Most of the discussions on here seem to be equity related. I was curious if anyone dabbles in bonds as well. I’ve recently been studying interest rate risk modeling to better understand the models and hopefully be able to build models in the future. I guess the material is filling the void until I start studying for CFA Level 2 in a week or so.
I was analyzing a portfolio and was amazed at the higher than peer yield with a lower than peer violatility. Turned out one of the ways the investor achieved this was buying premium, often callable, bonds in this low interest rate environment. I was curious if anyone here engages in buying premium bonds in the current market or any related strategies in fixed income that are interesting.
There is a lot I have to learn with regarding bonds, since its not as ‘sexy’ as equities.
Its a trade off, the more a bond is above par, the more convexity in the price/yield relationship, but it increases reinvestment risk at the same time. If interest rates fall, the high coupon/more convex bond has to be reinvested at lower rates, where as a zero coupon has no reinvestment risk but no convexity, which makes it more sensitive to interest rates.
Callable bonds (options attached or implied) both have negative convexity as price increases, even if they have positive convexity at lower prices… you’ll study more of this next year
Yea, this is true. But what I’ve been reading lately is that premium bonds are actually mispriced in the market slightly due to misconceptions of their yield. And if an investor think rates are moving upwards from current levels, the high coupon actually becomes a blessing because you get the good side of the reinvestment risk. I was just curious if anyone on this forum invested in debt and utilized this or any other similar strategies, given the strange interest rate environment we are in.
“And if an investor think rates are moving upwards from current levels, the high coupon actually becomes a blessing because you get the good side of the reinvestment risk”
Yeah, but there’s an opposite side, if they move down you don’t get as much capital gain and you have to reinvest at lower rates… its just a preference on which risk you’d like to take. Nothing free there, just pick your poison.
Yea, but from what my research shows, premium bonds often trade at a slightly higher yield for whatever reason. So you can pick up extra yield even if rates don’t move, you just suffer if rates move against you. Benefit if they stay still or move up.
Ehhhh… I didn’t read your links, but you’re the boss. From my experience (not research), high coupon bonds trading well over par tend to be sought after and more liquid, which causes them to trade at higher price/lower YTM.
That’s interesting. Is this a recent change due to low interest rate or has it always been that way?
For example: In exhibit 1, the yield on each bond is equal; however, some investors may be unwilling to purchase a premium bond with a final yield to maturity (YTM) equal to a par bond; so the market will generally price premium bonds to offer a slightly higher yield to compensate for the perceived disparity (source: http://www.fa.smithbarney.com/public/projectfiles/5ec1fd74-6f96-44c7-a34a-6e376bef60ee.pdf)
Most certainly not the boss. That’s why I’m curious your opinions Systematic
Higher coupon bonds tend to be cheaper (more yield) due to the higher loss given default, less liquidity, and higher capital outlay. As mentioned above, higher coupon bonds also have less duration. Considering that we haven’t seen Treasury yields this low since after the Great Depression, PMs are betting rates will back up from here and prefer higher coupons. However, a lot of investors are also underestimating the loss given default portion in higher coupon bonds. Bankruptcy court doesn’t care what you paid over par. You’ll often see bonds from the same company with similar maturities but different coupons having similar z-spreads. Just as an example, company XYZ with a ‘BBB’ rating and has a 6% and 8% bond maturing in 2020. The prices are 110 and 120 respectively but the z-spreads are 280 and 285 respectively. So you’re risking $10 more for only 5 bp. The math gets a little tricky in calculating a fair price but it’s definitely more than 5 bp. Anyway, my little rant into bonds.
What about higher coupon bonds that are due to being originated in another interest rate environment instead of credit quality and aren’t callable for whatever reason. Do you find these are less liquid as well?
@rawraw - I don’t quite understand your question but we’re in the environment we’re in because bonds were issued in higher rate environments. Looking back to early 2010, the on the run 10 year Treasury nearly hit 4% and is now trading today at 1.6%. When bonds are issued, the issuer will stay pretty close to par in most cases, paying an appropriate spread for their credit risk. So a ten year bond issued by XYZ company today will likely have a coupon near 4.5%. In terms of liquidity and a high coupon, it’s more about the things I mentioned above. Overall, there are a ton of other factors that will ultimately determine the liquidity of that issue.
