cancel
Showing results for 
Search instead for 
Did you mean: 
     
Highlighted
Participant ○○

retirement distribution portfolios

There are valuable posts on the old Discuss site of enduring value, sadly gone to waste because the failed search function there (and maybe that site won't even be accessible someday too soon).  Here is a post that I especially keep returning to as a guide in keeping a "retired" investing focus, the referenced book contains much data to support the noted conclusions:

 
Re: Ideas For an "All Bond Funds" Portfolio
06-30-2017, 8:24 AM | Post #3851247
 

For a great understanding of Sequence of Return Risk, I HIGHLY suggest reading "Unveiling the Retirement Myth - Advanced Retirement Planning based on Market History" by Jim C. Otar. It's 525 pages of highly detailed presentation and currently available online (PDF) for $6 from his website http://www.retirementoptimizer.com/.

Note. I have no association with Mr. Otar and provide this link only as a courtesy. I bought the PDF book a while ago and certainly got a lot out of it. The following was gently “lifted” from another review:

Otar is an engineer turned financial advisor. His book is filled with a great amount of data analysis and detail based on his research using historical market returns. If you are near or in retirement, you owe it to yourself to check it out.

Otar points out that retirement distribution portfolios have critical differences from accumulation portfolios; e.g., they have a finite lifespan of 30 years or less and they are a "wasting asset" instead of a "growth asset" as money is being withdrawn. These differences result in many of the tools and concepts used in financial planning for accumulation portfolios being essentially useless for retirement distribution portfolios.

For example, here are some of his conclusions:

1) Asset allocation and diversification, particularly within asset classes (e.g. stocks) make little difference for portfolio longevity.

2) Portfolio rebalancing, no matter how it is implemented, has little effect on portfolio longevity.

3) Conventional Efficient Frontier analysis does NOT incorporate cash flow and is useless for distribution portfolios.

4) Conventional Monte Carlo analyzers do NOT incorporate negative fat tails of market returns and produce results for distribution portfolios that are far too optimistic.

5) You should NOT select the optimal asset mix (the one that produces the longest portfolio life); instead you should select the tolerable asset mix (the one that produces the maximum loss over a given timeframe that you can tolerate without panicking).

6) Portfolio longevity and appropriateness are overwhelmingly determined by (1) Luck and (2) Withdrawal Rate.

7) If your withdrawal rate is below the SWR (generally 4% real) there is not a meaningful impact on portfolio survivability from factors other than luck.

8) Luck is comprised primarily of two elements: (a) sequence of returns, and (b) inflation. These two factors overwhelmingly determine the likelihood of survivability of a distribution portfolio.

> Luck: If you are unlucky enough to experience significant losses in your portfolio, particularly in the first few years of retirement, nothing you can do will restore your portfolio to its previous level during your lifetime. Unless you permanently reduce the level of your distributions your portfolio is very likely to run out too soon.

> Inflation: Significant inflation during your retirement will destroy the purchasing power of your income withdrawals. Stocks and nominal bonds provide poor protection from inflation; therefore you should incorporate meaningful allocations to real assets and TIPS.

--- 30 ---

0 Kudos
10 Replies
Highlighted
Frequent Contributor

Re: retirement distribution portfolios

       Having the experience of managing my parents assets from 1982-2017, I agree with that 100%. Working with their assets, $250k, a $600 pension and SS, I learned to invest for sure income and some growth as it occurs in spurts and take profits.

       Being from the Depression era we used muni bonds, utilities and Dividend Aristocrats although they didn't use that term yet for them. We use CEF's, Growth funds (2) and some growth stocks like AMZN and MSFT. We have sure monthly income to excess and backup appreciation of our portfolio.

        Still save after modest spending, a lesson handed down from the past.

0 Kudos
Highlighted
Frequent Contributor

Re: retirement distribution portfolios [Old Discuss]

@fpajerski . May be soon there may be time for posters to start moving posts from old forum to new as you have done, but I hope that it won't be necessary. Some of us have our own archives of selected old Discuss posts.

