Friday, September 14, 2012

The 4% Solution in Retrospect

11:48 AM

What would have happened if you retired 30 years ago?

In my previous post, we talked about safe withdrawal rates.

The withdrawal rate is a convenient approach that researchers use to determine how much a retiree can withdraw each year from a portfolio of assets and minimize the odds of running out of money (the "Probability of ruin").

Probability calculations are speculative: they are based on assumptions such as the expected annual return and volatility of the investments. Nevertheless, they are quite helpful for making decisions and evaluating options. They also may be the best approach to plan the future.

The magic period is often taken as 30 years to cover a typical retirement period, e.g. from age 65 to age 95. The withdrawal rate increases each year by the rate of inflation to maintain spending power constant.

In this post we ask: what would have happened if you retired 30 years ago?

The outcome

You recently got rid of your disco shirts and have $10,000 to invest on January 1, 1982. Will it be stocks or bonds?

If you invest in a diversified portfolio of stocks, you will have around $140,000. You will suffer anguish for a few years after the "dot com" bubble burst as your portfolio value turns south. After a steady recovery, your portfolio will suffer with the mortgage-backed security debacle. The stock market will have taken you for a roller coaster ride and you need resolve and discipline to ride out the rough times.

If you invest in a diversified portfolio of bonds, you will have almost $180,000 after 30 years. Most years your returns will be decent, sometimes disappointing, rarely spectacular, but you will have lost very little sleep ... and money.

Historical Returns Comparison -- Last 30 Years

In this illustration I made a few assumptions: equities earn the S&P/TSX Total Return Index. Fixed income returns are those of the DEX Universe Bond Total Return Index (and its predecessor the Scotia bond universe index). In both cases, there is no allowance for investment management fees.

In the calculations below, we'll adhere to these assumptions and also we'll use the actual rate of inflation to increase the annual withdrawal, not a fixed rate such as 2.5%.

The RetireWare Website has a free tool where you can compare returns over various historical periods and asset classes. You can try it here.

What happened in the past?

I will go a little longer than 30 years and look at historical returns since 1964, for the last 48 years. Here are a few interesting facts.

Range of returns

These are the range of returns that one could achieve by owning a portfolio closely matching their respective indices.

Fixed income
Average return
Lowest return
Highest return

Frequency of returns

Now let's look at the distribution of annual returns. The next table shows the number of years for various range of returns in the years between 1964 and 2011 inclusive.

Fixed income
Negative returns
Return between 0% and 10%
Return between 10% and 20%
Return above 20%
Years since 1964

A few observations:

  • Equities have earned about 2% more than fixed income. This is the risk premium. You take more risk, experience volatility and the higher returns makes it all worthwhile.
  • With 14 years of negative returns for equities, you need nerves of steel to stay the course and reap the years of big returns. If you got out of the market at one point, your performance could've ended up lower than bonds.
  • A 2% premium may not look like much but means there will be about 30% more money after 30 years.
  • The "bread and butter" return of fixed income is between 0% and 10%.
  • Equity returns revel with extremes: it is often all or nothing, not unlike ... the casino! 
  • Risk is real. With equities, if things go wrong your capital is at stake. Diversification will mitigate risk, but systemic risk will remain, such as an economic collapse, recession or depression.

As an aside, both historical return series (1964-2011) have almost no correlation. In other words, there is no pattern between bond and equity returns. They didn't tend to move together or away from each other. However, in the last 10 years, bond and equity returns have been negatively correlated. Negative correlation means that if one asset class does poorly, the other tends to do well. This is desirable as it dampers volatility and produces more stable rates of returns in a balanced portfolio.

Now what happens with withdrawals?

On January 1, 1965, would a 4% withdrawal rate do better with a portfolio of stocks or bonds?

With bonds your money would run out in about 25 years. Not bad. With stocks, your money would have lasted 48 years!

I tried retirement various starting years and it was hard to go bankrupt with stocks. Even starting in 1990, with a -14.9% return and assuming a 0% return for years after 2012, money would still last 47 years.

Should we invest all in a diversified portfolio of equities?

Well, unfortunately it didn't go like this all the time. If you retired on January 1, 1982, you would have done better with fixed income.

Allocation to
fixed income
Withdrawal rate that
lasted 30 Years

These are very high withdrawal rates. They are the result of a secular bull market and strong bond returns at the same time (and to a lesser extent ignoring investment fees). Of course, one needed to know the future to confidently withdraw such high amounts each year. But we can see that all asset allocations produced results that would work for even the highest spenders.

Recent times

What about recent experience? If I look at the last 10 years and repeat the sequence three times to form a thirty year series of returns, I will have more modest bond returns and more volatile stock returns.

This time, much lower withdrawal rates can be sustained.

to fixed income
Withdrawal rate that
lasted 30 Years

However, having 40% in stocks achieves the same result as 100% in stocks. 40% in stocks instead of 100% reduces the annual volatility of returns from 20% to less than 8%. Having a moderate portfolio with a lot of fixed income produces the same end result and is much more stable. It provides return-boosting exposure to equities and peace of mind at the same time.

What to make of this?

It is flawed to look at one or a few paths: they actually occured and will never come back. But in order to understand what can happen in the future, we have no choice but to look at the past.

The past has been full of unexpected events that impacted the economy and in turn affect capital markets, interest rates and inflation. The unexpected may very well continue to occur and our best tool to manage uncertainty is diversification.

If I was a betting man, I would say we will continue to see volatile equities and low yielding fixed income. I would also say that achieving a 4% withdrawal rate is a no brainer even in this environment.


Post a Comment


© 2018 Risk Blog by Equisoft Inc. All rights reserved. Designed by Templateism

Back To Top