PP withdrawals in retirement

[doodle]

My parents are in a PP and they are going to be retiring in a few months. They need to withdraw about 3% annually from the PP to cover living expenses. How is the best way to go about doing this? Should they simply live off the dividend income and sell some of the highest asset as a way of keeping things in balance?

[doodle]

Here is an idea that I had as well that is non-PP. More related to Joe Dominguez strategy. If my parents only need a 3% drawdown to live, what is to keep them from just piling all in to long bonds as an annuity type strategy while holding a bit of gold say 10% in case of inflation? Under this strategy they would never have to touch the principle and could exist merely off of interest. Barring hyperinflation (thats what the gold is for) they should be able to make it for 30 years on that...

[Lone Wolf]

My parents are in a PP and they are going to be retiring in a few months. They need to withdraw about 3% annually from the PP to cover living expenses. How is the best way to go about doing this? Should they simply live off the dividend income and sell some of the highest asset as a way of keeping things in balance?

How about just spending from the "Cash" portion of the portfolio and then rebalancing when they crack the lower 15% band?  (Or annually, if this is what they prefer.)

Here is an idea that I had as well that is non-PP. More related to Joe Dominguez strategy. If my parents only need a 3% drawdown to live, what is to keep them from just piling all in to long bonds as an annuity type strategy while holding a bit of gold say 10% in case of inflation? Under this strategy they would never have to touch the principle and could exist merely off of interest. Barring hyperinflation (thats what the gold is for) they should be able to make it for 30 years on that...

I am not a big fan of this "Your Money or Your Life" approach.  There's a good chance that it would turn out okay but it strikes me as very risky.

Imagine a period of "annoyingly high" inflation.  Let's say we return to a decade of 70s-style double-digit inflation and high interest rates.  With long bonds only paying 3%, a 10% inflation guarantees a 7% real loss even in the presence of no spending.  If they are spending at 3%, that's a real portfolio shrinkage of 10% (setting aside the gold.)  A few years like that could spell trouble, especially since as inflation takes hold, more than 3% will be necessary thanks to rising prices.

The gold could very well help but I'd be concerned that the effect might not be enough to counteract the damage the other 90% of the portfolio is suffering.  If interest rates keep pace with inflation, gold may not take off.  Furthermore, this will damage the value of all these LT bonds, closing off the option to sell them without taking terrible losses.

[Tyler]

Along these lines, I've been wondering:

What's a safe withdrawal rate for the PP to maintain principal in retirement?  Does the traditional 4% (usually cited for more stock-heavy portfolios) still apply?  Does the lower volatility mean you can withdraw more?  Or does the more modest growth pace in the good years mean you should withdraw less?

[craigr]

So here's a question I've been wondering about and couldn't quickly find an answer by searching the forums:

What's a safe withdrawal rate for the PP to maintain principal in retirement?  Does the traditional 4% still apply?  Does the lower volatility mean you can withdraw more?  Or does the tracking error against the stock market great years mean that maybe you should withdraw less?

That 4% number is based on a lot of things, and the past many years has gone out of favor with those using stock/bond allocations because the stock market has done so poorly relatively speaking.

But I've found the Permanent Portfolio has been able to generate between 3-5% real returns over rolling 10 year periods consistently (so far). So it is not unreasonable to think a 3-4% withdrawal could be used and not affect principal. But there are no promises. You just need to monitor it. Going above 4% you will face a good chance of eating into the principal. Which may or may not matter depending on your own circumstances and other savings you have (e.g. pension, etc.). A 3% withdrawal is more conservative and you can always spend more later if you find you have a lot of headroom left.

Personally when I'm between projects I'm withdrawing from the cash and will replenish it with the winner's profits if needed. However you can also have all dividends/interest go into cash so it is building itself back up as well which delays having to sell off your winners.

Now if we happen to hit a particularly bad year and the portfolio takes a real loss, then it makes sense to cut back your spending and just ride things out. Just as you'd do with any portfolio.

