06 September 2009
Bankers pay at G20
The FT Lex column (on ft.com) had an article about the G20s efforts to develop international financial regulation. This can be viewed at http://www.ft.com/cms/s/3/e0c5d822-9890-11de-807a-00144feabdc0.html 9Subscription may be required).
In response I wrote to the editor of the FT (published 8th Dec 2009, at least on ft.com) as follows:
Dear Sir,
The Lex column is quite wrong to describe the G20's focus on bankers' pay as "populist but often tangential". Bankers' behaviour is driven by pay, and so bankers' pay has to be absolutely central to regulation of behaviour in the financial sector. It is essential that the incentives that arise from pay lead to constructive, value creating behaviours and not to behaviours that undermine the system.
As Lex points out that, "the architects of boomtime credit innovations are returning to their desks, finding new ways to tinker with balance sheets and carve through rules that are still being developed. The regulated are already moving ahead of their minders." Such behaviour could all too easily send us back into crisis, and yet the behaviour arises because of the incentives in bankers pay.
Removing incentives for destructive behaviour from bankers pay is an essential first step before any other new regulation has a chance of succeeding.
Yours faithfully,
The Revd. Patrick Gerard
In response I wrote to the editor of the FT (published 8th Dec 2009, at least on ft.com) as follows:
Dear Sir,
The Lex column is quite wrong to describe the G20's focus on bankers' pay as "populist but often tangential". Bankers' behaviour is driven by pay, and so bankers' pay has to be absolutely central to regulation of behaviour in the financial sector. It is essential that the incentives that arise from pay lead to constructive, value creating behaviours and not to behaviours that undermine the system.
As Lex points out that, "the architects of boomtime credit innovations are returning to their desks, finding new ways to tinker with balance sheets and carve through rules that are still being developed. The regulated are already moving ahead of their minders." Such behaviour could all too easily send us back into crisis, and yet the behaviour arises because of the incentives in bankers pay.
Removing incentives for destructive behaviour from bankers pay is an essential first step before any other new regulation has a chance of succeeding.
Yours faithfully,
The Revd. Patrick Gerard
Labels: Bankers, G20, incentives, regulation
03 September 2009
In defense of the Tobin Tax
Willem Buiter, Professor of European Political Economy, London School of Economics and Political Science, often writes in the FT and has a blog on FT.com. His article in the FT, 1/9/09 about Tobin Tax caused me to write the below response.
The original article can be seen at http://blogs.ft.com/maverecon/ for 2nd Sept 2009 (subscription may be required).
I don’t think that you can argue that the financial sector is too large because government effectively subsidises its cost of capital by providing guarantees. This form of governmental help is less than a year old, but the problem of the oversized financial sector had developed well before that time. In fact it was the oversized “too big to fail” aspect of the financial sector, which effectively meant that government had no choice but to provide the guarantees.
It seems to me that your analysis of the problems is absolutely correct. You mention excessive churning, incentives that drive traders to make transactions, too much financial activity that is not just socially worthless but actually harmful, and too much speculation and not enough insurance. You imply that regulation should be used to directly restrict the undesirable features of contracts.
But how in practice could regulators do this? How could they keep pace with market innovation? How could they be sure that each regulation added does not create some new perverse incentive?
The first step for regulators must be to distinguish socially helpful financial transactions from unhelpful ones. It seems to me that there is no clear cut test for this. However as a general rule of thumb the nearer a transaction is to the real requirements of the real economy the more likely it is to be socially helpful. If a business needs to buy a currency in order to pay for a particular import, then this is a real requirement for a currency transaction. If a foreign currency is required in six months time for an import in six months time that must be accurately costed in the home currency now, then this is a real requirement for a currency futures transaction. Such transactions, driven by real requirements, create real value in the real economy and so are socially helpful.
In contrast a financial transaction that represents a nil sum game between the participants is much more likely to be socially problematic. When a trader takes a long or short position against another trader such that one will win money and one will lose money on the transaction then this is a nil sum game which adds no real value. A small number of such transactions are useful because they provide liquidity and facilitate the efficient spreading of risk. However a large number of such transactions actively destroy value because the transaction costs are high (traders are well paid) and risks inevitably flow towards places where they are hidden or not properly understood.
