Moody’s US downgrade is not worth losing sleep over

A significant proportion of operatives within bond markets don’t really understand yield curves. As once again revealed after Moody’s latest downgrade.

Moody's headquarters in New York (via CX56RK/ Alamy)

I am about to step out onto some very thin ice. And on Sunday, when the news broke that the credit rating agency Moody’s had stripped the sovereign Unites States of America of its “coveted” triple-A rating, that ice got a little bit thinner.

Why is it so thin and what do I mean by it? I frequently refer to myself as a superannuated bond dog which means that my head and my heart have for over four and a half decades lain in the fixed income markets and although I have for going on a decade been retired from the sharp end of bond trading I still mostly look at where short term money market rates are trading - as well as the key 2 year and 10 year US Treasuries, UK Gilts and German Bunds - before I look at the closing prices of the corresponding stock markets. I am happy to argue the toss over which market is the tail and which is the dog.

I hugely miss my late lamented friend Morris Sachs, formerly of the Inside Baseball with Old Chestnut podcast with whom I spent many hours on many phone calls discussing the private life of the yield curve. We agreed that the line “I understand equities, but I don’t understand bonds…” was fatuous but also that very sadly a significant proportion of operatives within bond markets don’t really understand them either. And much of that was once again revealed after Moody’s downgrade.

Although there is an inordinate amount of mathematics involved in bonds, most of it is really simple arithmetic. At the top of the pile sits the yield to maturity which for all intents and purposes is supposed to tell the observer what the real return on an investment in a given bond will be if he or she holds the security from the day of purchase until its maturity and reinvests all coupon payments to the matching final date. In reality, the yield to maturity is a complete fiction for the simple arithmetic formula assumes that every coupon payment – conventionally delivered every six months – can be reinvested for the residual period at the original rate at which the bond was first bought. Confused? Let me try to explain again.

Say the prevailing rate for a 10 year bond is on the day of purchase 5.00%. I will in my calculation assume, quite stupidly if you want to know, that I will be able to reinvest my first coupon payment at 5.00% for the remaining nine and a half years and the next one at 5.00% for the remaining nine years and so on and so forth until I reinvest my last coupon that is paid after nine and a half years at 5.00% for the remaining six months until the bond is redeemed and my invested capital is returned to me. The chances of this ever happening are infinitesimal.

The chances of one shuffling a pack of cards and them all coming up in the right order from ace down is 1 in 8.0658 × 10^67 which means that if one could shuffle the deck every second for the entire lifetime of the universe – thought to be around 13.8 billion years – it’d still never occur.

I sadly don’t happen to have a spare supercomputer hanging around in my garage but if I did and I put it to work my guess is that the chance of a perfect reinvestment outcome would be somewhere in the same category. Not that that really matters. It is not the precision of the algorithm that is important but the use of the same methodology on all bonds. Thus, and subject to the identical fundamental error, one can begin to compare one bond to another and ultimately the only scientific bit about bonds is in the comparison. As long as they are all subject to the same error, the comparison is valid. Please bear with me.

Now assume the same issuer launches at the same time two bonds with identical maturity dates but paying different rates of interest. For all intents and purposes they are identical, but due to the variation in the interest rate the cash flows are different. The bond maths is now used to level the playing field. Both are, say, nominally worth 100,000 dollars or pounds or euros. The market will now apply an upfront premium to the price of the one that pays more interest which in theory is equivalent to the extra interest the investor will receive over the lifetime of that bond. It’s not rocket science until one begins to try to adjust for so-called reinvestment risk. I don’t know where the rates will be at which I can reinvest my coupon – it won’t be as noted above at the same rate as the initial coupon on the bond but – I know it’s getting complicated – if rates are set to go higher I will want lots of coupon to reinvest at higher rates, if they go lower I want as little as possible to have to invest for a lower return.

Thus one eventually finds out that the bond with the lowest reinvestment risk, whether to the upside or downside is immaterial, is the bond that pays no interest at all. That is the zero-coupon bond. No need to panic. Instead of coupons being paid periodically, they are deducted from the price of the bond. It is sold to investors at a discount equivalent to the interest which would be received if they were attached.

Zeros (as they are known in the trade) are in fact dead cool because they don’t carry any investment risk but they also don’t produce periodic coupon payments which is for what most investors buy their bonds. Were I to go a level deeper, which I shall not, I would go on to explain that a 10 year bond with 20 semi-annual coupons could and should really be treated as one capital payment with 20 zero coupons attached. We have the technology to calculate the price of bonds that way and the interest rate swap market uses it but, alas, it is what it is.

The bond market has a variety of ways of arithmetically adjusting bonds with differing maturities and different coupons in order to make them as directly comparable to one another as is possible, but the methodologies are from the outset in effect all wrong. As in the fable of Plato’s cave, however, so long as everybody is living under identical misconceptions there is nothing to fear.

