1/2 of us won't see 90 really improves the chances of success if planning out to 90 or 95 in the calculator ..

[oldsoldier1976]

Originally Posted by mathjak107

we have been adding one year of life now every 4 years since 2000 . more folks are living to even ages like 90 than ever before . for a couple today the odds of a 65 year old couple having one of them see 90 is about the odds of a coin toss . that is pretty high .

but there is still a lot of padding in those calculators since they are based on pretty poor conditions already especially at 100% success rates . .

life expectancy influence success rates and makes them higher as well as in practice humans spend dynamically spending less when markets are down .as we age we tend not to need inflation adjusting yearly as well . so all these things tend to improve on the success rate .

if you have a high success rate all these things tend to make it even better with a bigger cushion .

i have yet to see anyone except pfau and kitces combine life expectancy statistics in with success rates .

the fact that almost 1/2 of us won't see 90 really improves the chances of success if planning out to 90 or 95 in the calculator ..

First let me say welcome back MJ. We missed you and your insights.

Now as for me since I am an early retiree and in fact I am not even 59 yet I am withdrawing at a 3% rate at the moment. Since starting my savings has gone up not down. It is good to know that we are on the right track. I will reassess in December but since DW finally decided to buy new SRX and told me that I was paying her car payments and she would pay off my car loan I can keep it where it is and not worry or I can take a tad bit more to save for my Z4. Hmmmmmm

[mathjak107]

thanks .