Benefits of Investment Diversification for Retirees

A few weeks ago I noted in my review of Moshe Milevsky's excellent book Are You a Stock or a Bond? that further diversification beyond the US stock, bond and T-bill portfolio he used could provide more benefit. I've done a bit of research and number crunching and take a look at these benefits!

Diversification can:

1. Raise the possible safe withdrawal rate from the investment portfolio by 1.5% a year, perhaps considerably more. Safe withdrawal rate is the percentage amount that can be taken out and spent every year while minimizing the chance of running out of money before dying - a 4% rate on a $100,000 portfolio means taking out $4,000 per year and raising that 1.5% means being able to spend $5,500 per year.
   
2. Reduce the chances of running out of money by anywhere from 3 to 10%.

Most of the benefit of diversification comes from the potential large reduction in volatility, or swings in the portfolio. Diversification helps enormously to reduce the sequence of returns risk (the chance of having bad markets at the start of retirement) which can send a portfolio on a downward spiral from which it never recovers. Retirees face this damaging risk, which is quite different from investors in the accumulation phase, because they are taking money out.

The table below uses Milevsky's formulas, which incorporate the risk of dying along the way, for calculating the risk of running out of money before death. It shows a few hypothetical (but based on realistic numbers) calculations and a couple of results using historical data.

Hypothetical Calculations:

• Weak vs Strong Diversification, keeping the same 4% withdrawal rate - the slight increase in return combined with the large reduction in portfolio volatility (standard deviation from year to year) means that there is a 10% better chance that the money will last till death - which would you rather have an 87% chance of not running out or a 98% chance?
• Weak vs Strong Diversification but boosting the withdrawal rate - for the same risk of 87% chance of successfully having the money last, an extra 2.7% could be withdrawn. That's $2,700 extra for every $100,000 in the portfolio.
• both these scenarios assume a 50% chance of living 19 or more years e.g. a 65 year old getting to 84, which is about the current number according to Milevsky

Historical Data

• TSX from 1958 to 2008 (i.e. including the highly "volatile" 2008!) produced a 2% higher return than the hypothetical scenarios but the volatility is about the same. The comparison portfolio is from the conservative 50 portfolio (50% fixed income) from IFA Canada, which is quite broadly diversified with Canadian, US and international equity holdings, fixed income and real estate, though it does not include other potential asset classes such as commodities and real return bonds. There is still a huge reduction in volatility in the IFA portfolio to 8% from the 15% of the TSX alone. This would have allowed the withdrawal rate to be 1.8% higher with the exact same 97% assurance of not running out.

By playing with the numbers in the spreadsheet, it becomes evident that the longer the expected time in retirement, the stronger the effects.

Bob Clyatt's interesting book on how to semi-retire, Work Less, Live More, shows similar results using the approach of reconstructing actual data for a US investor from 1927 to 2004. He estimates that an internationally-diversified, value- and small-tilted portfolio would allow investors to increase their safe withdrawal rate by 1.5% or more per year. His analysis addresses the needs of much younger people (the semi-retirees) who essentially need their portfolio to last indefinitely, 40 years or more.

PS: Those who want to try Milevsky's formula with their own numbers should read a A Gentle Introduction to the Calculus of Sustainable Income. If you use OpenOffice instead of Excel, be aware that the GAMMADIST function parameters are in the same order as for Excel, not as the OpenOffice help documentation says (that caused me some head scratching till I "reverse-engineered" the correct way to do it).

The moral of the story - in retirement, diversification is a huge opportunity to both boost income and/or reduce the risk of running out.