[Alanw]

So here's a question I've been wondering about and couldn't quickly find an answer by searching the forums:

What's a safe withdrawal rate for the PP to maintain principal in retirement?  Does the traditional 4% still apply?  Does the lower volatility mean you can withdraw more?  Or does the tracking error against the stock market great years mean that maybe you should withdraw less?

That 4% number is based on a lot of things, and the past many years has gone out of favor with those using stock/bond allocations because the stock market has done so poorly relatively speaking.

But I've found the Permanent Portfolio has been able to generate between 3-5% real returns over rolling 10 year periods consistently (so far). So it is not unreasonable to think a 3-4% withdrawal could be used and not affect principal. But there are no promises. You just need to monitor it. Going above 4% you will face a good chance of eating into the principal. Which may or may not matter depending on your own circumstances and other savings you have (e.g. pension, etc.). A 3% withdrawal is more conservative and you can always spend more later if you find you have a lot of headroom left.

Personally when I'm between projects I'm withdrawing from the cash and will replenish it with the winner's profits if needed. However you can also have all dividends/interest go into cash so it is building itself back up as well which delays having to sell off your winners.

Now if we happen to hit a particularly bad year and the portfolio takes a real loss, then it makes sense to cut back your spending and just ride things out. Just as you'd do with any portfolio.

Withdrawing a certain percentage from your portfolio is a question for each individual based on your individual circumstances and preferences.  For example, what is your current age, your life expectancy (if only we knew), your lifestyle, how much you wish to leave your heirs, do you have long term care insurance, could you go back to work if necessary, etc.

Personally, I am comfortable drawing 5% from the cash portion of the portfolio knowing that I will most likely be eating into the principal.  I will rebalance as necessary.  My reasoning is that I will be able to reduce that percentage later as I age and will no longer be able to do a lot of the things that I enjoy doing now and are quite costly.  These things will no longer be physically possible and the expenses associated with them will be reduced or eliminated.  As much as we hate to admit it,  this will eventually occur.

This is how I view the reality of living in retirement.  Potentially sacrificing a small part of the future for current lifestyle.  Being a cancer survivor has also helped in the decision of more personal consumption now rather than later.  Hopefully I'll be around for a while to see how it all works out.

[Reub]

Some of you may be familiar with it already but firecalc.com is a good place to get an idea of a safe withdrawal rate.

Clive, while you're here, do you think that this might be a "walk away in May" kind of a year for equities again? Or possibly even a "walk away in April"?

[pershing83]

HB gets his due!

More interest in PRPFX, PERM, and Harry Brown with world allocation ETF's thrown in. I thought this link might be interesting to the board. Results of various combinations with plain old HB's pp looking good.

http://www.myplaniq.com/articles/201203 ... lobal-etf/

[6 Iron]

Some of you may be familiar with it already but firecalc.com is a good place to get an idea of a safe withdrawal rate.

Reub, and others, what do you use on firecalc for return, standard deviation, and inflation rate? Do you count social security? For me, I use 8%, 8%, 3.5% and no soc. sec. income. Wonder of I am being conservative enough. I could sure use a crystal ball.

[Reub]

I just plug in how much I want to spend per year for my retirement needs, how much my portfolio is worth, and a life expectancy of 45 years. At a withdrawal rate of 3% firecalc tells me that there are zero instances where I would have run out of money over the past 96 years of data. At a 4% withdrawal rate I would have run out of money in 15% of the possible cycles.

[MediumTex]

HB gets his due!

More interest in PRPFX, PERM, and Harry Brown with world allocation ETF's thrown in. I thought this link might be interesting to the board. Results of various combinations with plain old HB's pp looking good.

http://www.myplaniq.com/articles/201203 ... lobal-etf/

pershing,

Thanks for posting that link.  There is some good information there.

[Xan]

Firecalc looks like a great tool, but I'm not convinced that its randomizer is working properly.  When I use the "random performance" feature, and click Submit multiple times with the same parameters, I get wildly different results.  One time I have 100% success, and another it might be 75%, with all the lines shifted much lower.  Either their randomizer is broken, or they're not doing enough runs for each graph they generate.

[Tyler]

Thanks Clive.  Interesting.

