Market Distortions, LIBOR Manipulation, and the Limits of Mispricing Models
Summary
The discussion asks how traders and modelers should respond when a widely used market rate is manipulated. It challenges the idea that market prices are necessarily correct and observes that valuation models are often calibrated to market inputs, including when those inputs imply unusual outcomes. The answer frames the central problem as practical: recognizing a distorted price does not guarantee the ability to trade against it or cause it to move toward a model-based estimate.
The response lists obstacles faced by participants who believed LIBOR was wrong, including incentives to participate, dependence on the rate in borrowing and lending, and limited confidence in reporting channels. It then generalizes the point to perceived option-skew mispricing: a trader can take a contrary position, but counterparties and convergence are uncertain. The document offers no statistical method for detecting manipulation or determining fair value, so it serves as a caution about model risk, incentives, and limits to arbitrage rather than a detection procedure.
Key ideas
- Market prices used to calibrate models can be distorted and should not be treated as inherently correct.
- Identifying a suspected mispricing does not ensure that a trader can act on it or profit from it.
- Funding needs, incentives, and reporting institutions can constrain responses to suspected benchmark manipulation.
- A model-based fair value may fail to attract counterparties or converge to market prices.
Tags
Full text
# if market is always assumed right, what happened when LIBOR was manupulated? # if market is always assumed right, what happened when LIBOR was manupulated? Recently Monetary Authority of Singapore (MAS) raps banks in rate-rigging. This is nothing new, LIBOR was also manupulated before, by some "major" banks. however, before the censorship, did any "minor" bank declare that the interest rate from LIBOR is distorted? seems modellers always assume the market is right, so models, or parameters used in the model are always first calibrated according to the market, then used to price products. so even there is a problem, people will try to find some work-around. for example if the term structure leds to a negative future rate, the curve is still accepted. is there some mechanism to detect such abnormality and flag "market is wrong" ? ## Answer by Matt Wolf (score 1, accepted) https://quant.stackexchange.com/a/8292 Nobody assumes the market is right. The issue with Libor was that nobody could do much about it. Well, not 100% correct, you had the following 3 choices: - Participate in the scheme and benefit (monetarily) - Be on the receiving end and despite you knowing the rate is not what it should be you have no choice, you gotta borrow/lend your funds at the end of the day no matter what. - You complain to regulators, the same regulators whose salaries are paid to a large degree by the very same bank conglomerates that set Libor every day. The choice was/is yours... So, even if you know that market prices are distorted what are you gonna do about it. Assume for a second that you know that the skew of most index options should be much more pronounced. You basically believe the market has it wrong and does not price risk appropriately. If you want to trade on your believes you can of course do so, however, whether anyone else trades with you is another question, entirely. And if the market does not converge to your own estimate of where you see it is fairly valued at then you will not even benefit from your "correct model".
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.