You Should Be Taking More Risk In Your Fixed Income Investments
“You should begin adding bonds in your portfolio as you age” - This is pretty conventional wisdom in the investment community, and advice you will likely hear from just about every investment advisor. Bonds provide less volatile, fixed returns that can help reduce risk in a retiree’s portfolio.
But we still have a lot of decisions to make when allocating a portfolio to bonds - “Bonds” is pretty vague. Do we add U.S. Treasury Bonds, which are generally considered “safest”, but offer the lowest yields? Do we add very highly rated corporate bonds, which yield slightly more than Treasuries, but technically have a default risk? Do we add “junk bonds”, which may yield 6, 7, or 8% today, but have a much higher chance of defaulting and causing significant losses? Or, do we add bonds somewhere in between, say A or B rated?
To start our journey to find the answer to this question, we are going to consider 6 different bonds of various riskiness based on credit agency ratings (S&P, Moody’s, etc.) that you could buy today:
1) A 10-year treasury note that yields 2.80%,
2) An ‘Aaa’ rated corporate bond that yields 3.83%, (which is based on the average current spread of Aaa bonds to treasuries according to the federal reserve):
3) An ‘A’ rated corporate bond that yields 4.30%, (A slight assumption based on holdings in various bond ETFs)
4) A ‘Baa’ rated corporate bond that yields 4.66%, (which is based on the average current spread of Baa to Aaa corporate bonds according to the federal reserve):
5) A ‘B’ rated corporate bond that yields 6.00%, (That is, once again a slight assumption but based on various holdings in bond ETFs)
6) And lastly, a “Junk bond”, C rated, yielding 6.27%, again the average yield based on Federal Reserve data:
Now, to begin to answer the question of “which bond is best?”, we need to quantify the risks of these different bond investments.
So, what are the risks of each of these bonds?
1) Treasuries are generally considered to be “risk-free”, as the government could always just print money to pay off the bond. For simplification, we are going to assume there is no default risk for U.S. Treasuries.
2) Top rated corporate bonds have also had very few defaults historically. Standard and Poors (S&P) has never had a company that it rated AAA go bankrupt. Moody’s has had 5 companies with its top rating of Aaa go bankrupt in the last 48 years (3 of which were Icelandic banks). We are going to go with Moody’s numbers in this example in order to consider a “worst case scenario”.
Quick point for clarification: One primary decision for a bondholder who is investing in Aaa-rated debt would be what to do if the company’s rating is downgraded. Do you continue to hold, or sell (most likely at a loss)? For this post, we are going to assume that our investor buys an ‘Aaa’ rated bond and holds, regardless of how far the bond gets downgraded.
Under that assumption, the largest 10-year cumulative default an investor would have experienced historically was buying Aaa-rated corporate bonds in 1983, where 3.526% of their bonds would have defaulted:
(This chart, and subsequent ones below come from a Moody’s research report, here: https://www.moodys.com/researchdocumentcontentpage.aspx?docid=PBC_1018455 . There is also an equivalent S&P report available here: https://www.spratings.com/documents/20184/774196/2017+Annual+Global+Corporate+Default+Study/a4cffa07-e7ca-4054-9e5d-b52a627d8639 .)
We are going to assume the worst and say that a fixed income investor in Aaa rated corporate bonds today will face the same cumulative default rate as an investor in 1983.
3) ‘A’ rated corporate bonds worst-case default data, whose worst 10-year period of cumulative losses was 1985, when 5.207% of bonds defaulted:
4) ‘Baa’ rated bonds have had a significantly higher rate of defaults. The worst 10 year history of performance for Baa rated bonds was 1982, where cumulative losses after 10 years were 9.808%:
5) ‘B’ rated bonds, whose worst 10-year cumulative default rate occurred in 1983, when 53.346% of bonds defaulted:
6) Junk-rated bonds. As you can probably imagine, there are periods in history when Junk bonds have performed terribly. 1970 would have been the worst time to own junk bonds, when over the next 10 years a staggering 90.625% of bonds defaulted:
So assuming these worst-case scenarios repeat, and considering the increase in yields you receive for riskier bonds today, where should you invest your fixed income allocation for the best return?