For most calculations you quote, you often reference the worst PP years as being in the 1930s to 50s.  However, is that truly a fair comparison, since the decoupling of gold and the dollar is a key factor in why the PP works the way it does?  It seems to me that simply using silver as a proxy for gold prior to 1971 doesn't fundamentally model how those assets operate in a post-gold-standard economy.

Maybe I'm wrong, and perhaps I'm misunderstanding your chart, but I'm under the impression that the PP real returns since 1971 are closer to 4% than 2.  

[Tyler]

Thanks for the info, Clive.  I understand that there is no perfect plan for any investment when you look at history.  I imagine that during WW2 people had more to worry about than just safe withdrawal rates.

What does the chart look like from 1971 - present? 

[craigr]

Craig's Vanguard Questionnaire Says 100% Stock is Great Idea and Unicorns are Real blog/article suggests that 100% all stock is not an appropriate asset allocation. For some however the choices may be limited. Looking at the prospects of being able to withdraw a 4% income for 30 years for instance in real terms (after inflation) indicates that all-stocks have a reasonably high chance of success

If you hit a pocket of dead air on stocks you can be in just as much trouble as any other asset. Substantially since the early 2000s I have used my portfolio for living expenses as I transitioned between careers. If I had been drawing down a stock heavy portfolio there is almost no chance it would have supported 4% without eating into the principal significantly. Stocks the past 10 years have been very volatile and unreliable. However the Permanent Portfolio has always given me at least *one* asset that could be tapped to provide either income or capital appreciation to use.

Generally the greater stability provided by the PP comes at the expense of greater risk of drawdown via income withdrawals if your income needs are any higher than a 2% withdrawal rate.

Yet all the data for the last 40 years clearly shows that 3-5% real returns/withdrawal rates would have been much more reliably achieved with the Permanent Portfolio than with stocks alone.

The problem with assuming 4% stock withdrawals is that some markets like the US have had very big bull markets right in the middle of the primary dataset (1980-2000). Without that huge hump of returns then the outlook for stocks is a lot less certain. I feel much more confident owning stocks, bonds, cash and gold than something that is concentrated in stocks alone. Stocks have a mystique that I think is not totally deserved. They are not a sure thing investment and historically they've had really bad protracted periods of returns that could really affect a retiree relying on them for portfolio growth alone. IMO.

I think one of the biggest attributes of the Permanent Portfolio is it gives a retiree options to choose from for withdrawals that a concentrated portfolio does not. Sometimes you need options to respond to markets that don't work out as planned.

[Tyler]

So over the past 40 years it sounds like Craig and Clive are pretty much on the same page with the PP SWR.  Craig says 3-4% (leaning towards 3), and Clive offers 3-3.6% (1-1.6% over a 4% drawdown, minus 2% tailwind).

Of course no plan is bulletproof, but that seems like sound, reasonably conservative advice.  Works for me - Thanks for the input.  

[moda0306]

Clive,

For the bajillionth time, not only is your 1936-1952 PP NOT the PP, the negative real rates after the war would have likely made gold skyrocket.. also, it probably wouldn't have fallen as much in 1937... take a look at 2008), we weren't a soveriegn fiat currency then, so even though gov't bonds were likely the safest bonds available, they weren't free of default risk. 

You can't completely alter the nature of our currency, remove 25% of it and replace it with a different "similar" asset, and expect to not at least have to put an asterisk next to your statement.

The change in risk profile of our currency and the nature of silver vs gold make your pre-1970 PP calculations start to approach irrelevence, IMO.

[craigr]

Of course there are good and bad periods for all assets/blends. The fact that the PP has performed well over the last three or four decades is no assurance that will continue to hold in the future.

The point really is by owning widely varying assets with little sharing of risk I have a better chance of adapting to the future than gambling heavily on one asset if that bet turns out wrong. The US markets are the outlier in terms of world history. Betting on stocks doing as well as they have in the past is at least as risky as widely diversifying, perhaps more so. I like having more options and feel safer spreading my money around outside of the stock market exclusively (or even heavily).