All Post view in old Discuss shows 5,263 pages, or 5,263 x 30 = 157,890 posts. There are probably more as the Off-Topic forum was taken off-line recently. While many of those posts had just heat, some had some light. I hope that M* makes a decision soon on what happens to old Discuss - either moving the content to new Discussions, or making old Discuss a searchable archive linked from new Discussions. But it hasn't even been locked to new posts [that I assume can be done easily], so there may be something going on at M* that only @RyanM and @Darrin know. 

YBB
0 Kudos
Highlighted
Contributor ○○

Re: retirement distribution portfolios


 

For example, here are some of his conclusions:

1) Asset allocation and diversification, particularly within asset classes (e.g. stocks) make little difference for portfolio longevity.

 

4) Conventional Monte Carlo analyzers do NOT incorporate negative fat tails of market returns and produce results for distribution portfolios that are far too optimistic.

 

7) If your withdrawal rate is below the SWR (generally 4% real) there is not a meaningful impact on portfolio survivability from factors other than luck.

8) Luck is comprised primarily of two elements: (a) sequence of returns, and (b) inflation. These two factors overwhelmingly determine the likelihood of survivability of a distribution portfolio.

> Luck: If you are unlucky enough to experience significant losses in your portfolio, particularly in the first few years of retirement, nothing you can do will restore your portfolio to its previous level during your lifetime. Unless you permanently reduce the level of your distributions your portfolio is very likely to run out too soon.

> Inflation: Significant inflation during your retirement will destroy the purchasing power of your income withdrawals. Stocks and nominal bonds provide poor protection from inflation; therefore you should incorporate meaningful allocations to real assets and TIPS.

--- 30 ---


I mostly agree with what is suggested but I'd observe that asset allocation is going to matter.  There is a huge difference in the projected risk of a stock portfolio vs a bond portfolio.  I don't think one can glibly say 'it doesn't matter'.  

Monte Carlo simulation is not perfect.  If one is having a concern with it, some options are to incorporate a more complex probability distribution for the returns - instead of assuming a Gaussian maybe consider some Weibull elements.  Or one can do what practitioners typically do - just boost the volatility estimates.  

Inflation is a problem for all investors, but especially for those who no longer have their 'labor capital'.  Labor prices are well correlated to inflation, so its a natural hedge, but retirees are not 'long' this shadow asset anymore.  If anything they are 'short' that, and thus short inflation, as they buy personal service - home care, health care, and so on.  This accentuates the value of inflation protected payment streams - like Social Security or COLA linked pensions (admittedly rare now).  

Great post to share and I think the essential takeaways are 

a) stay diversified

b) respect the value of lower risk portfolios once drawdowns start

c) unfortunately, spending reductions may be the only solution to market riots, especially in the early years

0 Kudos
Highlighted
Community Manager Community Manager
Community Manager

Re: retirement distribution portfolios [Old Discuss]


@yogibearbull wrote:

@fpajerski . May be soon there may be time for posters to start moving posts from old forum to new as you have done, but I hope that it won't be necessary. Some of us have our own archives of selected old Discuss posts.

All Post view in old Discuss shows 5,263 pages, or 5,263 x 30 = 157,890 posts. There are probably more as the Off-Topic forum was taken off-line recently. While many of those posts had just heat, some had some light. I hope that M* makes a decision soon on what happens to old Discuss - either moving the content to new Discussions, or making old Discuss a searchable archive linked from new Discussions. But it hasn't even been locked to new posts [that I assume can be done easily], so there may be something going on at M* that only @RyanM and @Darrin know. 


I know Darrin wanted to keep the old forums open to new posts for a time to give people ample time to switch over. It's not my decision what happens to the legacy Discuss posts in terms of migrating/archiving/locking, but I expect there will be a decision on all this in the next few weeks.

0 Kudos
Highlighted
Frequent Contributor

Re: retirement distribution portfolios [Old Discuss]

@RyanM . OK. 

