For many years I have lobbied and asked for more detail on the arrears position by sector, by product and by lender type to no avail, but the decision by the FSA could make this a reality in the coming months.
The headline figures, although useful as benchmarks, hide a wide variety of performance levels across sector, region, property, lender type and even vintage of loan.
It has always puzzled me how when two lenders have the same average level of arrears across their portfolios, this appears to make the portfolios of equal risk. In practice, diversity of performance across portfolios is as important as the current average level of arrears.
Interestingly, the arrears averages of many sub-prime portfolios bear a striking resemblance to more traditional prime lending books during the boom. The recent flight to quality lending owes more to gut reaction than scientific investigation as industry benchmarks and histories are difficult to find.
Data for the first six months of this year shows that overall arrears remain low, with three-month plus arrears at 1.3% of mortgage balances, only a fifth of the levels seen in 1992. The figures within the Basle II definition of default (six-month plus) are only at 0.6%.
But when one looks at the actual number of mortgages in arrears the numbers become a little more difficult to swallow. In 2008, we estimate that around 156,000 customers will be three or more months in arrears and this will rise sharply in 2009 but to a level well below the 650,000 we experienced in the early 1990s.
These headline numbers conceal a range of performances. Table 1 below shows arrears by sector, with prime lending remaining dominant.
However, the non-conforming sector has seen short-term arrears rise to more than 11% (and more than 30% in US) compared with around 2.5% in the prime sector.
This has been accompanied by a large rise in the probability of default which has reached 2% and is heading higher. This has become more serious as the potential losses for these cases have risen sharply on the back of lower house prices, increased mortgage margins and increased time to sell.
It should come as no surprise that this difference in short-term arrears has translated into much higher levels of repossessions for a given average level of arrears as some lenders have adopted more aggressive business strategies. The distribution of arrears has had a significant impact on the outlook for total repossessions.
We expect repossessions to rise sharply in 2009 without any intervention from the government, reaching over 60,000 compared with a little over 26,000 in 2007.
The latest data from the FSA also highlights that building societies and banks are far less affected than specialist lenders. Arrears among securitised loans are nearly double the level for the whole mortgage book. Although this reflects in part newer loans with higher LTV ratios, it also should have a big bearing on the required capital lenders should put behind such loans.
As the mortgage market has become more complicated and fragmented the need for more detailed market data and insight has increased.
Relying on peer group lenders for reassurance about absolute risk is folly. Maintaining your relative position may look attractive in the short term but the industry can easily get into a systemic problem by not taking a detailed view of the problems and risks.
We need a more mature and transparent approach to risk in the mortgage market, with lenders adopting more cautious lending practices in the upturn and more flexible and appropriate ad-justments in the downturn.
Basle II risk information (see Table 2) seems to have had a big effect on lenders’ attitudes in the downturn but played little or no role in the decisions made in the upturn. The devil is in the detail – and much of that detail was sadly absent when it mattered.