A couple of thoughts on the potential portfolio impact of a US downgrade

There has been some attention to the potential fallout of the downgrading of US Treasuries in terms of the follow-on impact on institutional investors that I wanted to expand on. In particular, a New York Times article from July 20^th included this paragraph (h/t Jane Hamsher).

One of the worst possibilities that people in the financial industry, like Mr. Lengsfield, have been discussing is that scores of insurance companies, pension funds and mutual funds might be forced to dump their Treasury holdings. Some investors have rules that they cannot hold assets that are rated below AAA. It was this sort of rule that drove the forced selling of mortgage bonds during the financial crisis.

An even more apt description from the Daily Beast (h/t Barry Ritholtz):

…. frankly, the power of their agencies is the product of endless buck passing: first on the part of the SEC at various points in the 20th century, then of everyone from bank regulators to insurance regulators, and in the institutional world, pension boards and brokers, all of whom wrote into their due-diligence contracts that only “investment grade” bonds could be held for certain purposes. Rather than force the fiduciaries of each of these to do their own due diligence, they were all given the out that if one of the designated ratings agencies certified the credit worthiness, that was enough.

These quotes seem to conflate two kinds of portfolios, those managed for total return and those which are (for lack of a better name) statutory portfolios. The impact on the former is likely to be less serious in light of a default than on the latter, I think.

On the less serious side, many mutual fund and other investment portfolios (particularly pension funds) are managed for total return according to formal – but internally chosen – guidelines that specify U.S. government debt as a separate category, not as part of a generic AAA bucket. As a result, a change in rating should theoretically have little impact for such investors, who also have the option (as the Times article described) of changing their guidelines. I used to write investment guidelines for pension funds many moons ago, and even way back then (late 1990s) they were moving away from proscribed exposures to specific credit tiers and toward a more commonsense approach of specifying X% in investment grade, etc., as noted in the Daily Beast quote above, which implies that (hopefully) there aren’t too many pension funds left that have AAA requirements.

The much bigger concern in terms of institutional money will be statutory portfolios. The largest of these in the US are insurance companies and major banks, whose exposures to specific credit tiers are monitored by regulators and (you guessed it) rating agencies. This may have been what the Times quote was referring to in the last sentence. If/when the US loses AAA, it will be a major concern to determine how best to figure out whether insurance company A’s portfolio still merits an AM Best rating of A+, or whether Global Bank B’s capital reserves still meet Basel Whatever Number We’re On Now.

There is another concern that I haven’t seen addressed explicitly, and that has to do with the nature of the downgrade. The U.S. itself is being downgraded as a sovereign entity, not simply as the issuer of Treasuries. That means (I think) that any security which carries a AAA rating backed by the full faith and credit of the US government is also at risk. If this expanded definition is true, then there is much more at stake than Treasuries.

It has been a while since I looked at the composition of the US bond market, so I looked up the sector weightings in the Barclays U.S. Aggregate Index, which is meant to capture the broad U.S. investment grade debt universe. They are:

Source: Barclays Capital fact sheet

The chart on the left shows the composition by sector of the U.S. investment grade bond market. The bottom tier in each bar is Treasuries, which account for around a third of the market, and which have been the primary focus of most commentary.

However, there are two other major categories in the market – Government-Related (mostly agencies) and MBS – which are predominantly made up of securities whose AAA ratings are based on that of the federal government now that Fannie and Freddie have been formally backstopped (since it’s the Aggregate, I think we can assume that the MBS segment does not include CDOs but rather the underlying bonds from which they were constructed). By implication, all such securities should logically be downgraded if/when the US is. After adjusting for underlying composition within these buckets, the chart on the right shows that “only” around 78% of the US investment grade bond market would face such a downgrade.

Here’s hoping none of this comes to pass, but it seems a possibility.