Financial stability on track, or on a detour?

That the Reserve Bank devoted five pages of yesterday’s Financial Stability Review to the ABX.HE credit default swap indices was a statement in itself – these things are important.

In fact, they are at the heart of the credit crisis and deserve at least five pages of any review of financial stability.

The combination of mark-to-market accounting, introduced by the Securities and Exchange Commission in 1997, and the ABX.HE indices, which were launched by a small research firm named Markit Group about three years ago, has been absolutely explosive.

Bank and investment bank sub-prime losses have all resulted from using these indices to value their sub-prime mortgage books.

The volatility and repricing of risk in credit markets has, to a large extent, been an unforeseen consequence of what seemed like, and were, two good ideas at the time; more transparent accounting rules and the introduction of indices that would allow derivative trading, and therefore hedging and risk mitigation, in the mortgage-backed securities market.

But in yesterday’s Financial Stability Review from the Reserve Bank of Australia there is this remarkable conclusion: “There is… a growing concern that the prices of the ABX.HE indices, particularly the ‘AAA’ sub-indices, may be giving an unrealistic signal of the losses likely to be sustained on the underlying RMBS, which is prompting some to question the use of these indices in valuation models.

“According to one estimate, the recent prices of the ABX.HE indices imply cumulative losses of about one third on the constituent RMBS. One way this could occur would be if two thirds of mortgage holders defaulted, and the average recovery rate was only half of the mortgage values.”

This, to say the least, implies a very bad housing downturn – the worst ever. It almost certainly implies a worse depression than occurred in the 1930s. So have banks been valuing their portfolios – and reporting losses – based on realistic pricing? You decide.

At the very least an examination of the ABX.HE indices, and how they might be improved, should run alongside the other reviews that are aimed at dealing with the crisis.

For example, US legislators, including former Wall Street banker and now Treasury Secretary Henry Paulson, are pushing for Federal Reserve regulation of investment banks following the bail-out of Bear Stearns, and there is plenty of talk about the need for governments to now buy mortgage backed securities (MBS) or guarantee them in order to restart some kind of securitised home lending in competition with banks.

But in its Financial Stability Review yesterday, the RBA is clearly suggesting that the whole basis for measuring the extent of the downturn in MBS is flawed.

Specifically, the ABX.HE indices capture just 20 MBS, compared to the 50-100 that are typically used in the MBS collateralised debt obligations (CDOs) that have been traded by the banks and investment banks in recent years, and which are now causing all the write-downs.

This, says the RBA, raises questions about whether the pricing is “prone to distortion”.

Meanwhile “mark-to-market accounting” was introduced in financial accounting standards no.157 in 1997, and while there is some debate about what the standard actually requires, it is undoubtedly responsible for the loan write-offs that have so far been announced.

Looking back at the genesis of the crisis, it is hard to determine which is the chicken and which the egg, and certainly a vicious cycle got going; two Bear Stearns sub-prime funds marked their books to market in July and defaulted on loan covenants (they were obviously sailing closest to the wind), and from there the MBS markets progressively shut down. Every time the ABX indices declined, more losses were announced which in turn caused further declines in the indices and more losses.

The basic problem is that mark-to-market accounting rules applied to fixed income securities makes them behave to some extent like equities – that is, making the pricing volatile.

Shares have traditionally carried an “equity risk premium”, but given the fact that they always perform better than bonds over the long term, there is no logic to that premium except to compensate for the psychological impact of short-term volatility.

If fixed-interest pricing becomes as volatile as equities, then a premium will have to apply to them as well (and arguably should have already, given what has happened).

Alternatively, and more likely, fixed-interest securities backed by mortgages and priced according to derivative indices will be treated as equities.

So in some ways, the credit crisis that we are now going through might be explained as the painful transition of mortgage-backed securities from being fixed-interest to being equities.

One thing’s for sure – there can be no going back to less transparent accounting. Mark-to-market is, and should be, here to stay. It’s not the “mark” part of the phrase that’s the problem – it’s the “market”.


This story first appeared on BusinessSpectator