Seeking Alpha
  • Font Size:
  • Print

In my first post on saving for retirement, I looked at how to calculate the amount of money you’ll need to save annually based on your retirement goals. Now we’re ready to tackle our second topic: how asset allocation decisions can affect your returns.

There have been several major academic studies on asset allocation, including Brinson, Hood & Beebower (BHB) in 1986, which compared pension fund returns with broad market index returns and concluded that 93.6% of the volatility of returns can be attributed to asset allocation. Their research provoked a lot of discussion and debate. In a subsequent study in 1991, Brinson, Hood & Beebower examined pension fund returns through 1987 and found a similar 91.5% allocation-determined link to return volatility.

Yale professor Roger Ibbotson and Morningstar researcher Paul Kaplan went a step further, analyzing both pension fund and balanced mutual fund data and came to a similar conclusion, finding that approximately 90% of the variability of returns over time is explained by a fund’s asset allocation decision. In other words, asset allocation can be a largely accurate predictor of returns over time.

So, asset allocation is perhaps the single most important investment decision you’re going to make, which is why advisors spend a lot of time with clients finding the appropriate allocation at the start. But that doesn’t mean that you can put asset allocation decisions on autopilot. You’ll need to revisit your asset allocation on a regular basis to make sure you’re not taking too much or too little risk in order to meet your goal.

In general, the longer your time horizon, the more risk you can take. That’s why in the retirement calculator example in my last post, I used that 9% return assumption from a 70% stock/30% bond allocation. With 30 years to go until retirement and then another 20-30 years in retirement, I have a relatively long time horizon in which to absorb the volatility of the equity markets.

As you get closer to retirement, however, you should generally be reducing equity risk in favor of bonds or other lower volatility asset classes because you don’t have as much time to make up for a large surprise on the downside. The table below (click to enlarge) shows how even a balanced portfolio can experience negative returns during a bear market.

Source: Bloomberg*

So what asset allocation is right for you? The conventional approach on this question has been to select an allocation (assuming a well diversified portfolio) that seeks to balance the return potential needed to meet your goal with the risk level. There’s also a newer school of thought on this question that factors in your “job sector risk” in a qualitative way to make your asset allocation more or less risky based on the risks inherent in your employment. We’ll discuss job sector risk further in my next post.

*Methodology

  • Stock return data: Wilshire 5000 Index (which is, by definition, only price return, not total)
  • Bond return data: Barclays Capital Aggregate Bond Index
  • Inflation data: non-seasonally adjusted CPI-U (data used by UST when adjusting TIPS principal)
  • Cumulative return = Stock return + Bond return – Cumulative inflation

Asset allocation and diversification may not protect against market risk.

Index returns are for illustrative purposes only and do not represent actual investment performance. Index performance returns do not reflect any management fees, transaction costs or expenses. Indexes are unmanaged and one cannot invest directly in an index. Past performance does not guarantee future results.

The assumed rate of return and other assumptions included in the retirement calculator example referenced above are strictly for illustrative purposes, should not be construed as investment advice and are not representative of any actual performance outcome. Actual results will vary and investors should consider their specific situations when making an investment decision.

