One Coupon Callable Bonds

Did I buy a billion of bonds, as notional amounts, in my career? Probably. For sure if I consider various bond buyback exercises. Did I buy every single one on Bloomberg? YES (or…no? I recall once I implemented TradeWeb but I am not 100% certain I ever use it outside some tests. I changed too many jobs).

By bonds I mean bonds, not bond ETFs or bond funds. Meaning each time I had to check issuer, maturity, rating, seniority, yield to maturity, spread and so on while comparing available alternatives. For this, Bloomberg is worth its cost; plus, Bloomberg is the only place where you can trade this stuff at reasonable spreads so, no real alternative.

Last year I changed job (again) and I do not have access to Bloomberg anymore. A friend wrote me saying that a new, weird bond type had become quite popular (in Italy, at least) and if I could check it out. I found myself like the first time I brought my 4yo daughter ice-skating: not that at ease. Was it because I am too accustomed to using Bloomberg for this type of analysis or because the free tools are so bad?

There may be websites out there that are great for comparing bonds (and tell you if retail investors can buy them and where, because that’s another big piece of the puzzle) and I do not know their existence. But it is also possible that retail investors, who seem so eager to buy single bonds (again, at least in Italy), think they are making an informed choice whereas they are like a Chinese lad browsing China’s heavily regulated internet: there’s a whole world out there they do not even know about.

Why Single Bonds?

The legend narrates that single bonds are an effective way to hedge chunky liabilities in the short to medium term.

As for assets, you can calculate the duration of your liabilities: if you plan to spend 2k in 2 years and 3k in 4 years, your liabilities duration is equal to the average duration of the assets that match exactly those maturities. Something slightly higher than 3 years in this example. As time passes by, the duration of your liabilities goes down, making fixed-duration bond ETFs a blunt instrument to hedge your liabilities.

But…

Aside from very special events like saving for your wedding (and even that, let’s talk about it), your liabilities, even chunky ones, tend to have this annoying habit of…being quite recurrent. The moment you get the keys for your new car, it is already time to start saving for the next one (or for the repairs, if you plan to keep it for a decade); you might change the kitchen in your house once in a lifetime but if you now include in that bucket the bathroom(s), the roof, the windows…voila’, another recurring outflow.

Do you get where I am going? Items go out and on your belt conveyor of future liabilities, turning your liabilities duration range-bounded and definitely never zero (unless you die, obv). So let’s say your liabilities duration stays in the range of 2 to 5 years, matching them with a 3 to 5 years fixed-duration ETF would do the job. The duration mismatch, while always there, will never be material.

But…

This would mean that when you have to divest from your short-to-medium saving pot (the ETF), you might have to sell while the pot is in a drawdown. If interest rates go up, the value of the ETF goes down.

Interest rates influence single bond prices as well but, somehow, some savers got convinced that in this case it does not matter because their bond will always mature at 100. The ETF is simply pulling or pushing forward the same cashflows while you “do not see” this effect with single bonds because you do not see the opportunity cost of being stuck with an asset that pays 2%/year when you could now get 4% (what the ETF is showing). Future liabilities just became less expensive to fund (which is also wrong because if rates went from 2% to 4%, most likely inflation went up as well and with it your liabilities but let’s not go there).

The strategy of using single bonds to match liabilities is removing the displeasure of having to sell an asset in a drawdown (obv you could find yourself in the opposite situation, selling while seeing a gain), while it is simply cashflows that move closer or farther in time. Savers face the same volatility on the liability side: sometimes your car breaks down way earlier than expected and, on other occasions, it might last longer than your forecast. In those situations, even the single bond strategy might lead to unpleasant situations: if you have to sell earlier, your bond might be worth less than 100 and if your bond matures before you need the funds, you might re-invest at a lower rate. But we choose to see the latter as somehow better than selling an ETF at a loss, even if they are exactly the same thing.

If you are still confused, I understand. Maybe try with Cliff Asness, point 10 of his pet peeves.

I AM still confused because I know more about single bonds than 99.99% of you out there and I never bought one for my personal portfolio. Never found the reason. But I also have 4 different ways to match those liabilities: my monthly salary, the annual bonus, my portfolio asset allocation and…avoiding those liabilities in the first place. For example, I do not need a car (that’s what happens when you do not live in a place taken hostage by 3 taxi drivers with decent public infrastructure) by “life design”, a choice that has nothing to do with frugal bs.