[Gumby]

not only is your 1936-1952 PP NOT the PP, the negative real rates after the war would have likely made gold skyrocket.. also, it probably wouldn't have fallen as much in 1937... take a look at 2008), we weren't a soveriegn fiat currency then, so even though gov't bonds were likely the safest bonds available, they weren't free of default risk.  

You can't completely alter the nature of our currency, remove 25% of it and replace it with a different "similar" asset, and expect to not at least have to put an asterisk next to your statement.

The change in risk profile of our currency and the nature of silver vs gold make your pre-1970 PP calculations start to approach irrelevence, IMO.

That example was a near pure PP moda - with 25% gold, not silver.

Well, we've been over this countless times. What on Earth would have been the point of holding physical gold in a portfolio when the currency was already pegged to gold?? That would be like having a portfolio with 50% cash — of course it would underperform. So, this long backtest through a gold-standard era doesn't make any sense. There's no logic in it. Why bring out that data when it's so inherently flawed?

The Permanent Portfolio was created for a fiat world. It makes no sense in a gold-standard world.

[D1984]

For the bajillionth time, not only is your 1936-1952 PP NOT the PP, the negative real rates after the war would have likely made gold skyrocket.. also, it probably wouldn't have fallen as much in 1937... take a look at 2008), we weren't a soveriegn fiat currency then, so even though gov't bonds were likely the safest bonds available, they weren't free of default risk. 

You can't completely alter the nature of our currency, remove 25% of it and replace it with a different "similar" asset, and expect to not at least have to put an asterisk next to your statement.

The change in risk profile of our currency and the nature of silver vs gold make your pre-1970 PP calculations start to approach irrelevence, IMO.

Moda, I have tested this with silver, with gold (fixed at $35 an ounce so it didn't lose anything or gain anything in 1937 or in the postwar years), with commodities (DJ commodities index which I have data for back to 1933), and with platinum.. No matter what "inflation protection" asset you choose, the results in real inflation adjusted terms for the entire PP are basically as bad as Clive described.

Yes, in a fiat currency system LTT bonds would have behaved differently and they probably would have started out with MUCH higher (think Volcker-style 20%+) rates from their peak (in 1920) and thus would have had more "punch" during the Depression (see my post from a few months back about this and about how in 1931 they would have almost certainly acted like they would've in 2008) as yields fell...HOWEVER, the 1937-38 recession was partly due (the rest was due to FDR's austerity and attempts to balance the budget in a still depressed economy) to the Fed tightening the money supply (like in 1920 and 1981) and this quite possibly would have hurt long bonds (and maybe even hurt gold).

For the sake of argument, though, I'll concede you would be correct...let's assume that we did have a fiat currency, that gold was free market priced, and that it DID stay fairly stable in 1937-38 and then went up during 1941-42 (wartime inflation before price controls and rationing took full effect) and then again during 1946-47 when the Fed kept rates low while inflation soared to double digits. Gold would have helped save the PP then, but what would happen from circa 1950 to the late 60s? We all know that gold does very poorly in an environment of prosperity with positive real rates...and that is what we had for almost all of the postwar boom from the early 1950s at least until 1970. Gold would fall (as would LTTs as rates gradually rose from 2% levels) and you would have traded one period of PPPPPP (Pathetically Piss Poor Permanent Portfolio Performance) in 1937-53 for another one from 1950 to circa 1970. Hardly a choice I'd like to make.

In all honesty, the best way to "withdraw" from the PP from 1937-1953 might have been not to withdraw at all but to borrow; assume borrowing rates were at STT yields plus 1.25% (which Interactive Brokers basically offers today so it's not too far-fetched to assume that if such a broker was available then such a feature might have been offered); you could have borrowed 4% each year (adjusted for inflation) and let the debt accrue at margin and you'd still be about 35-40% better off in real terms than you would have been if you'd withdrawn an inflation-adjusted 4% each year. Why? Because while REAL returns of the PP were not able to beat inflation during this time when depleted by withdrawals (i.e. you had negative real rates working against you) the nominal returns almost DID beat inflation and when you borrowed in a suppressed rate environment you had financial repression working for you since the artificially lowered rates were even less than the PP's nominal return and were certainly much less than inflation for most of this period. Better to swim with the current than trying to swim upstream against it IMO.

