From: Jurriaan Bendien (adsl675281@TISCALI.NL)
Date: Sun Nov 18 2007 - 11:07:05 EST
As far as punditry goes, I think it is exaggerated. As far as I know, there's about $10 trillion of US mortgages in total, of which nearly 14% subprime. Of those 14% a third is subject to variable interest rates, which is just under 5% of the total mortgages. There were a lot of foreclosures already in previous years, as Bernanke pointed out, but now the number of foreclosures is roughly doubling, that's the first problem. In 2007 about 40% of the mortgagers reached their higher interest rate, and in 2008 another 40% or so faces the same. The Dutch Reserve Bank estimates the aggregate losses in the US at $200 billion max, Deutsche Bank $400 billion max, and the Royal Bank of Scotland $500 billion max, presumably spread over several years. Meantime, something like $60 billion has actually been written off. Obviously the mortgagers may have some equity in the foreclosed property, and the homes themselves have a market value, but for all the financial claims additionally staked on the original loans there is no collateral, because they hinge on a future earning power which has disappeared. Neighbourhoods with the highest rates of subprime lending have black and Hispanic workingclass majorities, while the areas with the lowest rates are mainly white middleclass, and it's mainly the black and Hispanic areas in which the defaults are occurring. When the Federal Reserve from 2004 required lenders to provide specific data on subprime loans, the finance industry fought successfully to keep the risk profile of borrowers, including credit scores, secret. Many subprime loans were in fact not even intended to be affordable over the long term, but rather, to be refinanced, before their initial "teaser" rates rose, on the basis that the market value of the houses would have increased by then. The lenders had no particular stake in extending loans which they knew would be defaulted on. But if market prices for houses drop, and credit facilities tighten, the borrowers are stuck with loans they cannot repay. The real problem is (1) that for roughly half the "bad" loans it is not even clear who owns the contracts staked on them, and because nobody wants to be associated with bad lending practice, reliable information is hard to get and (2) the problem triggers all sorts of financial fears, which leads to a credit squeeze and to investors withdrawing funds all over the place, because of loss of confidence in the markets. (3) it combines with a number of other circumstances, such as the declining exchange rate of the US dollar, rising oil prices, reduced inflow of foreign capital, inflationary pressures, and a fall-off in consumer spending. Credit is a marvellously flexible instrument which allows one to displace the costs of current activity constantly in space and time. But all depends on the capacity of borrowers to repay, and the compulsion for them to repay. If they cannot repay, and they are not compelled to repay, the credit system starts to squeeze. Lenders become more conservative, at the very least. When they talk about "healthy economic fundamentals" they just mean that people can continue to pay sufficiently. Although there is a lot of "slack" in the capitalist system (idle funds), the overall effect of a more conservative stance must be declining output growth, on this all are agreed. By itself, I think the financial magnitudes involved in the subprimes are not sufficient to cause a recession by themselves, but in combination with other, more important factors, it must produce another downturn. EU real GDP growth next year is likely to be around 2.1% or so, and it would not surprise me if US GDP growth fell below 3%. But as Fred Moseley commented, nobody really knows the total effects of the events. A philosopher said to me once, "America likes to gamble with the future of the world, inviting all to take some more risk". Well, let's see how well America can gamble. PS - gambling is not a democratic sport. Jurriaan
This archive was generated by hypermail 2.1.5 : Fri Nov 30 2007 - 00:00:03 EST