It seemed to me you were suggesting higher coupons are result of lower credit quality. I was asking about the higher coupon bonds with a high credit quality requiring them to be sold at a premium, like the 2010 Treasury. When you said loss of excess of par paid in default, I assumed that to mean investing in mid to low tier debt that trades at a premium, but still has high yields.
I think spun was saying that because all rates have come down, basically every bonds trades above par, regardless of credit quality and at the time of issuance they’re sold around par. You’ll see alot of bonds that trade a $1.10 on the dollar today, these are bonds issued at par that have a 10% capital gain. Conversely, in 1982 everything was trading below par due to higher rates.
Spun’s first post are all reason why YTM’s wouldn’t be compareable to bonds trading at par. Perhaps in my few years of work exp I am seeing more liquidiity in these than historically due to more PM’s prefering high coupons/more convexity as spun said. Also, taxable investors would prefer to receive more capital gain and less ordinary income due to preferences in tax treatment, another reason for differences in yield. But none of this is free IMHO, you’ll pay back the few extra bps you get in taxes or its premia rewarded for the higher potential loss of capital in default and etc
Sorry if I wasn’t more clear in my first example. I was talking about one company with different bonds but similar maturities. This takes out the credit quality factor so we’re purely looking at different dollar prices. Sometimes the differential can be more pronounced because you have a bond that was issued over 20 years ago compared to something recently issued. So as an example, let’s say we have IBM bonds. Both have seven years until maturity but one is trading near par and the other is trading around $140. If you notice in the picture below, the OAS on the IBM 8.375% is over 55 bp wider. These are definitely less liquid because folks have a hard time swallowing that much premium. The bid/ask spread is another good indicator.
So in this IBM example, why such a large OAS difference? Would that a result of that issuance being callable – either through refinancing or a call option. Or is it mostly due to an aversion to buying premium bonds? Thanks for the example
The 8.375% issue is a bullet (no call option) and the 1.875% has a make whole call so they might be slightly more expensive (less yield) if at all.
There’s definitely a lot of spread here even for the higher loss given default. Liquidity and the steep capital outlay are probably the biggest factors. To give you an example of liquidity, for trades above 250k, the 1.875s traded 19 times in the past 30 days while the 8.375s have only traded 3 times. These are both benchmark issues (over $500 million) so there’s plenty of paper out there to be traded.
One “unobvious” friction to higher premium bonds - PMs having to explain amortization to clients. On statements, it looks like you’re taking losses on your bonds and it looks worse the more premium you have. I’m sure there’s a threshold were a PM knows they’ll get a call.
I’m still confused by what you keep taking about default. Isn’t refinancing the debt and paying it off at par (loosing the premium paid to par) more likely than IBM defaulting ?
And yes, the client reason is what I’ve read is one of the reasons for the pricing disparity.
Even though it is unlikely IBM will default on its obligations, the probability is still there. It’s the reason an investor will get some excess spread when they buy IBM bonds. For the 8.375s, you’re shelling out $40 more so your loss will be greater in the slight chance IBM defaults. This is part of the risk that needs to priced when you buy premium bonds.
We could make the discussion even more interesting by looking at the current issues in the IBM examples and comparing them to the IBM swap spreads.
Consider for a moment just the 1.875’s. If we went long the 1.875’s and purchased the IBM default swaps, wouldn’t we expect to recieve roughly what 7 year treasuries are paying?
Now consider the 8.375’s. Since they have a higher yield but have the same default risk, why can’t we simply purchase the 8.375’s and buy the IBM default swaps for an arbitrage profit?
Perhaps by going long the 1.875’s and buying the swaps, we actually get less than the return we get on treasuries. But if that’s the case, we could then simply short the bond and sell the swaps and arbitrage a profit that way, right? Well there’s just one gaping hole in this whole strategy that my comments fail to account for…transaction costs. If we were actually going to try to make this happen and contacted a swap dealer, Im betting the bid/ask spread and other costs would render this strategy useless on all ends (unless of course we’re content to collect the rfr).
A little off track, I apologize. But interesting nonetheless and still in the same context. I think the concept of two issues with similar maturities from the same issuer having materially different yields is an interesting one. By looking at swap spreads, it offers another perspective to the same problem.