[moda0306]

Where's the thread with the year-by-year returns of the 1930's thru 1970's again?

I need to take another look at this.

[craigr]

Guys this historical data is all well and good, but the time periods discussed also includes big-time cases of price and production controls not just during the 1930s but also WWII. Plus there are other impacts as well. Since gold was the dollar it is simply impossible to know how the portfolio would have performed with backtested data to those dates.

All you can do in that case is to look at economic cycles and try to apply the theory of how each asset would likely behave unencumbered. Plus you'd also have to look at other countries to see how assets behaved in isolation under various economic climates and glean some useful data. But looking at a raw spreadsheet just isn't going to do it. IMO.

[HB Reader]

Well, we've been over this countless times. What on Earth would have been the point of holding physical gold in a portfolio when the currency was already pegged to gold?? That would be like having a portfolio with 50% cash — of course it would underperform. So, this long backtest through a gold-standard era doesn't make any sense. There's no logic in it. Why bring out that data when it's so inherently flawed?

The Permanent Portfolio was created for a fiat world. It makes no sense in a gold-standard world.

That's right. The PP was created for the world monetary and economic system that has existed since 1971.  Will it change in my lifetime or yours?  Maybe, but until it does I don't plan to change my general investment framework.  I don't feel any need to make it conform specifically with a world that existed before then. 

Many of us seem to have different views about how our current system works, but other than occasional calls for a return to a gold standard (which can have many different variants) I've heard precious little serious discussion anywhere about fundamental monetary and economic change.  Of course, it may well be that prodding or beating the dead, comatose or irrelevant horse in front of us is a lot easier intellectually than coming up with a new way to view the world.

   

[Gumby]

Guys this historical data is all well and good, but the time periods discussed also includes big-time cases of price and production controls not just during the 1930s but also WWII. Plus there are other impacts as well. Since gold was the dollar it is simply impossible to know how the portfolio would have performed with backtested data to those dates.

All you can do in that case is to look at economic cycles and try to apply the theory of how each asset would likely behave unencumbered. Plus you'd also have to look at other countries to see how assets behaved in isolation under various economic climates and glean some useful data. But looking at a raw spreadsheet just isn't going to do it. IMO.

Exactly. This idea that we could somehow determine what the assets might have done if the pre-1971 dollar wasn't pegged to gold is totally bonkers, since all of the dollar-denominated assets and investments from the pre-1971 era played out exactly as they did because the dollar was pegged to gold.

To back-test the Permanent Portfolio strategy (or any portfolio, really) pre-1970 is to ignore Chaos Theory, which tells us that the slightest change in the initial conditions (and the conditions at any given point in time) cause enormous and widely diverging outcomes over time.

Chaos theory is a field of study in mathematics, with applications in several disciplines including physics, engineering, economics, biology, and philosophy. Chaos theory studies the behavior of dynamical systems that are highly sensitive to initial conditions, an effect which is popularly referred to as the butterfly effect. Small differences in initial conditions (such as those due to rounding errors in numerical computation) yield widely diverging outcomes for chaotic systems, rendering long-term prediction impossible in general. This happens even though these systems are deterministic, meaning that their future behavior is fully determined by their initial conditions, with no random elements involved. In other words, the deterministic nature of these systems does not make them predictable. This behavior is known as deterministic chaos, or simply chaos.
Source: http://en.wikipedia.org/wiki/Chaos_theory

The stock market, the bond market, the commodities markets, the decisions of politicians, the decisions of central bankers, the decisions of consumers... everything would have turned out differently if we had a fiat currency pre-1971. So, it's totally flawed to draw conclusions from pre-1971 backtesting.

[craigr]

I understand the desire to analyze these things and I don't think it's good to use the word "bonkers" to describe it. I'm just not sure backtesting is going to work here. It is more productive for me to look at economic cycles in various places and see how various assets performed when put under stress and good conditions respectively. It's not perfect, but you can come away with some interesting observations on portfolio construction even if the data is incomplete.