Accumulation is Different

Research on Safe Withdrawal Rates

Moderator: hocus2004

hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

"Please put forward some suggestions even if they are incomplete or fuzzy."

There are two investors, Investor A and Investor B. Both are 45 years old, both hope to retire within five years, both possess $1 million in accumulated assets, and both expect to spend $40,000 per year in retirement. Investor A is familiar only with conventional methodology studies. Investor B is familiar both with conventional methodology studies and with data-based studies. Stocks are at the valuation levels that applied in January 2000 (they are providing a 1.6 percent SWR). TIPS are paying a 4 percent real return (that translates into an SWR of about 5.8 percent).

Investor A figures that he can afford to take on more risk while he is in the accumulatiion stage than can retirees. So he goes with a 100 percent S&P allocation rather than the 74 percent S&P allocation identified in the REHP study as "optimal." Investor B likes stocks but finds the SWR being currently offered by TIPS so appealing and the SWR being currently offered by stocks so unappealing that he determines that he cannot afford a stock allocation of greater than 10 percent for the time-being. In an attempt to still his strong inner desire for a higher stock allocation, he promises himself that he will up his stock allocation when the difference between the SWRs of the two asset classes is not so great.

Investor B would like if possible to be able to take 4 percent out of his portfolio each year without ever seeing a diminishment of his long-term net worth. He understands that, while he could pull this off with a 100 percent TIPS allocation, he cannot do it with a 10 percent stock allocation at the then-current price levels. But he wants to have some skin in the stock game because he has seen historical data showing that stock prices are unpredictable in the short term. He believes that, were stocks to go up 30 percent in the next year, he might feel an emotional pull to invest in stocks at the worst possible time for doing so. He views the 10 percent stock allocation as "insurance" against the possibility that stock prices will rise in the short term. He persuades himself that he will not need to pay much of a price for this insurance because the data tells him that, so long as he does not sell any stock shares, his long-term return on the money allocated to stocks will be not all that bad (something less than the 3.3 percent real that applies at today's valuations). He cannot imagine any scenario in which he would have to sell stocks to cover living costs since he has 90 percent of his portfolio in non-stock investments. He views the not-so-bad long-term return expectation for the stock money to be good enough as a worst-case scenario.

In the next year, stock prices decline by 50 percent. To accurately determine how each of the two investors is likely to do over 30 years, we need to make an assumption as to whether the two investors will sell or buy stock shares at their new lower prices. The conventional studies assume zero stock sales by Investor A. I find this assumption unrealistic. Common sense tells me that an investor who suffers a 50 percent loss in portfolio value is going to feel diminished enthusiasm for the asset class that caused this loss. What little data we have access to supports this common-sense observation. Middle-class investors have always in the past increased their stock allocations when prices were high and then decreased them when prices were low. It seems unlikely to me that the future will turn out entirely unlike the past. So I think that a more reasonable assumption is that Investor B will lower his stock allocation.

In contrast, Investor B will increase his stock allocation. He did not like the idea of going with a 10 percent stock allocation and did so only because the SWR for stocks was so low and the SWR for TIPS was so high. The fall in stock prices will probably bring down the real return paid on TIPS (since demand for TIPS will go up with a big drop in stock prices). Investor B is in a position to get a premium payment for his 4 percent TIPS and to increase his stock allocation at a time when he can buy many more shares per dollar of accumulated capital than he could have when the SWR for stocks was only 1.6 percent.

Lets say that Investor A lowers his stock allocation to 50 percent. His total portfolio is now worth $500,000. So he now has $250,000 invested in stocks. Let's say that Investor B increases his stock allocation to 50 percent. He suffered a loss of 5 percent of his portoflio value ($50,000) because the value of his 10 percent stock allocation was cut in half, but he had a 4 percent real return on the money invested in TIPS, so he is only down about $14,000. He is not feeling at all bad about that because the other side of the story is that he now can afford to purchase stocks at prices which are consistent with his retirement-plan take-out requirements. The SWR on stocks is now at a level far higher than it was before. And he understands that the SWR number reflects a worst-case scenario. He now has reasonable protection against a worst-case scenario while also enjoying the prospect of big gains in the event that something better than a worst-case returns sequence pops up for him.

Investor B now has far more invested in stocks than Investor A. What are the dollar losses at the end of 30 years for Investor A as a consequence of his sales of 50 percent of his stock portfolio, under a worst-case, best-case, and average-case scenario? What are the dollar gains at the end of 30 years for Investor B as a consequence of his purchases of stocks at prices consistent with his retirement-plan take-out requirements?

The Stock Selling Penalty is a penalty imposed on investors who over-invest in overvalued asset classes. It's clear enough at this point that the penalty is often a severe one. Still, I think it would be helpful to quantify the price that is likely to be paid by investors who overinvest in overvalued assets, presuming that stocks perform in the future somewhat in the way in which they have performed in the past. It is one thing to say that the Stock Selling Penalty is a major risk factor in the ownership of stocks. It is something else to tell the story with numbers from the historical stock-return data.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

I have now addressed this (imperfectly) in my new post More about Imperfect Allocation Switching.

It turns out that the interest rate on the TIPS is more important than I had thought.

It is important to make comparisons among investments, not simply to focus on stocks by themselves.

Have fun.

John R.
Post Reply