[caption id="attachment_38614" align="alignright" width="300" caption="Talking to ITV News about changes to financial regulation"]
[/caption]As I told ITV news this lunchtime, separating out retail and investment banking can be actively dangerous. To understand why, imagine you separated them completely into relatively safe retail banks and relatively risky investment banks. The retail banks would still be able to take risks. Banking is an inherently risky activity as you often lend for the long term and borrow for the short term. Research has even found that banks with investment banking divisions can be more reliable as they are more diversified. Northern Rock didn't have an investment banking arm, after all. Alex Tabarrok, one of the authors at the Marginal Revolution blog, sets out why that could be the case here:
Proponents of the Glass-Steagall Act argued that separating commercial and investment banking would increase the safety and reduce bank and customer conflicts of interest. Neither of these arguments bares close scrutiny here. At the most basic level, it is clear that many securities (stocks and bonds) are less risky than are loans. Security investments are also liquid and publicly observable. Liquidity lets banks quickly rebalance their portfolios to avoid runs, and public observability improves the efficiency of bank monitoring by depositors and bond holders. Even if all securities were riskier than all loans, forbidding banks to invest in securities could increase bank risk because of the benefits of diversification (see Macey 1991).
If a bank is told that it is a safe, retail bank which the Government can't possibly allow to fail then that will encourage them to take risks. We will be offering the retail banks a very explicit heads you win, tails we lose, bet.
It isn't clear yet how much risky activity will be contained within the retail part of the banks, as we don't know the Government's definition of retail and investment banking. But Robert Peston has written about one definition, and the extent to which major risks are still contained within that notional retail banking sector:
A recent suggestion by HSBC - which would base the break-up on a new accounting standard differentiating a bank's trading activities from traditional banking (in the jargon, those assets valued on an accrual basis would be in the ring-fenced retail bank, and those marked to market would be outside) - would see some 60% or 70% of a bank inside the ring-fence.
Now some bankers argue that HSBC's proposal would tend to keep taxpayers far too exposed to potential bank losses - since, in practice during the recent crash and recession, losses for banks on HSBC's definition of retail banking have been massively greater than investment banking losses (by a multiple). For example, HBOS's insanely loss-making loans to property businesses would have been inside the ring fence, on HSBC's definition.
It could be that we do need to carve out a safe space for retail depositors. It is far from clear that the Government is doing that though and half measures could be the worst possible outcome.
There are two missing ingredients so far. The first is ensuring that bondholders bear the risk when they lend to banks. The decision to give them a free pass, as we did in the bank bailout, creates huge moral hazard. If someone lends money to a company, and it goes bust, they should lose it. Otherwise there is no incentive for lenders to insist that executives at the banks are careful about the risks they take on, and be careful about only lending money to sound institutions. Andrew Lilico has led the charge on this arguing that bondholders should be bailed in, rather than banks being bailed out, and he wrote about that for ConservativeHome.
The other thing they need to do is look at global regulations that are increasing the systemic risk to the financial system. As that regulation becomes more and more homogenous, the result is a kind of monoculture which is far more vulnerable to disease. When something goes wrong everyone has the same vulnerabilities and tries to respond in the same way, becoming a huge danger to the world economy. There is a lot more on this in a short research paper we released with the Legatum Institute.
For all these shortcomings, there was some very good news at the speech. The Government is going to auction Northern Rock as soon as possible. We need to get our money back, and we need to get the Government out of the business of investing in banks.
[iframe http://www.youtube.com/embed/LzimYGoBEiM 500 314]The reforms set to be announced by George Osborne in his Mansion House speech tonight, aiming to ring-fence retail banking, are an incomplete response to the financial crisis, and if the Government aren't careful they could be actively unhelpful. While the details are very complex, there are some simple issues that taxpayers should be aware of in judging if reforms are going to make it more or less likely that bankers come begging to them for a handout again.
[caption id="attachment_38614" align="alignright" width="300" caption="Talking to ITV News about changes to financial regulation"]
[/caption]As I told ITV news this lunchtime, separating out retail and investment banking can be actively dangerous. To understand why, imagine you separated them completely into relatively safe retail banks and relatively risky investment banks. The retail banks would still be able to take risks. Banking is an inherently risky activity as you often lend for the long term and borrow for the short term. Research has even found that banks with investment banking divisions can be more reliable as they are more diversified. Northern Rock didn't have an investment banking arm, after all. Alex Tabarrok, one of the authors at the Marginal Revolution blog, sets out why that could be the case here:
Proponents of the Glass-Steagall Act argued that separating commercial and investment banking would increase the safety and reduce bank and customer conflicts of interest. Neither of these arguments bares close scrutiny here. At the most basic level, it is clear that many securities (stocks and bonds) are less risky than are loans. Security investments are also liquid and publicly observable. Liquidity lets banks quickly rebalance their portfolios to avoid runs, and public observability improves the efficiency of bank monitoring by depositors and bond holders. Even if all securities were riskier than all loans, forbidding banks to invest in securities could increase bank risk because of the benefits of diversification (see Macey 1991).
If a bank is told that it is a safe, retail bank which the Government can't possibly allow to fail then that will encourage them to take risks. We will be offering the retail banks a very explicit heads you win, tails we lose, bet.
It isn't clear yet how much risky activity will be contained within the retail part of the banks, as we don't know the Government's definition of retail and investment banking. But Robert Peston has written about one definition, and the extent to which major risks are still contained within that notional retail banking sector:
A recent suggestion by HSBC - which would base the break-up on a new accounting standard differentiating a bank's trading activities from traditional banking (in the jargon, those assets valued on an accrual basis would be in the ring-fenced retail bank, and those marked to market would be outside) - would see some 60% or 70% of a bank inside the ring-fence.
Now some bankers argue that HSBC's proposal would tend to keep taxpayers far too exposed to potential bank losses - since, in practice during the recent crash and recession, losses for banks on HSBC's definition of retail banking have been massively greater than investment banking losses (by a multiple). For example, HBOS's insanely loss-making loans to property businesses would have been inside the ring fence, on HSBC's definition.
It could be that we do need to carve out a safe space for retail depositors. It is far from clear that the Government is doing that though and half measures could be the worst possible outcome.
There are two missing ingredients so far. The first is ensuring that bondholders bear the risk when they lend to banks. The decision to give them a free pass, as we did in the bank bailout, creates huge moral hazard. If someone lends money to a company, and it goes bust, they should lose it. Otherwise there is no incentive for lenders to insist that executives at the banks are careful about the risks they take on, and be careful about only lending money to sound institutions. Andrew Lilico has led the charge on this arguing that bondholders should be bailed in, rather than banks being bailed out, and he wrote about that for ConservativeHome.
The other thing they need to do is look at global regulations that are increasing the systemic risk to the financial system. As that regulation becomes more and more homogenous, the result is a kind of monoculture which is far more vulnerable to disease. When something goes wrong everyone has the same vulnerabilities and tries to respond in the same way, becoming a huge danger to the world economy. There is a lot more on this in a short research paper we released with the Legatum Institute.
For all these shortcomings, there was some very good news at the speech. The Government is going to auction Northern Rock as soon as possible. We need to get our money back, and we need to get the Government out of the business of investing in banks.
[iframe http://www.youtube.com/embed/LzimYGoBEiM 500 314]