Now, not all borrowers are equal and a further adjustment in the yield has to be made in order to compensate for a varying possibility or even probability of any given borrower not being able to either meet periodic coupon payments or, heaven forbid, the principal amount of the debt.

It is hard for investors like you or me to assess the default risk but fortunately we have the ratings agencies which place an alpha-numeric score on most big borrowers which reflects the probability of them defaulting over a given period of time.

There are hundreds of ratings agencies, but bond markets broadly look at three of them, Moody’s, Standard & Poor’s, and Fitch. All three rate from triple-A to double-A to single-A to triple-B, to double-B and so on down to single-C at the very bottom after which comes D for default. Then in between each category comes in + and – as in AA+, AA and AA-, so there is some subtlety in the ratings which is refined even further by the addition of the rating being classed as stable or on credit watch for a possible future up- or downgrade. Although credit ratings are assigned in letters and numbers they are actually just an expression of an assumed probability of future default.

Moody’s estimates are that a Aaa rated credit carries a 0% chance of defaulting within 3 years which rises to 0.30% over 10 years. A single-A credit for the same period is assume to carry risks of 0.02% and 0.79%. A C-rated credit is presumed to carry a 19.9% risk of defaulting within a year and a 41.18% probability of turning up its toes.

That is without a doubt a proper junk bond but where does junk begin? Strictly, junk bonds are ones with a rating of below BBB- which are not treated as “investment grade” and therefore ineligible for discretionary investment or, in the trade, for widows and orphans. The correct term is “speculative grade” although that has in common parlance been replaced by “high yield” which sounds less toxic and more palatable to those to whom the bonds are being flogged.

So where does Moody’s shock-horror US downgrade of Sunday sit in the greater scheme? The answer is nowhere. Fitch and Standard & Poor’s have long had US sovereign debt rated in the AA+ category and Moody’s has had the AAA rating on watch for a downgrade since Adam was a boy. The statistical backdrop for each ratings cohort is indeed complex and Moody’s separates between unadjusted and withdrawal adjusted default probabilities but suffice to say that in the unadjusted category, AA rated credits are – brace yourself – less likely to default over a 10 year period than AAA rated credits, albeit by only 0.02%. Withdrawal adjusted the default probability increases by 0.02% from 0.56% to 0.58%. So why all the fuss?

I did get an email from a Frankfurt based reader that asked whether the downgrade might nor force some pension funds to sell. Institutional investors are generally held to applying the lowest of the three agencies’ ratings – some apply as a benchmark only two, some only one agency - and the USA has for a very long time no longer been an AAA/Aaa/AAA credit, so the Moody’s outlying AAA has not been that important.

And yet, the media went nuts and doom and gloom was predicted for all markets yesterday. Risk asset markets did indeed open lower, and the US Treasury market also opened lower which is with higher yields. It didn’t take long for equities to recover, the Dow closing the day up by 0.32%, the S&P500 by 0.09% and the Nasdaq by 0,02%. The 10 year US Treasury note had closed on Friday at around 4.43%. On the open is spiked up to 4.56% - that’s a lot in bond market terms – but by the close of business was right back where it had started.

So, will the cost of borrowing for the US government rise following the downgrade? In practice we will never know for the price of Treasuries is principally influenced by market expectations for future monetary policy which is in turn driven by the twin factors of the employment and inflation outlook and how the members of the Federal Reserve’s monetary policy committee see the future. On a comparative basis one might be able to measure the price of US Treasuries against other US dollar denominated sovereign or semi-sovereign debt but that again is influenced by the differential in the liquidity of Treasuries versus other dollar debt and to be frank there is no comparison whatsoever. In reality,99.9% of bond market price making is based on comparison, on relative value and on relative liquidity. It’s not a quiz but as in fantasy football it helps to know the competition.

We are of course largely dependent on what we garner from the media to know what’s going on and, bless, there are very few people writing on bond markets who truly understand how it all works. Of the few who do, and they truly are few, most are hamstrung by editors who don’t. Then there are the odd bods like myself with no editorial constraints who have the freedom to call a spade an effing shovel but know that as right as they may be it is the market that decides in which direction it’ll go.

In the end, the bond market will always be right although it might take it a while to get there. Money is made not only by being first to know where that will be but also supported by enough capital backing to be able to hold one’s breath until the rest have worked it out too.

In his book about the collapse of the subprime mortgage market “The Big Short”, Michael Lewis beautifully describes the struggle some of those who saw the crash coming went through in order to hold their short positions against a market that had yet to appreciate what a mess it had got itself into.

Those who could hang on made out like bandits and became heroes. How many got squeezed out before the brown stuff finally hit the propellor is not recorded but I can assure you that there were many. Moody’s downgrading the USA? Not really worth losing any sleep over. Buying bonds when the rest were naively selling? Priceless.

If you have got to here and have understood what I have just told you, go and reward yourself with a glass of champagne.