Here is how a $10,000 investment in each rating would play out under these worst-case default scenarios, with today's interest rates:
Here are the numbers for those interested:
Even if the worst conditions we have seen in the last 48 years repeat, at today’s interest rates an investor would be better off buying A (and Aaa) rated corporate debt over U.S. Treasuries.
This will go against how a lot of (maybe every?) financial advisors or asset managers allocate the bond portion of a client's portfolio.
Traditionally, very high allocations are made to U.S. Treasuries.
For example, an advisor who uses a total bond fund, such as Vanguard’s Total Bond (Ticker: BND) or iShare’s Aggregate Bond Fund (Ticker: AGG) is essentially allocating ~40% of a client’s assets in U.S. Treasuries. Here are the current allocations of AGG and BND:
So, sell your treasuries and buy high rated corporates?
Not so fast, there are a few extra considerations to make first. Here would be some added issues by overweighting corporate bonds compared to U.S. Treasuries:
1) Increased Costs
Many brokerages charge no commission for purchasing U.S Treasury bonds directly at time of auction (and you invest on your own and don’t have an investment advisor, could also use Treasury Direct which charges nothing to buy Treasuries at auction).
Buying corporate bonds on the other hand, will likely not be free. And in order to replicate these results here exactly, you would have needed to buy dozens if not hundreds of individual bonds.
But all is not lost!
Could an ETF or mutual fund help you?
iShare’s has an ETF called ‘iShares Aaa – A Corporate Bond ETF’ (Ticker symbol: QLTA) that holds corporate bonds rated only AAA, AA or A (which are technically S&P ratings, but the fund does say that ratings have been converted from Moody’s to S&P, so it should line up with our findings above).
Best yet, this fund holds primarily A rated bonds, which we found to have the highest return under worst-case scenarios above:
The fund has an expense ratio of 0.15%. I factored in a 0.15% reduction in yields from our chart up above (In other words Aaa bonds have an “after-cost yield” of 3.68% instead of 3.83%, and A rated bonds have an “after cost yield” of 4.15% instead of 4.30%).
Even with the ETF’s added expenses, an investor in QLTA would have outperformed an investor in “free” treasuries even under the worst historical default rates in the last 48 years!
So, whether you have a huge portfolio where it is economical to buy many individual bonds, or a small portfolio where it would make sense to use an ETF, based on historical data it still seems to make sense to have a higher allocation to high rated corporate bonds over treasuries – even after increased costs to implement the strategy!
Another issue when moving investments from treasuries to corporates – liquidity. This is improved somewhat by the use of an ETF for the average individual investor, however the liquidity of the fund’s underlying holdings could still create a problem.
I don’t think this is as big of a problem for most, as a proper bond allocation should be set up where maturities match a timeframe when the investment is needed…but the world never goes according to plan. If you have to sell your corporate bond holdings at a time when liquidity is at a minimum you may be forced to take a loss on your investment.
Also, if QLTA experiences huge redemptions, it could negatively impact an investor in the fund. I am not sure how to exactly quantify these “black swan” types of risks, but it is certainly there.
3) “Past Performance is no Guarantee of Future Returns”
All of this is based on the past 48 years of data only. Will we ever experience a period where higher numbers of highly rated corporations default on their bonds? It could very well be likely. If that was the case, Treasuries would likely be the better bet.
Historically, an investor would have been well rewarded by taking additional risk in their fixed income portfolio by adding highly rated corporate bonds over U.S. Treasuries.
Today, many investors heavily allocate the fixed income portion of their portfolios to U.S. Treasuries for “safety” and give up valuable returns that they could receive by investing in highly rated corporate bonds instead.
I’m starting to think that this may be a mistake…
Like most things in life, the best course of action is likely to avoid extremes and take the “middle of the road”. Would I advocate getting rid of 100% of an investor’s Treasury bonds? Probably not. But should I advocate a higher allocation to highly rated corporate bonds than they currently have? Probably….