Don’t forget to enter the contest to win a copy of the first authorized biography of Warren Buffett – “The Snowball: Warren Buffett and The Business of Life”. Click here for details – free to enter!
Over long periods of time, the conventional wisdom of diversifying your bond portfolio with longer-term bonds and/or lower-credit bonds (in order to get higher yields) seems silly on a risk-adjusted basis. But let me explain the logic behind this assertion.
A few assumptions
- A diversified basket of equities might earn 8% nominally (meaning before factoring loss in purchasing power to inflation).
- A diversified basket of shorter-term, investment grade bonds (BBB or higher) might earn 4% nominally.
- Longer-term and/or higher credit risk bonds (BB and below) might earn 6.5% nominally.
Let’s assume we have an investor with a 50% equity / 50% fixed income portfolio, and that the fixed income portion consists of the shorter-term and high credit quality bonds. The investor might have a long term return of 6%, which is just the weighted average of 50% equities x 8% return added to 50% fixed income x 4% return.
Someone might advise this investor that they can add high-yield fixed income to their portfolio in order to increase their return – and they might take their 50% fixed income and split it into 25% high quality fixed income and 25% high yield fixed income. Therefore instead of earning 4% on their fixed income portion they would earn 5.25% on the fixed income portion of their portfolio (50% x 4% + 50% x 6.5%). If we then re-calculate the expected long term return of this new portfolio we would get 50% x 5.25% + 50% x 8% for a total expected return of 6.63%.
An Alternative
An alternative to adding high-yield fixed income to increase returns would be to simply have more exposure to equities. Doing some quick math we find that a 66% exposure to equities and a 34% exposure to high quality fixed income will give us a 6.64% long term return (66% x 8% + 34% x 4%).
Adding Equity Exposure Is A Better Solution
What’s described above does not tell you much about the risk-adjusted returns, so let’s explore that some more.
First, the equity premium (return of stocks over t-bills) has been 8.25% for the US stock market between 1927 and 2005. The “maturity” factor (excess return of long term government bonds over short term government bonds) is only 2.09%, and the “default” factor (excess return of long term corporate bonds over long term government bonds) is only 0.36% for the same 78 year period. To put it another way, you have to spend much more risk (in the form of lengthening terms or decreasing quality) to eek out higher returns from fixed income investments.
Secondly, short term US fixed income has very poor correlation with the S&P500, but the correlation gets larger as the maturities of the fixed income extends.
Maturity Correlation with S&P 500
1 Month -0.08
6 Month +0.01
1 Year +0.05
5 Year +0.22
20 Year +0.30
If you have multiple asset classes (each with net positive long term return expectations), the overall risk-adjusted returns will be higher if the correlations between asset classes are lower.
Conclusion
Taking this all together, the risk adjusted returns will be much higher by increasing the equity exposure as opposed to trying to seek a higher return through high-yield fixed income (by extending maturities or decreasing credit quality).
Preet
Very timely post, as I have been mulling the very same topic recently.
I would be interested in your views on how dividend stocks and preferred shares factor into this picture (especially given their after-tax advantages).
Regards
q
Preet…does the current available yields on some reasonably safe yet currently high yielding debt change your opinion at all. Historically I agree with your analysis but we are in unique times and some investment opportunities are available to us today that may never be available again in our lifetimes.
@45free – an excellent question. It doesn’t change my thinking, but nor does that preclude taking advantage of the high spreads of corporates over govs, since many investment grade corporate issues still fall into the category of higher quality and are also providing high yields. Long term, I would expect that equities will outperform bonds and even given the unique circumstances now that is probably just as true as ever since stock prices are also unique.
Also, I subscribe to a core-satellite approach where 80% goes into a disciplined portfolio, and 20% is dedicated to individual picks and active trading. Allows me to have structure while appeasing my human nature! :)
@q – I was watching the election in the background so my mind was all over the place while writing the post, I remember having the intention to specifically address the tax effects as well – obviously the preferential tax treatment of equities further supports the argument, but adds nothing for tax-sheltered accounts.
Now, preferred shares are a bit of a different animal. I don’t think there are too many big fans of perpetuals out there, so the same restraints on what slice of the preferred share market you add to your portfolio is probably not a bad idea, i.e. keep them higher credit quality and shorter maturity (or at least shorter time-to-reset). Other than that, I’m fine with adding prefs to the fixed income mix in moderation.
Preet A follow-up. Where do you see dividend paying stocks (with DTC advantage) in this mix. Same as other equity, better worse? Thanks q
Preet, some excellent points that match my thoughts.
Another source of information along the same lines is one of my favourite books, David Swenson’s ‘Unconventional Success’. He suggests any desire to earn anything above government bond rates should be achieved by increasing exposure to equities, not investing in riskier bonds. Some reasons for this include:
– corporate management teams interests are more aligned with stockholders than bond holders. The management team typically holds stock or has stock options. So if they have to choose whether to negatively impact bond holders or stock holders, they’re likely to choose bond holder.
– the return on most corporate bonds have only one way to go.. down. They typically have clauses that allow them to recall the bonds (and typically reissue at a lower rate) if interest rates go down or their credit rating goes up. If their credit rating goes down or interest rates go up, you’re holding a suboptimal investment.
I think there were some more points as well, but I’d have to re-read the section on corporate bonds to remember them. But your conclusions are essentially the same as his: the extra risk of high yield corporate bonds doesn’t compensate with as much expected return as the same amount of risk taken in equities.
@qas – I’m a big fan of dividends! :) The question of dividend paying stocks being better/worse than other types of equities is probably worthy of a longer, more in-depth response, but in a nut-shell: they would work very well with this mix. Having capital gains only would shelter the growth (and defer taxes), but historically dividends have been a key driver behind wealth creation.
@Returns Reaper – I have not yet read Swensen’s book, but it comes highly recommended by many. Your points mentioned are excellent, and may help explain why you need to spend proportionately more units of risk trying to earn higher units of returns from bonds versus equities.