The 4% Rule is not as good as we hoped – Part 2: Bond expected returns

Part 2 of our multi-part series deals with bond returns. During the 1871 to 2016 time span, 10 year government bonds returned over 2% in real terms. It’s hard to see how bonds can continue that trend in the foreseeable future.

Part 2: Bond Expected Returns

If you thought stock returns might disappoint it looks even worse for bonds. At least for equities, hope springs eternal and the bull market that started in March 2009 might as well go on for another few years, maybe another 10 years. As we showed in Part 1 of this series, even with an equity earnings yield (CAPE) as low as it is today (under 4%), it wouldn’t be outrageous to see real equity returns north of 6% over the next 10 years.

In the bond world, unfortunately, your low returns are already baked in right now. See the table below, where we list bond ETF yields as of 3/31/2016 according to iShares. Today, safe government bonds have a yield just under or maybe barely above anticipated inflation rates. If we assume a 2% inflation rate over the next 10 years you are left with a negative real return with bonds. Going to longer maturities you might squeeze out some extra yield, maybe even a slightly positive real yield with the 20+ year maturities, but that also comes at substantially more risk than the 7-10 year bonds. So, trying to diversify your equity risk with an investment in bonds you would have to give up a lot of return from the already low expected equity return, see part 1 of this series.

Bond ETF yields (iShares as of 3/31/2016)

With corporate bonds you can get a little bit of extra yield. AGG is a mix of government bonds and investment grade corporate bonds, which still offers only 2.33% nominal yield. LQD adds a full extra percentage point of nominal yield but at the expense of now having all corporate bonds with a slightly bigger correlation to equities and, no doubt, also higher default risk. For very, very safe corporate bonds, the yields are going south. Check this recent headline in the WSJ: “Unilever Almost Borrowing for Free Following ECB Bond-Buying Plans.” Basically, Unilever (European Anglo-Dutch corporation) recently issued a 4 year bond at zero % coupon at a little bit below par for an annualized yield of 0.08%. Not 8%, but 0.08% annual yield! That’s for a corporate, very high credit rating bond!

High yield, i.e., sub-investment grade bonds, also called “junk bonds” yield a lot more. After 2% inflation they have a pretty decent real expected yield, though still below 4%. But there are two caveats: 1) the yield is before any defaults, which could eat into your return very substantially, 2) junk bonds have a high correlation with equities, so the diversification benefit is limited. Not a pretty picture for bonds.

Are low or even negative realized bond returns something extraordinary? Not at all! In the past, realized 10Y real returns have been even worse, all the way to -4%. So, a zero expected real return is nothing out of the ordinary! How can bonds have poor returns, even supposedly safe government bonds? Simple: inflation can erode your real returns and rising bond yields lower your bond prices. Of course, the one upside of bond funds is that if interest were to go down even further the value of bonds would go up. That would sound like good news, but it isn’t really:

  1. The gains would be only a short-term blessing because future yields would then be even lower than they are today. From the frying pan into the fire!
  2. The reason for the substantially lower yields would likely be another recession, so for every dollar you make in your bond portfolio you will likely lose five in your equity portfolio. Remember, even with a 60% equity and 40% bond portfolio, the vast majority of your portfolio volatility still comes from equities. Specifically, a 60% S&P500, 40% Barclays US Aggregate Bonds portfolio had a correlation of above 0.98 with the S&P500 over the last 30 years.

So, if you apply today’s yields of around 1.8% and subtract 2% expected inflation you are looking at a zero or even slightly negative real yield, see the green line that extrapolates 2% inflation over the next ten years. This would be bad news for the early retiree who relies on bonds as a diversifier.

Realized CPI adjusted bond yield (10Y average CPI less 10Y yield from 10Y ago)

This is all making the 4% withdrawal rate less sustainable. Over the past decades or even century, bonds had an average return about 2-2.5% above what you can expect over the foreseeable future. Back then, sprinkling in maybe 20-25% of bonds into your stock portfolio had the beneficial effect of diversifying equity risk while still keeping the overall portfolio expected return well above the 4% withdrawal rate. Simulations on cFIRESim and from the Trinity Study looked pretty good when using a small share of bonds. But going forward, assuming a 3.5% real equity return and 0% real expected bond return, a 75/25 portfolio has an expected return of 2.63%. A 50/50 portfolio has an expected return of only 1.75%. Even without any market volatility we would deplete our portfolio pretty rapidly at the 4% “safe” withdrawal rate, and with volatility potentially even faster.

What are the remedies?

  • A lower withdrawal rate, at least until equity expected returns have recovered again
  • A low (or zero!) bond allocation, which in turn means having to take on more equity risk
  • Explore other non-traditional investment options:
    • Real estate still has the potential to generate a rental yield of more than 4% (without leverage, after taxes and costs). Your returns will likely be mostly uncorrelated to the equity market. REITs are not a panacea because their returns are highly correlated with equities and the dividend yields are not as generous as they used to be
    • Juicing up yield through equity option writing strategies
    • Buying bonds for diversification on margin (through US Treasury Futures). This is can avoid the large opportunity cost of bond investing while still getting the diversification benefits. But it might sound too scary and risky, so it’s probably not for everyone
    • As always: stay away from the scams!
  • Flexibility in your budget: have a plan B, plan C, and so on to be prepared for the some of the expected and unexpected challenges in retirement
  • Own your primary residence (outright without a mortgage). Not having to pay rent reduces your monthly mandatory expenses and gives more wiggle room when you have to reduce spending

Stay tuned for future parts of this series!

Intro: Pros and cons of different withdrawal rate rules

Part 1: Equity expected returns

Part 2: Bond expected returns

Part 3: The small-sample problem in historical simulations

Part 4: More bad news on equity expected returns

10 thoughts on “The 4% Rule is not as good as we hoped – Part 2: Bond expected returns

  1. Might want to make this “fixed income” instead of bonds. CDs are also a bit of a losing proposition, but at 2.42% (best current 5 year CD) they seem like rather a better deal than bonds. Fact that you can break them for a few months interest cost really reduces the risk…

    1. Good point. Definitely better rates than 5y bonds and if you play it smart and stay under the FDIC max, completely free of default risk. But still not a viable option for getting your portfolio return towards CPI+withdrawal rate (2%+4%=6%)

  2. I agree that real estate is a good option. I am currently assessing a commercial rental opportunity that should give a net return after tax of 4.7%. I plan to get the ball rolling on it ASAP since frankly it is unlikely I can find something to beat that kind of return at the moment.

    1. We like RE too. At least it’s mostly uncorrelated to equity risk (if you do the RE investment directly, not though a REIT). 4.7% return would also likely be inflation-adjusted, so much better than the sustainable safe withdrawal rate from equities these days!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.