All considered, I hardly see the return in investing the large required amount of time to learn which bonds to buy and where to buy them. (Yes I understand that once you have done it, the maintenance cost, at least system-wise, is pretty low)

One Coupon Callable Bonds

Here is an example description of one. I am not sure if this is a Europe-wide phenomenon or if the frenzy is limited to Italy. Issuers are global banks (Societe Generale, Barclays, Goldman Sachs) so…might go either way. In any case, I will try to generalise this post as a guide on how to evaluate non-plain vanilla bonds with limited tools.

Let’s translate the characteristics of the bond I took as an example: it is a Senior bond issued by Societe Generale with a fixed coupon of 6.5% and maturity in 2038 (the bond was issued August last year and at issuance it was a 15-year bond). SG can also repay the bond at par (price=100) every August from 2024 until 2037.

The jargon:

• ONE COUPON means the bond pays…drum roll…just one coupon at the maturity of the bond. No annual or semi-annual or quarterly payments. Also, this is crucial, the accruals on that coupon are not compounded. If you buy a plain vanilla bond, you can re-invest the coupons you receive during the life of the bond, allowing you to enhance your overall returns. Actually, that’s the action implied in the calculation of a bond yield-to-maturity. Since this is the market convention, if SG wanted to be “fair”, they would have added the coupon accruals to the bond notional after each year, a mechanism to allow the compounding of interest. But they don’t. So that 6.5% you see is not comparable to the 6.5% you see attached to a plain vanilla bond. The “real” CAGR of the bond is as follow (the longer the bond is not called (see later), the lower the CAGR goes):

• CALLABLE: linked to the bond, SG is buying what is called a “string of options” from the bond investor, i.e. the opportunity to repay the bond at par each year according to a pre-defined schedule. It is called a string because there is one option for each year.

The bond has also a quite interesting KID. According to the doc, the bond has an “entry fee” of 2.4% and an “exit fee” of 0.3%. If you are surprised by these costs, welcome to the party. It might be that the regulator wants the issuer to include “proxy costs” in the KID, like the worst possible scenario to consider bid/ask spreads, broker fees and whatnot. Or maybe these are real costs. My point is that you hardly find this s*&t when dealing with big ETFs because others did this type of due diligence for you. When dealing with single bond issuances, you have to pay attention to all the details if you do not want to have surprises. Any additional research you do, costs you, at least in terms of time.

We already saw why the ONE COUPON feat can be deceiving. Let’s have a look at the CALLABLE one.

By selling optionality to SG, the bond investor should be compensated by receiving a premium. According to the arbitrage-free framework, the value of a bond with an embedded option is equal to the arbitrage-free values of its parts—that is, the arbitrage-free value of the straight bond and the arbitrage-free values of each of the embedded options. Any investor can create a “synthetic” version of this bond by buying a SG plain vanilla senior bond with equal maturity, selling a string of options on the same bond and…a swap to replicate the one coupon feat (not really relevant here, I just added for the sake of completeness). The point is that the synthetic version and the original one have to cost the same, otherwise I can go long one and short the other and create a risk-free profit.

Because the call option is an issuer option, the value of the call option decreases the value of the callable bond relative to an otherwise identical but non-callable bond. So if the price goes down, the yield goes up. Meaning, when you see that 6.5% you should not think “OMG free money!1!” but “I am being compensated for the optionality I am selling, am I receiving fair compensation?”.

Usually, the reply to that question is NO. First, because not many investors understand the concept I just described. And second, because pricing those options is impossible if you do not have Bloomberg.

An easy thing to do is to check how big is the callable coupon compared to the straight bond. You just have to type SOCGEN Corp <GO>…ah no, without Bloomberg. On SG’s website I can check if they issued a senior bond with maturity close to 2038 but then I’ll not have the updated price. Koyfin and TradingView do not have a suitable search function for corp bonds and I do not want to pay for YCharts.

The issue with bonds is that they mainly trade OTC, over the counter, not on an exchange. I checked on EuroTLX, a bond exchange for retail investors, knowing that what I find there is only a very small subset of outstanding bonds. I found other SG bonds but nothing with a comparable maturity. I went on IB to see what I could trade there and finally found this one:

It is a similar Senior Bond that matures on the 25/01/36 and rated A1/A/A+ (order is Moody’s, S&P and Fitch). BPCE is another French bank. This is our SG bond, priced at maturity:

The premium the callable bond investor receives is roughly 4.48% – 3.83% = 0.65%. I’ll let you judge if all the time we spent analysing this thing is worth that premium; I would also let you compare those 65bps with the initial reaction you had reading about a bond that seemingly was paying 6.5%, i.e. how good SG marketing dept is.

