Thanks For The Tip

Making the rounds a bit amongst the blog shamans this week is the headline (not the actual contents, God forbid) of an article from Expansión. Below is a paragraph-by-paragraph précis/translation.

Headline: One in five mortgages has a high risk of default.

Paragraph 1: Banks set themselves up to receive the next wave of defaults which will come from families as summer employment and government stimulus measures come to an end.

P2: It is expected that the unemployment rate will resume its upward trajectory, undermining the solvency of households in the process. Despite this, the banking industry believes it has improved its systems for handling defaults and will be able to confront this without seeing its default ratios soar.

P3: There are 670 billion euros of property-related loans outstanding in Spain. The default rate is currently 2.88%. The latest rate of increase in this figure is 0.9% - compared to the 8.1% of a year ago. Problems are expected from loans with the highest LTV and those whose service requires the highest percentage of family incomes.

P4: The LTV should not exceed 80%. The prudent limit for payment-to-income (depending on the method of calculation) is 35 to 40%.

P5: It is difficult to come by reliable data with which to calculate the curent state of affairs. The Banco de España reports that the percentage of transactions involving an LTV of greater than 80% has fallen from 17.6%, in 2006, to 11% in 2008.

P6: Prior to 2006, it is difficult to find reliable figures. A study by Genworth Financial (ed: a provider of mortgage insurance) notes that LTV's had increased from 1995 to 2005 along with the increase in the price of housing with respect to incomes.

P7: Mortgages with an LTV of greater than 80% constituted 12.2% of the market in 1996, and 26.4% in 2005. As of the second quarter of 2009, the average percentage of income required to service Spanish mortgages stood at 38.6%, without considering tax treatment.

P8: These loans may also be backed by guarantees other than the property itself.

P9: This figure has dropped from the 46% registered in 2006-2008, owing to the drop in prices and interest rates.

P10: The payment-to-income ratio was between 33 and 37% in the 1990's. On average, 9% of mortgages granted between 1995 and 2005 required more than 50% of family income to service.

P11: The figures emitted by individual lending institutions show that high risk loans, despite the increases registered, do not weigh heavily on their books. In the case of Santander*, at the end of 2008, 8% of mortgages showed an LTV of between 80 and 90%, and 7% of greater than 90%. LTVs of greater than 80 comprise 11% of La Caixa's book, 6% of Bankinter's and 5% of Banco Pastor's.

P12: These figures represent the reality at the time the loans were issued and the situation will have changed over time due to factors such as: the amount of debt paid down, the price of housing and the increase in the unemployment rate. Further, on occasion (today almost always) these loans may be guaranteed by co-signers or insurance policies.

More astute readers will immediately note that it is, in fact, difficult to draw any conclusions from the mish-mash of data presented. But, that there is not one single mention of the headline's conclusion in the body of the text (let alone anything to prove that '20%' of mortages are 'high risk' - definition of this forthcoming, we suppose) should command the reader's attention - particularly if that person proposes to be an interface between an ignorant and fearful public and ugly reality. Bloggers, by any other name.

*People of the punting persuasion will, on the other hand, find a bit of information value in the article's treatment of Banco Santander. The reporter has very delicately split this bank's exposure to high LTV mortgages into deciles, as opposed to the crude and disprespectful quintiles doled out to the other examples. Our trusty calculator reveals that a peaches-to-peaches comparison awards Sr. Botín's bank a 15% exposure to plus-80 LTV loans.

-----------------------------