Retire Early Home Page
Hocus's SWR Questions

Format for Printing

Format for printing

Request Reprints


By JWR1945
August 13, 2002

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

For CatherineCoy and others who have expressed interest.

This may help you understand what hocus is after. When he mentions a variety of Safe Withdrawal Rates (SWR) for different asset classes, think along these lines:

1. The thing that keeps a stock market SWR down to 4% (after adjusting for inflation) is volatility. Maintaining a constant income during retirement requires that you sell more shares when prices are low than when prices are high. In essence, we are talking about dollar cost averaging. It is a great idea during accumulation. But it works against you when you are making withdrawals.

2. In many cases investing in stocks is the best long term investment choice. It certainly is if we restrict ourselves to a few specific investment choices. Stock market investments include a growth element that is missing from bond and other fixed income investments.

3. Because of volatility in the stock market, it is a good idea to have some bonds or money market funds or some other fixed income investments. If stocks were to drop suddenly and sharply (such as having the Dow Jones Industrial Average drop to 1600), you would still have some money after you make your withdrawals and you would end up buying a few shares of stock at bargain prices when you rebalance your portfolio. (Actually, you would be selling fewer shares than otherwise because you should withdraw and rebalance at the same time.)

Now let us expand our outlook.

Since the problem with stocks is their volatility, why not look at alternatives that have lower volatility than the overall stock market but still have an element of growth as well? I know that rkmacdonald and others have invested in REITS to reduce volatility. Along the same lines, some people have selected stocks with high or steadily growing dividends. Stock prices and earnings fluctuate considerably, but dividends tend to remain steady or grow steadily over long periods of time. Their volatility is very low.

Remember that during retirement we are looking for a steady income stream (after adjusting for inflation). Some of this may be by selling stocks. But part of it can be from bonds. Part of it can be from dividends.

Remember also how direct ownership of bonds differs from owning bond funds. If you own a bond, you can choose to hold it to maturity and receive a well defined income stream. The prices of bonds fluctuate considerably as interest rates fluctuate. But you can maintain your income stream without worrying about those fluctuations. With bond funds, however, payouts fluctuate because new bonds are purchased and old bonds are sold or cashed in at maturity all of the time.

REITS and stocks with high dividends have some of the characteristics of bonds, some of the characteristics of bond funds and as well as some of the growth features of stocks in general. Even if their prices were to fall dramatically (again equivalent to having the Dow drop to 1600), their dividends (ideally) would remain the same. With these asset classes, the percentage of your retirement income derived from capital gains would be relatively small. You would need to sell a smaller percentage of your REITS and high dividend stocks than otherwise. It is likely that you would be able to use a longer time for rebalancing your portfolio and that you would have a little bit of flexibility in choosing when to sell. Although you should not expect to get the very best prices when you sell, you might be able to avoid selling when prices are ridiculously low.

There may be subtle differences in risk versus reward tradeoff characteristics. There might be other characteristics that would allow people to retire earlier. For example, if you had a high probability of success but not certainty, you might choose to retire right away if you could be certain to know your ultimate financial outcome within two or three years. If you could not be sure whether your retirement income would be secure until ten years later, you would probably choose to delay your retirement.

Now the embarrassing question.

What would be a Safe Withdrawal Rate (SWR) for a portfolio that includes REITS and high dividend stocks as additional investment classes?

The accurate answer.

We don't know.

An elaboration of the answer.

We can't know. The historical record on REITS is very short and the dividend payout data are confounded by the effects of taxes.

What can we know?

Quite a bit.

For example, in the Retire Early Safe Withdrawal Rate study, intercst has made a sensitivity study (i.e., he has looked at some "what if" type questions) related to portfolio diversification. That addresses the issue of volatility...but from a different vantage point than we are talking about here.

There are models that can project the effects of hypothetical investment categories. For example, we could hypothesize three of four possibilities of future REIT characteristics. The models could then give us an idea of what to expect in each case. But much more importantly, the models could help us know right away if we guessed wrong. It would allow us to make changes and salvage our portfolios. Of course, it would also help us to know whether those models are any good and when they can be used safely and when they cannot.

Along these lines I again direct you to the Retire Early study. intercst looked at the importance of the length of historical record. If you had projected stock returns based on the years immediately after World War 2, you probably would have concluded that a 10% withdrawal rate was a good idea and you would have ended up with a busted retirement.

The usable history of REITS and high dividend stocks (in today's tax environment) is short. There is a significant element of risk in making projections. We need reasonable estimates. We need reasonable projections. It would be a real benefit to a whole lot people if we could introduce these asset classes into our withdrawal strategies. The key is being able to spot trouble right away.

Why not just propose a solution?

I think that you can see why we should not expect hocus or any other single person to propose a comprehensive solution. There are some highly technical issues involved and, at the same time, there are a whole lot of opportunities for people with non technical backgrounds to help out as well. Whenever projections are made or statistics are used, the most valuable contributions are the common sense observations. These are often called sanity tests or the emperor has no clothes type arguments. Although hocus does not claim technical credentials, he has identified hidden flaws in many investment strategies. There is also a lot of value in personal experiences. This includes identifying our own blind spots for different investment classes as well as how individual circumstances influence the appropriate criteria for investment selection and retirement withdrawal strategies. (This includes much more than risk tolerance alone.)

This is not a comprehensive description of the issue.

My intention is to improve your understanding of hocus's questions. I want you to understand that hocus's usage of phrases such as the "Safe Withdrawal Rates for different investment classes" is appropriate even though the underlying issues are quite complex and often subtle and obscure. I have restricted my comments to a very small subset of what hocus is interested in.

I also realize that CatherineCoy and FoolMeOnce and others who invest in real estate have a lot to offer and a lot to learn as well. I think that they would like to see Safe Withdrawal Rate information that includes rental property. Of course, many of us would prefer to avoid the hassle factor in rentals.

Although I have only mentioned it in passing, the nature of any risk is important. Having a mean, median and variance usually provides a good starting point. But there are a whole lot of other details that become more and more important as we demand greater confidence in our conclusions. The Retire Early study is very good in this respect. But we are trying to extend it.

Have fun.

John R.

Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.