The Call

Fact is, those 65bps are not even guaranteed. That’s the best possible scenario, i.e. the bond is not called in the next 14 and change years. What I found really curious is that my friend told me these bonds are sold as a way to profit from the (not-really-so-sure) incoming interest rate cuts. The (very weird) logic is that if the ECB cuts IR, SG will be incentivised to call the bond in August 2025 and investors would receive a 6.5% return for one year.

Even without knowing anything about the bond, this would be the first time ever I see a seller of an option being happy that their option is maturing in the money.

In fact, if the ECB indeed cuts rates, the maximum gain you can have with the called SG bond is 6.5% (let’s leave aside the costs written in the KID): you bought it at 100 and it is reimbursed at 100 one year later.

But the price of the BPCE bond would also increase because rates went down. If, for the sake of argument, we assume it has a duration of 10, its price will increase 2.5% for every 25bps cut in IR. This means the total 1yr return for the investor would be 3.875% (coupon) + 2.5% (capital gain) = 6.375%. With just a single 25bps cut (assuming a parallel shift in the whole curve, I’d spare you the fact that different points in the curve move in different manners), the total return on the 2 bonds is almost the same! The kicker is that the more the ECB cuts, the more the BPCE price will go up while the SG one is pinned at 100. That’s the upside the option seller is giving away.

“What are you saying, the SG bond price today is ABOVE 100”

Yes, but that’s because we are still quite far from the call date AND the market is not pricing a high probability that SG would call. There is anyway a cap on the max price this bond can reach because the lower rates go, the more the bond is priced with a duration linked to the first call date instead of its natural maturity. In other words, the max a buyer is willing to pay is a price that will not convert into a return, until the first call date, lower than the risk-free rate.

[For the lads interested in more technical stuff. When the option is near the money, the convexity of a callable bond is negative, indicating that the upside for a callable bond is much smaller than the downside]

Rule #1: if you want to bet on something, find the most straight way to express that view.

In this case, if your bet is that rates will go down, buy a long-duration government bond. That’s it.

There are also some nasty, albeit fringe, possible situations.

Maybe rates will indeed go down but for the right reason: the economy would be in a really bad situation. Guess what will go up? Banks’ credit spread. Maybe rates won’t go down but SG manages to bring to the spotlight another Jerome Kerviel (I’ll let you Google who he is). Guess what will go up? SG’s credit spread.

In those situations, SG will be more than happy to not call the bond and sit on what is now cheap financing for them. While investors hold the hot potato without even the booby price of some interim coupon payments. But I understand the mood, it has been ages since the last big European bank needed a rescue and some of its bondholders (the junior one, to be fair) got wiped out. The FT had very nice words about the bank a few months before these bonds were issued:

“He [the new CEO] added that although SocGen had been through a series of “nightmarish” situations during Oudéa’s reign — including a perilous eurozone crisis, heavy losses from its once-vaunted trading division and a €3bn loss on the sale of its Russian subsidiary — the bank had shown its resilience.“

Let’s actively try to buy an instrument that clearly would thrive if rates go down and it is riskless otherwise! 😉 😉 😉

65bps on a yearly basis is a material premium: if you are a bond fund manager, this investment might offer you a legit way to beat the benchmark, if you hold the view that rates are staying put or might go up. But you are not a bond fund manager. In particular, the universe of assets you can invest in is not constrained at all to just bonds and, more importantly, you have no clue where rates are going.

You can achieve that “premium” by designing an asset allocation that fits your objectives and stick to it. But again, if you want to be clever and express your view on where the market is going, keeping it simple should be always the rule.

[the mindful between you might argue that 65bps should reflect the correct market value of the string of options, at least once the bond is traded on the secondary market, because of the arbitrage-free framework. My theory is closer to passive investing has broken the market. I cannot even find the total size of this issuance, damn it Bloomberg!, but I suspect you do not need many price-insensitive buyers, i.e. lads that are lured by the higher coupon without understanding the product, to keep the price artificially high. Also, shorting these bonds is neither easy nor cheap…and the size of the opportunity is, again, constrained. But if you want non-financial advice, never buy this stuff on the primary market, just to be sure]

What I am reading now:

Follow me on Twitter @nprotasoni