And some people want more regulation?
How Basel II allowed the banks to get it wrong
- Posted by Jon Stanford
The number of shocks to the banking system in advanced economies recently such as the nationalisation of Northern Rock Bank in England; the losses by a rogue trader from SocGen; the reconstruction of two smaller German banks after over-exposure to the subprime crisis; the increase in short term interest rates as banks are increasingly reluctant to lend to each other in money markets; the increase rates of interest on business and mortgage loans in excess of official central bank changes. The biggest one of all was the sale of Bear Sterns, a prominent New York investment bank and a leader in the sub-prime mortgage securitisation, to JP Morgans in a bargain basement sale officially supported by the Federal Reserve.
All this take together is indicative of a banking system which has misjudged risks, has mispriced risk, has been unable to assess the creditworthiness of other banks.
Is this just a case of banks having overextended themselves and made poor calls at critical points?. On the contrary I would argue that changes in international banking regulation brought about under the auspices of Basel II have allowed banks systematically to underestimate risk and loss allowances and to make inadequate provision of capital reserves to meet unexpected losses. Banks have been implicitly encouraged to place growth ambitions ahead of prudent management of their affairs.
There are two changes under Basel II which deserve consideration. The first is to allow banks to employ credit ratings to measure individual banksâ credit risk.
The second is to allow banks to use advanced statistical models to determine risk of default or failure in the event of external shocks.
Concerns about credit ratings are not new; they were previously raised during the Enron scandal of 2002 and the Asian Financial Crisis of 1997. But under Basel II banks have been allowed to use credit ratings to determine the risk attaching to their loans and hence to calculate minimum capital requirements. An example of how this worked is: banks sell mortgages off their balance sheet to a securitisation vehicle which is rated AAA. In consequence, the bank needs to keep minimal capital reserves against this. If the bank lends the securitisation vehicle the funds to complete the purchase, the bank needs to make minimal, if any, provisions for capital reserves. When the securitisation loses its AAA rating the bank is then required to make substantial provisions for capital reserves up to eight per cent of the loan. If the loan to the securitisation vehicle is considered doubtful, then the bank must make explicit provisions for the expected loss. What started out as a clever piece of financial engineering has put the bank in difficult circumstances. The position is made worse if the sale of mortgages was not a âclean saleâ i.e. the bank provided implicit undertakings to purchase loans back.
Also banks have been allowed to carry out their own risk assessment of other parts of their operation by using sophisticated statistical techniques which aim to calculate the risk of loss in the event of some external shock e.g. a rise in market interest rates, a shock to financial markets or a recession in the economy. As one might expect, banks used assumptions which cast their assessed risks in the best possible light. The general effect has been to allow banks to come up with models showing they face a lower risk which means they have to hold smaller amounts of capital. Banks like to hold less capital because it doesnât dilute control it increases leverage (able to use more borrowed funds) and hence profits when things are going well. However, when things are not going well, as they are in 2008, banks have to overcompensate by holding higher than the minimum capital reserves. Since they are unable to raise new capital on the open market (think how an issue of shares would be regarded when the price of existing bank shares have taken a severe tumble), banks have to cut back on new loans and try to increase their profit margins i.e. increase interest rates by more than official increases.
The process is contagious. Banks become unwilling to transact with other bank because each suspects that the others have been up to the same tricks as they have so that balance sheets lose transparency. The interbank market come to a halt and liquidity dries up especially as banks now need to hoard cash.
Is this analysis speculation? Well, the Presidentâs Working Group on Financial Markets on March 11, in a classic case of shutting the stable door after the horse has bolted, has made extensive recommendations for the reform of the credit rating industry, the use of statistical risk models and has called for certain of the Basel II to be sent back to the Bank for International Settlements for another look.
How Basel II allowed the banks to get it wrong
- Posted by Jon Stanford
The number of shocks to the banking system in advanced economies recently such as the nationalisation of Northern Rock Bank in England; the losses by a rogue trader from SocGen; the reconstruction of two smaller German banks after over-exposure to the subprime crisis; the increase in short term interest rates as banks are increasingly reluctant to lend to each other in money markets; the increase rates of interest on business and mortgage loans in excess of official central bank changes. The biggest one of all was the sale of Bear Sterns, a prominent New York investment bank and a leader in the sub-prime mortgage securitisation, to JP Morgans in a bargain basement sale officially supported by the Federal Reserve.
All this take together is indicative of a banking system which has misjudged risks, has mispriced risk, has been unable to assess the creditworthiness of other banks.
Is this just a case of banks having overextended themselves and made poor calls at critical points?. On the contrary I would argue that changes in international banking regulation brought about under the auspices of Basel II have allowed banks systematically to underestimate risk and loss allowances and to make inadequate provision of capital reserves to meet unexpected losses. Banks have been implicitly encouraged to place growth ambitions ahead of prudent management of their affairs.
There are two changes under Basel II which deserve consideration. The first is to allow banks to employ credit ratings to measure individual banksâ credit risk.
The second is to allow banks to use advanced statistical models to determine risk of default or failure in the event of external shocks.
Concerns about credit ratings are not new; they were previously raised during the Enron scandal of 2002 and the Asian Financial Crisis of 1997. But under Basel II banks have been allowed to use credit ratings to determine the risk attaching to their loans and hence to calculate minimum capital requirements. An example of how this worked is: banks sell mortgages off their balance sheet to a securitisation vehicle which is rated AAA. In consequence, the bank needs to keep minimal capital reserves against this. If the bank lends the securitisation vehicle the funds to complete the purchase, the bank needs to make minimal, if any, provisions for capital reserves. When the securitisation loses its AAA rating the bank is then required to make substantial provisions for capital reserves up to eight per cent of the loan. If the loan to the securitisation vehicle is considered doubtful, then the bank must make explicit provisions for the expected loss. What started out as a clever piece of financial engineering has put the bank in difficult circumstances. The position is made worse if the sale of mortgages was not a âclean saleâ i.e. the bank provided implicit undertakings to purchase loans back.
Also banks have been allowed to carry out their own risk assessment of other parts of their operation by using sophisticated statistical techniques which aim to calculate the risk of loss in the event of some external shock e.g. a rise in market interest rates, a shock to financial markets or a recession in the economy. As one might expect, banks used assumptions which cast their assessed risks in the best possible light. The general effect has been to allow banks to come up with models showing they face a lower risk which means they have to hold smaller amounts of capital. Banks like to hold less capital because it doesnât dilute control it increases leverage (able to use more borrowed funds) and hence profits when things are going well. However, when things are not going well, as they are in 2008, banks have to overcompensate by holding higher than the minimum capital reserves. Since they are unable to raise new capital on the open market (think how an issue of shares would be regarded when the price of existing bank shares have taken a severe tumble), banks have to cut back on new loans and try to increase their profit margins i.e. increase interest rates by more than official increases.
The process is contagious. Banks become unwilling to transact with other bank because each suspects that the others have been up to the same tricks as they have so that balance sheets lose transparency. The interbank market come to a halt and liquidity dries up especially as banks now need to hoard cash.
Is this analysis speculation? Well, the Presidentâs Working Group on Financial Markets on March 11, in a classic case of shutting the stable door after the horse has bolted, has made extensive recommendations for the reform of the credit rating industry, the use of statistical risk models and has called for certain of the Basel II to be sent back to the Bank for International Settlements for another look.