Thinking twice about a possible Treasury downgrade

I was probably too cavalier this morning in my dismissal of the risks of a Treasury downgrade. In discussions over lunch today with my most excellent former colleagues, Steve and Ken, I came to appreciate the deep concerns that hover over the institutional bond market community. My point this morning was that a downgrade of US Treasury debt is essentially a downgrade of all debt—since Treasuries are the bedrock upon which all debt is priced—so a downgrade doesn't really mean much.

Their point, however, is that a downgrade of Treasury debt has huge and little-understood implications for many large institutional bond portfolios. To understand why, consider the case of a billion dollar bond portfolio that is currently underweight Treasuries, overweight MBS and corporate bonds, and skewed to lower-quality debt given the steepness of the credit curve. Given the steepness of the yield curve and the still-generous level of corporate spreads, overweighting MBS and corporates has been a very profitable strategy in recent years and promises to continue to be so. Moreover, such a strategy recognizes that Treasuries are very fully valued (and quite possibly overvalued), and thus minimizes a portfolio's exposure to rising Treasury yields. If a manager is even modestly optimistic on the economy's prospects, and if he believes that the Fed's accommodative monetary policy stance should, by making liquidity relatively abundant, result in relatively low default rates and facilitate economic growth, then positioning this portfolio at the low end of his client's acceptable credit quality range makes a lot of sense. (And isn't the Fed trying to encourage people to take on more risk?) In short, there are many good reasons for large, diversified, institutional portfolios to be structured today with a relatively low average credit quality.

But here's the catch: If Treasuries are downgraded, then a portfolio's average credit quality, assuming it holds at least some Treasuries, will fall. Even if the portfolio holds MBS and no Treasuries, one could argue that a downgrade of Treasuries perforce implies a downgrade of current-issue MBS, since their AAA rating depends crucially on the assumption that the implied Treasury guarantee of principal is bullet-proof. In short, if you downgrade Treasuries you downgrade just about everything, as I was arguing this morning.

But if you downgrade Treasuries and MBS, you need to understand that the average credit quality of our diversified, billion dollar portfolio will fall, and meaningfully. If a portfolio is currently at the low end of its acceptable average quality, a downgrade of Treasuries and MBS could quickly put it in violation of its credit quality guidelines. The solution to this problem illustrates the quandary that the bond market is very concerned about: The only way to bring the average quality of our billion dollar portfolio portfolio back up to its minimum required level, if Treasuries are downgraded, is to sell the cheapest bonds (e.g., the low-quality corporates) and buy the most expensive bonds (e.g., Treasuries). How would you like to call your billion dollar client and explain to him that as a result of the downgrade of Treasury bonds, of which you held very little or none because you think they are overvalued, you now have to buy more? And that in order to buy more Treasury bonds you have to sell the bonds that yesterday you thought were the cheapest and most attractive?

When this example is multiplied over the hundreds of billions and even trillions of dollars of diversified bond portfolios with strict quality guidelines, you suddenly realize that the bond market might theoretically be forced to dump massive quantities of low-quality bonds (thus raising the borrowing costs of hosts of struggling companies) in order to buy equally massive quantities of Treasury bonds (thus keeping Treasury yields very low), should Treasuries be downgraded. To drive home the absurdity of this, consider that the worse the deficit situation of the US becomes, the more our billion dollar bond portfolio would have to invest in Treasuries. This may sound surprising and even counterintuitive to bureaucrats, rating agencies, and the layman, but it would be eminently in keeping with The Law of Unintended Consequences, which is always lurking in the background, waiting to make fools of those who live by rigid rules and not by logic.

Memo to the bond market: it's past time to make quality guidelines more flexible.