In real life it would be almost impossible for regulators to distinguish socially helpful transactions from unhelpful ones. Any attempt to do this would create an unhelpful incentive to disguise transactions to make them look socially helpful. However a Tobin tax does have a real chance of making the correct distinction. Basically a transaction that is driven by a real requirement in the real economy can usually afford to pay a small Tobin tax. In contract a nil sum game transaction cannot, because it becomes a negative sum gain after the tax has been deducted.
The question you quite rightly ask is “What distortion is a tax on financial transactions targeted at?” The answer is that we have far too many nil sum game transactions, and a Tobin tax targets these because it makes them economically unattractive.
The original article can be seen at http://blogs.ft.com/maverecon/ for 2nd Sept 2009 (subscription may be required).
I don’t think that you can argue that the financial sector is too large because government effectively subsidises its cost of capital by providing guarantees. This form of governmental help is less than a year old, but the problem of the oversized financial sector had developed well before that time. In fact it was the oversized “too big to fail” aspect of the financial sector, which effectively meant that government had no choice but to provide the guarantees.
It seems to me that your analysis of the problems is absolutely correct. You mention excessive churning, incentives that drive traders to make transactions, too much financial activity that is not just socially worthless but actually harmful, and too much speculation and not enough insurance. You imply that regulation should be used to directly restrict the undesirable features of contracts.
But how in practice could regulators do this? How could they keep pace with market innovation? How could they be sure that each regulation added does not create some new perverse incentive?
The first step for regulators must be to distinguish socially helpful financial transactions from unhelpful ones. It seems to me that there is no clear cut test for this. However as a general rule of thumb the nearer a transaction is to the real requirements of the real economy the more likely it is to be socially helpful. If a business needs to buy a currency in order to pay for a particular import, then this is a real requirement for a currency transaction. If a foreign currency is required in six months time for an import in six months time that must be accurately costed in the home currency now, then this is a real requirement for a currency futures transaction. Such transactions, driven by real requirements, create real value in the real economy and so are socially helpful.
In contrast a financial transaction that represents a nil sum game between the participants is much more likely to be socially problematic. When a trader takes a long or short position against another trader such that one will win money and one will lose money on the transaction then this is a nil sum game which adds no real value. A small number of such transactions are useful because they provide liquidity and facilitate the efficient spreading of risk. However a large number of such transactions actively destroy value because the transaction costs are high (traders are well paid) and risks inevitably flow towards places where they are hidden or not properly understood.
In real life it would be almost impossible for regulators to distinguish socially helpful transactions from unhelpful ones. Any attempt to do this would create an unhelpful incentive to disguise transactions to make them look socially helpful. However a Tobin tax does have a real chance of making the correct distinction. Basically a transaction that is driven by a real requirement in the real economy can usually afford to pay a small Tobin tax. In contract a nil sum game transaction cannot, because it becomes a negative sum gain after the tax has been deducted.
The question you quite rightly ask is “What distortion is a tax on financial transactions targeted at?” The answer is that we have far too many nil sum game transactions, and a Tobin tax targets these because it makes them economically unattractive.
Labels: regulation, Tobin Tax, trading
18 October 2008
Regulation of Bankers Pay
On Wednesday 15th October the Financial Times published a very good article by Jamie Whyte about the diffiulties of regulating bankers pay. The article can be read (by subsribers?) at http://www.ft.com/cms/s/0/62601d32-9a51-11dd-bfe2-000077b07658.html . The article discusses the principle-agent problem (how does an owner get a manager to work for the owner's interests and not his own) and advocates inovation in devising new arrangements for performance related pay.
I responded by writing to the FT letters column. My letter was not published but is included below.
Dear Sir,
I appreciated Jamie Whyte’s excellent analysis in “Why regulating bankers’ pay is still a bad idea” (FT 15/10/08) even if his conclusions are not quite right.
Mr Whyte points out that if a business owner wants to get good performance from a greed free manager then the owner must rely on the manager’s desire to what is best for the owner. Mr Whyte then suggests that it is over optimistic for an owner to assume that he has found such a manager. Clearly to “assume” this is over optimistic, so the real challenge is to find ways of building trust between the owner and the manager such that, over time, the owner comes to know that the manager really is working for the owner’s best interests.
Trust and fiduciary duty are fundamental to success of capitalism because they are the only satisfactory solution to the “principle-agent problem”. It is hard work to sustain trust and fiduciary duty and as concepts they might be profoundly unfashionable, but we shall not escape the financial crisis until they have been re-established.
Mr Whyte prefers the alternative which is to devise remuneration schemes that align the interests of the managers with the interests of owner. In adopting this approach the owner is seeking to harness the managers’ greed to his own advantage. This drives the principle-agent relationship towards mutual exploitation and away from trust. The approach breaks down because the managers have a strong incentive (which remuneration consultants collude with) to move remuneration practice along to make it easier for managers to secure higher rewards. The moving along of remuneration practice is often presented as “innovation”, but the innovations that are easiest to agree and get implemented are the ones that work best for the managers.
Under the incentive model, owners should insist on stable long term incentives that align the managers interests with their own. One reason why they fail to do this is because owners are themselves really managers (fund managers) who are themselves seeking higher rewards from principles, so they find it convenient to collude. In reality remuneration schemes like the FILLIP, which are really serious about aligning owner and manager interests, are of little interest in the market place of remuneration ideas.
Mr Whyte’s criticisms of regulation have some validity, but regulation that imposed and kept stable real long term incentive alignment between management and owners might well be the lesser of many evils.
The real solution however is to build trust with talented and hard working managers who are willing to work for the ownership interest. How does the manager build trust? Well accepting a flat salary with no extras of, say, US$500,000 would be a very convincing start.
Yours faithfully,
Revd Patrick Gerard
I responded by writing to the FT letters column. My letter was not published but is included below.
Dear Sir,
I appreciated Jamie Whyte’s excellent analysis in “Why regulating bankers’ pay is still a bad idea” (FT 15/10/08) even if his conclusions are not quite right.
Mr Whyte points out that if a business owner wants to get good performance from a greed free manager then the owner must rely on the manager’s desire to what is best for the owner. Mr Whyte then suggests that it is over optimistic for an owner to assume that he has found such a manager. Clearly to “assume” this is over optimistic, so the real challenge is to find ways of building trust between the owner and the manager such that, over time, the owner comes to know that the manager really is working for the owner’s best interests.
Trust and fiduciary duty are fundamental to success of capitalism because they are the only satisfactory solution to the “principle-agent problem”. It is hard work to sustain trust and fiduciary duty and as concepts they might be profoundly unfashionable, but we shall not escape the financial crisis until they have been re-established.
Mr Whyte prefers the alternative which is to devise remuneration schemes that align the interests of the managers with the interests of owner. In adopting this approach the owner is seeking to harness the managers’ greed to his own advantage. This drives the principle-agent relationship towards mutual exploitation and away from trust. The approach breaks down because the managers have a strong incentive (which remuneration consultants collude with) to move remuneration practice along to make it easier for managers to secure higher rewards. The moving along of remuneration practice is often presented as “innovation”, but the innovations that are easiest to agree and get implemented are the ones that work best for the managers.
Under the incentive model, owners should insist on stable long term incentives that align the managers interests with their own. One reason why they fail to do this is because owners are themselves really managers (fund managers) who are themselves seeking higher rewards from principles, so they find it convenient to collude. In reality remuneration schemes like the FILLIP, which are really serious about aligning owner and manager interests, are of little interest in the market place of remuneration ideas.
Mr Whyte’s criticisms of regulation have some validity, but regulation that imposed and kept stable real long term incentive alignment between management and owners might well be the lesser of many evils.
The real solution however is to build trust with talented and hard working managers who are willing to work for the ownership interest. How does the manager build trust? Well accepting a flat salary with no extras of, say, US$500,000 would be a very convincing start.
Yours faithfully,
Revd Patrick Gerard
Labels: alignment, banking, Fiduciary Duty, incentive, Jamie Whyte, regulation, trust
05 October 2008
The New World for Banking
On 4th October 2008 a poll by ft.com asked the question, "Will the rescue plan work?". This was a reference to the $700bn rescue plan for the financial services industry finally approved by the US Congress earlier that day. I posted in the following comment, which can be viewed at
http://www.ft.com/cms/6c2bf1ce-91b7-11da-bab9-0000779e2340.html?a=tpc&s=646099322&f=851094803&m=9531017771&r=9531017771 .
The $700bn rescue plan can, at best, only help in the short term. Unfortunately the underlying long term attitudes which caused the credit crunch are still very firmly in place.
We should certainly hope that the $700bn might last long enough to allow development of new and tighter regulation for banks that is internationally agreed. Sadly it is hard to find solid grounds for this hope, because the regulation that banks really need is likely to work against the instincts and vested interests of most of the people involved in the discussions.
Banking need to be much, much simpler so that financial markets are more readily understandable. Complex derivatives proved to be much more effective at hiding risk that at managing it efficiently. Banks need to be much smaller so that failures are more manageable. Banks need to be less leveraged so they are safer and have more of a utility feel. The capital banks have available for speculation must be linked to market making obligations. Speculative capital should be strictly limited so that huge movements in short term capital cannot cause market lurches.
The incentives that drive banking behaviours need far more attention. The incentives that arise from holding a "long" equity position are much more constructive to the economy as a whole that the incentives which arise from holding a "short" position. The incentives that arise from pay need careful consideration. Top bankers should not be eligible for annual bonuses; all incentives should be on a long term basis. The top mangers in a bank must all have common incentives so that they work together, share information and form a common mind on the banks position and the state of the market. Above all the pay of top bankers must be much lower so that shareholders can feel confident that the top bankers are working for the shareholders' benefit not their own benefit. Very high pay increases the likelihood of ruthless and self-seeking characters at the top of the organisation; if you pay gold you get pirates!
In summary banking needs to become much, much more boring! Banking careers should appeal to steady and consistent people. Just like top athletes, top bankers should be subjected to regular drugs tests. The supercharged performance currently expected is not human and its puts inhuman pressures on other parts of the system.
Unfortunately we are still a very long way from a safe and boring banking system. And in the meantime what are the whiz-kids doing? Well I expect that the big prizes right now are for finding the best schemes to persuade government to take over and pay too much for the very worst assets. All this overcharged pursuit of money has killed many of our financial institutions. Are we going to allow it to kill our public finances?
http://www.ft.com/cms/6c2bf1ce-91b7-11da-bab9-0000779e2340.html?a=tpc&s=646099322&f=851094803&m=9531017771&r=9531017771 .
The $700bn rescue plan can, at best, only help in the short term. Unfortunately the underlying long term attitudes which caused the credit crunch are still very firmly in place.
We should certainly hope that the $700bn might last long enough to allow development of new and tighter regulation for banks that is internationally agreed. Sadly it is hard to find solid grounds for this hope, because the regulation that banks really need is likely to work against the instincts and vested interests of most of the people involved in the discussions.
Banking need to be much, much simpler so that financial markets are more readily understandable. Complex derivatives proved to be much more effective at hiding risk that at managing it efficiently. Banks need to be much smaller so that failures are more manageable. Banks need to be less leveraged so they are safer and have more of a utility feel. The capital banks have available for speculation must be linked to market making obligations. Speculative capital should be strictly limited so that huge movements in short term capital cannot cause market lurches.
The incentives that drive banking behaviours need far more attention. The incentives that arise from holding a "long" equity position are much more constructive to the economy as a whole that the incentives which arise from holding a "short" position. The incentives that arise from pay need careful consideration. Top bankers should not be eligible for annual bonuses; all incentives should be on a long term basis. The top mangers in a bank must all have common incentives so that they work together, share information and form a common mind on the banks position and the state of the market. Above all the pay of top bankers must be much lower so that shareholders can feel confident that the top bankers are working for the shareholders' benefit not their own benefit. Very high pay increases the likelihood of ruthless and self-seeking characters at the top of the organisation; if you pay gold you get pirates!
In summary banking needs to become much, much more boring! Banking careers should appeal to steady and consistent people. Just like top athletes, top bankers should be subjected to regular drugs tests. The supercharged performance currently expected is not human and its puts inhuman pressures on other parts of the system.
Unfortunately we are still a very long way from a safe and boring banking system. And in the meantime what are the whiz-kids doing? Well I expect that the big prizes right now are for finding the best schemes to persuade government to take over and pay too much for the very worst assets. All this overcharged pursuit of money has killed many of our financial institutions. Are we going to allow it to kill our public finances?
Labels: banking, incentives, regulation