But I think that if posters start to move/copy posts selectively from old Discuss, or from their personal archives, that will be less than ideal. Those moved/copied posts will have personal touches, and not be comprehensive. Hopefully, a decision by M* will be made soon.

YBB
0 Kudos
Highlighted
Frequent Contributor

Re: retirement distribution portfolios [Old Discuss]


@yogibearbull wrote:

@RyanM . OK. 

But I think that if posters start to move/copy posts selectively from old Discuss, or from their personal archives, that will be less than ideal. Those moved/copied posts will have personal touches, and not be comprehensive. Hopefully, a decision by M* will be made soon.


Yes, I would surely like to also have access to the old forum posts.

But I am glad I kept copies in a library of mine, of key posts made by me and others!!!

R48

 

0 Kudos
Highlighted
Frequent Contributor

Re: retirement distribution portfolios

In the spirit of not reinventing the wheel, the most important conclusion is number 3:

3) Conventional Efficient Frontier analysis does NOT incorporate cash flow and is useless for distribution portfolios.

Remembering, of course, that the whole foundation of modern retirement portfolio withdrawal strategies (asset allocation, rebalancing, safe withdrawal rates, hence Monte Carlo and, to a much lesser extent, Trinity) is based upon Efficient Frontier analysis.

The alternate, of course, is a withdrawal strategy based upon cash flow, specifically, withdraw and spend less than the cash flow of your portfolio, regardless of whether you are retired or not.

In terms of M* discuss (old AND new), there are two whole forums that deal with this subject.  This forum (Investing During Retirement) and Income and Dividend Investing.  Today's count of the OLD forum postings show 101,056 for IDR, 183,099 for D&I.

Let me suggest to both @RyanM and @Darrin that the wealth of knowledge contained in those two old forums, usually presented in a light, not heat, manner and self regulated, would be lost forever unless a link to each was provided, at least, or a full migration, from old to new, at best.

JMO, H as always, from someone who has been around on BOTH of these old forums whenever they first came about.

ElLobo, de la casa de la toro caca grande
0 Kudos
Highlighted
Participant ○

Re: retirement distribution portfolios

I've always said it's better to be lucky than good!

 

6) Portfolio longevity and appropriateness are overwhelmingly determined by (1) Luck and (2) Withdrawal Rate.

7) If your withdrawal rate is below the SWR (generally 4% real) there is not a meaningful impact on portfolio survivability from factors other than luck.

8) Luck is comprised primarily of two elements: (a) sequence of returns, and (b) inflation. These two factors overwhelmingly determine the likelihood of survivability of a distribution portfolio.

> Luck: If you are unlucky enough to experience significant losses in your portfolio, particularly in the first few years of retirement, nothing you can do will restore your portfolio to its previous level during your lifetime. Unless you permanently reduce the level of your distributions your portfolio is very likely to run out too soon.

From the Morningstar Home page today-

Investment writer Morgan Housel on coping with behavioral biases, the evolution of the investment industry, and the role of luck in investing.

0 Kudos
Highlighted
Participant ○

Re: retirement distribution portfolios

If you are into the math of all of this, Wafe Pfau's book is probably the ultimate source on the subject. It's filled with charts and explanatory text. Personally, I like the Guyton-Klinger spend-down method, supplemented with a small bucket of emergency cash, and a life insurance policy to cover me until i start my pension as late as possible (to collect the mortality credits). https://www.amazon.com/gp/product/1945640022/

0 Kudos
Highlighted
Explorer ○○○

Re: retirement distribution portfolios

The following comments by aubergine are worth mentioning again

"I think the essential takeaways are

a) stay diversified

b) respect the value of lower risk portfolios once drawdowns start

c) unfortunately, spending reductions may be the only solution to market riots, especially in the early years"

Rather than talking about a "safe" withdrawal, the emphasis should be on "safe" spending. Also note that the scheme imbedded in the IRS rules for Required Minimum Distribution will ensure that you don't outlive your accumulation because it requires you to reduce spending if bad things happen. After twenty years into retirement I would add "don't panic" and sell at the bottom of a market drop.

0 Kudos
Announcements