This article is tagged with: Investing for Income, Retirement, United States

You may also like

Comments (10)
Add a comment
  • if the 9% return is real, then i believe it is wildly optimistic whatever the length of the holding period.
  • The 9% assumption (actually, 9.1% but I rounded down) is from looking at historical returns on a 70% stock / 30% bond portfolio from 1926 - 2010. It's just meant to be an illustration using historical data, not a forecast on future expected returns.
  • i don't know why you would use this historical data, almost like a problem-solving teaser, if you agree it is not realistic -- this is the source of lots of problems with pension funds and other forward-forecasting entities. you project out 9% based on historical data and all seems well, contributions on pension funding are kept low, then you get 4-6%
  • Ibbotson's later work states that asset allocation only accounts for 15% of your returns. He has recanted on his earlier 90% research.
    I will quote Ibbotson:
    The time has come for folklore to be replaced
    with reality. Asset allocation is very important, but
    nowhere near 90 percent of the variation in returns
    is caused by the specific asset allocation mix.
    Instead, most time-series variation comes from
    general market movement
    Link: www.cfapubs.org/doi/pd...
  • Sorry, but this article is very misleading for the following reasons:
    1. Most readers are looking at investment horizons over 40-50 years or less.
    2. You cannot predict the future based on the past as the last 20 years have shown.
    3. As a previous commentator has mentioned, asset allocation, in itself, is not a great predictor of returns.
    4. You ignore a cash flow model for retirement that can include bonds, preferred stocks and interest paying common stocks. Some advocates would call this a dividend-growth model.

    I switched to a dividend growth model three years ago and have 47% in intermediate bonds (Treasuries, corporates and muni's), 48% in dividend paying stocks and MLP's and 5% in cash. As a result I have sufficient cash flow to pay my bills and sleep well at night while ignoring the gyrations of the stock market. My portfolio is not risk free, but my cash flow is reasonably steady.
  • Yet again that old gem of "asset allocation determines 94% of your returns". The way advisors use the metric and interpret it is pure bunk. They want to sell you their services of AA. Discussion www.retailinvestor.org...
  • Emerald, I would have to agree with the dividend growth model. I have finally found my strategy and have stuck to it for over three years as well. During that time, thanks to the downturn, some good stock choices, and some great dividend increases I have managed to increase my income stream 108%. Granted, I am young and still working so I was starting from a rather small starting point (I have been investing since I was 12 but it took me until I was 29 to finally decide on a strategy). Focusing on dividend growth and the ability of a company to cover its dividend has helped me remain focused (and even excited) during market downturns. In 2001 watching my mutual fund (I only had one at the time) drop over 30% was scary but today I find 30% drops as buying opportunities to increase my dividend stream. It has taken much more of the emotion out of investing and I feel better for it.

    Now, I am not saying my method is superior to others, just that following the dividend-growth strategy (for the most part) has led me to rather nice gains! Anyone who frequently reads my comments will note my bias towards dividend investing!
  • Thanks for all the great comments above. Always nice to generate a stimulating discussion. Actually, I cover cash flow in my first post -- 'getting your number right' -- which 40-50 years into the future is not that easy a task. From a financial planning perspective, understanding future cash flow needs allows you to then find the adequate savings rate into retirement that generates the future value you'll need, ideally at the lowest risk level necessary to achieve that goal.

    In terms of total return assumptions, there are various approaches to use, all with their strengths and weaknesses. I tend to use historical data for illustrations, mostly because, over long time periods (i.e. 40-50 years), mean reversion tends to produce similar return series. Also, keep in mind that these are nominal returns.

    Notwithstanding our current economic challenges, one question I would ask is why should we expect the next 50 years to be significantly different than the last 50 years in terms of broad market returns? I purposely included the years of the Great Depression in my earlier data to balance even more challenging economic times with decades of great prosperity like the 1950s. The U.S. has made it through some pretty rough patches in the past and I'm optimistic over the long-term that we'll find our way through this current period of challenge.

    Stay tuned for thoughts on dividend growth strategies and how to forecast expected returns on this strategy. I'm working on a new blog on that topic this week.
  • One difference between the last 50 years and the next 50 years is that we did not have a welfare society where nearly 50% of our population expecting something for nothing. People in the 40s and 50s expected to work until death or almost near death rather than retiring at 45 or 55 on a government or union pension. We did not have Social Security facing a shortfall or other government assistance programs ran into insolvency.

    But, despite my negative comment, I do look forward to your thoughts on dividend growth strategies. The more discussion the better!
  • "93.6% of the volatility of returns can be attributed to asset allocation."

    The most important word here is "volatility".

    Note that the quote is NOT "93.6% of the RETURNS can be attributed to asset allocation."
About this author: