The revelations that have surfaced in the past couple of weeks over the setting of LIBOR (London inter-bank offer rate) have been nothing short of amazing, especially for a fairly simple-minded and even sometimes naive economist such as myself.
When we teach economics we spend a lot of time talking about markets and how they work, and we usually discuss markets in terms of the supply side, the demand side, and the (equilibrium) price. Curiously, though, when it comes to the details of price setting, we are often a bit vague, relying on the 'market mechanism' to set proper prices without thinking terribly much about what exactly the price should be. If we do give prices some thought, it is usually to argue that competition between firms will make sure any price distortions cannot last long - if a price is too high, new firms will enter the market; if it is too low, firms will fail and leave the market. And mostly this simple story seems to work quite well.
However, in markets where there are not multiple price setters, the story can break down badly as we discovered in the case of LIBOR. LIBOR is actually a really fundamental 'price' in the financial markets, since massive volumes of contracts (including some mortgages) pay an interest rate that is expressed as so many basis points (100 basis points make a 1% interest rate) above LIBOR. So if the way LIBOR is set is wrong, lots of folk - both firms and individuals - will end up paying the wrong price (interest rate) on their borrowing. Mostly, I imagine, they would end up paying too much (thereby raising bank profits), but sometimes the error could go the other way, and they would pay too little. Thus it's quite important that the LIBOR is set correctly and fairly.
Yet despite its importance, LIBOR is one of those prices to which most economists have paid absolutely no attention. Hence it was interesting, and quite a revelation (to me), to discover that it is set daily by the British Bankers Association (BBA) on the basis of submissions from their member banks. Each bank is supposed to submit data on the interest rate it has to pay that morning for inter-bank borrowings on the wholesale market for funds. The BBA apparently drops the four highest and the four lowest submissions, then averages the rest to set the LIBOR for that day. The whole process seems to me surprisingly informal, and it relies heavily on the truthful submission of the relevant data by the member banks. In other words, setting LIBOR is based on an assumption that bankers are 'gentlemen' in the old fashioned sense of being completely honest, faithful to the official 'rules of the game', and not at all out to play the system for personal gain.
Recent events revealed what one might have guessed if one had investigated the LIBOR price setting process at all carefully, which no one apparently did - though the regulators had at least started to ask the right questions, we gathered. This is that the bankers are not in fact the gentlemen they were naively assumed to be, and that they did, over an extended period, systematically strive to distort LIBOR by providing incorrect data to the BBA. In other words, they lied.
Why might banks do this? Well, there are two reasons that come to mind. First, the interest rate a bank has to pay to get funds in the wholesale market is sometimes viewed as an indicator of the health of the bank. If a bank can borrow cheaply, it is considered healthy; if it has to pay a high interest rate, this is deemed to reflect a market judgement that the bank is in poor shape. So this line of thinking might make a bank try to achieve a lower value for LIBOR than is really correct, by claiming that it doesn't have to pay much to get wholesale funds. Second, the bank might want to get LIBOR higher than it should be so that its customers then have to pay more for their loans and the banks make higher profits.
These factors push in different directions, but either way one can see that banks might have incentives not to be totally truthful. And with LIBOR set in such an informal and non-transparent way, it wasn't easy for anyone to verify what was really going on, not even the BBA itself. As a result, it was all too easy for this price-setting process, based on the regular and truthful submission of information, to break down.
Two lessons can be drawn from this sad and sorry episode. One is that economists need to think more carefully about the specific ways in which important prices in the economy are set, and the associated incentives that affect the various players in the relevant markets. Doing this could lead to a significant programme of rather important and interesting economics research. The other lesson is simply that setting LIBOR surely cannot continue to be done as it has been hitherto.
When we teach economics we spend a lot of time talking about markets and how they work, and we usually discuss markets in terms of the supply side, the demand side, and the (equilibrium) price. Curiously, though, when it comes to the details of price setting, we are often a bit vague, relying on the 'market mechanism' to set proper prices without thinking terribly much about what exactly the price should be. If we do give prices some thought, it is usually to argue that competition between firms will make sure any price distortions cannot last long - if a price is too high, new firms will enter the market; if it is too low, firms will fail and leave the market. And mostly this simple story seems to work quite well.
However, in markets where there are not multiple price setters, the story can break down badly as we discovered in the case of LIBOR. LIBOR is actually a really fundamental 'price' in the financial markets, since massive volumes of contracts (including some mortgages) pay an interest rate that is expressed as so many basis points (100 basis points make a 1% interest rate) above LIBOR. So if the way LIBOR is set is wrong, lots of folk - both firms and individuals - will end up paying the wrong price (interest rate) on their borrowing. Mostly, I imagine, they would end up paying too much (thereby raising bank profits), but sometimes the error could go the other way, and they would pay too little. Thus it's quite important that the LIBOR is set correctly and fairly.
Yet despite its importance, LIBOR is one of those prices to which most economists have paid absolutely no attention. Hence it was interesting, and quite a revelation (to me), to discover that it is set daily by the British Bankers Association (BBA) on the basis of submissions from their member banks. Each bank is supposed to submit data on the interest rate it has to pay that morning for inter-bank borrowings on the wholesale market for funds. The BBA apparently drops the four highest and the four lowest submissions, then averages the rest to set the LIBOR for that day. The whole process seems to me surprisingly informal, and it relies heavily on the truthful submission of the relevant data by the member banks. In other words, setting LIBOR is based on an assumption that bankers are 'gentlemen' in the old fashioned sense of being completely honest, faithful to the official 'rules of the game', and not at all out to play the system for personal gain.
Recent events revealed what one might have guessed if one had investigated the LIBOR price setting process at all carefully, which no one apparently did - though the regulators had at least started to ask the right questions, we gathered. This is that the bankers are not in fact the gentlemen they were naively assumed to be, and that they did, over an extended period, systematically strive to distort LIBOR by providing incorrect data to the BBA. In other words, they lied.
Why might banks do this? Well, there are two reasons that come to mind. First, the interest rate a bank has to pay to get funds in the wholesale market is sometimes viewed as an indicator of the health of the bank. If a bank can borrow cheaply, it is considered healthy; if it has to pay a high interest rate, this is deemed to reflect a market judgement that the bank is in poor shape. So this line of thinking might make a bank try to achieve a lower value for LIBOR than is really correct, by claiming that it doesn't have to pay much to get wholesale funds. Second, the bank might want to get LIBOR higher than it should be so that its customers then have to pay more for their loans and the banks make higher profits.
These factors push in different directions, but either way one can see that banks might have incentives not to be totally truthful. And with LIBOR set in such an informal and non-transparent way, it wasn't easy for anyone to verify what was really going on, not even the BBA itself. As a result, it was all too easy for this price-setting process, based on the regular and truthful submission of information, to break down.
Two lessons can be drawn from this sad and sorry episode. One is that economists need to think more carefully about the specific ways in which important prices in the economy are set, and the associated incentives that affect the various players in the relevant markets. Doing this could lead to a significant programme of rather important and interesting economics research. The other lesson is simply that setting LIBOR surely cannot continue to be done as it has been hitherto.
No comments:
Post a Comment