LIBOR INTRODUCTION

The London interbank offered rate (LIBOR) is a primary benchmark for short term interest rates globally and is used as the basis for settlement of interest rate contracts on many futures and options exchanges. It is used in Used in loan agreements throughout global markets Mortgage agreements Considered as a barometer to measure the health of financial money markets.

Although reference is often made to the LIBOR interest rate, there are actually 150 different LIBOR interest rates. LIBOR is calculated for 15 different maturities and for 10 different currencies. The official LIBOR interest rates (bbalibor) are announced once a day at around 11.45am London time by Thomas Reuters on behalf of the British Bankers association (BBA). American dollar - USD LIBOR Australian dollar- AUD LIBOR British pound sterling - GBP LIBOR Canadian dollar- CAD LIBOR Danish krone - DKK LIBOR European euro - EUR LIBOR Japanese yen - JPY LIBOR New Zealand dollar - NZD LIBOR Swedish krona - SEK LIBOR Swiss franc - CHF LIBOR

Each day, the BBA surveys a panel of banks, asking the question, At what rate could you borrow funds, were you to do so by asking for and then accepting inter-bank offers in a reasonable market size just prior to 11 am?. The BBA throws out the highest and lowest portion of the responses, and averages the remaining middle. The average is reported at 11:30 a.m. LIBOR is actually a set of indexes. There are separate LIBOR rates reported for 15 different maturities (length of time to repay a debt) for each of 10 currencies. The shortest maturity is

overnight, the longest is one year. In the United States, many private contracts reference the three month dollar LIBOR, which is the index resulting from asking the panel what rate they would pay to borrow dollars for three months.

WHAT WENT WRONG AND HOW DID THEY MANIPULATE?

The BBAs rate-setting process has certain inherent subjectivities. It is a self-reporting process as each submitting bank provides to the BBA an estimate of the interest rate at which the bank believes it could borrow from other banks. It is an estimate rather than an actual interbank borrowing cost; by its very nature, it is not a precise, market-based price. The global credit crisis in 2007 and 2008 exemplified the limitation in deriving LIBOR based on estimated borrowing rates. During the peak of the crisis in which there was considerable dislocation in the credit markets, LIBOR-submitting banks short-term financing composition shifted from interbank borrowings to government-issued financing. The decline in interbank borrowing may have made it more difficult for banks to accurately assess their interbank borrowing rates. In addition, LIBOR is determined based on a trimmed mean approach. Trimmed mean averaging reduces the effects of outliers and can provide a more robust estimation of the average than an arithmetic mean of all data points. In most instances, a single misstated LIBOR submission would have very little or no impact on the calculated trimmed mean. For example, in instances where both the manipulated submission and correct submission were to fall into the same top or bottom quartile, there would be no impact on the resulting LIBOR calculation. Two unlikely circumstances would have to occur for a single misstated LIBOR submission to have a material impact on the calculation of the trimmed mean i) a substantial differential between the top and bottom quartiles; and ii) a bank would have to submit a misstated rate that would cause the rate either to be included in the trimmed mean calculation as stated or to fall in a different quartile than it would have had it not been misstated. An example is illustrated below to explain how a small variation in the submitted rate by a bank can alter the overall LIBOR rate.

As illustrated in Panel A of Exhibit 2, based on the rates submitted by the 18 banks, using the trimmed mean approach, rates for Banks 1 through 4 (top 4) and Banks 15 through 18 (bottom 4) would be excluded and the resulting average would be 4.15%. In Panel B, we assume that the correct rate for Bank 2 is 4.9%, but is submitted as 3.35%. Panel C demonstrates how this incorrect submission affects the ultimate computed rate. With the submission at 3.35%, Bank 2 becomes the second lowest among the 18 submitted rates, and the rate of 3.6% submitted by Bank 15, which was correctly excluded under the Panel A calculation, is now included and part of the averaging process. The rate of 4.6% submitted by Bank 5, on the other hand, is now excluded from the averaging process. As a result, the calculated LIBOR has dropped to 4.05% (10 basis points lower than what it should be).

WHEN DID THEY START MANIPULATION?

Starting in the fall of 2007, almost five years ago, the financial press started writing about potential abuses, reflecting what their sources were telling them. Gillian Tett in the Financial Times appears to have been the first: On September 4 of that year she quoted a banker as calling Libor a fiction. The Wall Street Journal in April 2008 also ran the first of several articles about Libors reliability. At the same time, suspicions started being aired in more official circles, including at meetings of bond trading associations and even in government publications, using data from the markets themselves. How so? Largely it has to do with the nature of Libor and Euribor themselves.

These rates are set by panels of banks who meet daily to disclose the average rate at which they can obtain unsecured funding for a given period. The Libor fixings, for example, cover 10 currencies, and the periods range from overnight up to a year. The top and bottom 25% of these submissions are eliminated, and the rate is calculated using the average of the 50% that are left. But, critically, the rates that the banks submit in these sessions are not the rates they are actually paying to borrow money, but rather the rates that they estimate they would have to pay. Until the 2007 crisis, the Libor and Euribor rates tracked quite accurately the rates that were actually transacted, and which are noted by central banks. But suddenly, after Lehman, there was a very significant divergence between the benchmark Libor and Euribor rates, and the actual transaction rates. Jean-Franois Borgy spotted that at once. Hes a French former swaps trader with a long record in the markets, having worked for Credit Lyonnais, Banque Worms and Natixis. He was so struck by the change that, in October 2007, he gave a presentation to the annual meeting of the European Bond Commission. In a series of charts, he showed how the Euribor three-month rate had, since the 1999 introduction of the euro as a currency, almost without exception had been mirrored by the effective European overnight rate based on actual transactions. The spread or difference between this Eonia (Euro OverNight Index Average) rate and the Euribor rate had consistently been 6.3 basis points, or 0.063%. But with the crisis it suddenly jumped to 40 basis points, or 0.4%. For a trader, thats a huge and obvious change. As Borgy explained in his 2007 presentation, the Eonia rate is drawn from the same 47 banks who are involved in the Euribor fixing; the difference is that Eonia reflects real transactions by the banks, while Euribor simply reflects what the banks say they will do.

IMPLICATIONS ON GLOBAL FINANCIAL MARKETS

Parties involved in LIBOR linked financial instruments are numerous banks, Investors, hedgers, Pensions funds, borrowers globally, mortgage borrowers etc. In general, lower LIBOR hurts banks as lenders, benefits the borrowers and hence its against banks interest to keep LIBOR low. However during the crisis, LIBOR levels have become a signal to the

market. At that time irrespective of the gains/losses on the lending, a number of submitting banks had interest in submitting a lower rate and keeping the LIBOR low. IMPACT OF SCANDAL During the crisis period investors and lenders lost money for LIBOR being artificially low and borrower saved money In pre-crisis period, its difficult to judge who gained and who lost since LIBOR fixing could be higher or lower than fair levels. Calculating fair level itself s difficult. If a bank underreported by 20 bps we have to see whether the banks submission was included or excluded. If included then they would have had only 2 bps impact on LIBOR. However when multiple banks gives data in such a way that it get included in that LIBOR will be higher. Gross vs. Net Impact In many instances, an institution that experienced losses on certain transactions as a result of an artificially low LIBOR could have also experienced gains on other transactions. For example, many corporations and financial institutions that purchase floating-rate investments also issue floating rate debt. The key question in these instances surrounds whether the financial impact should be assessed on a gross or net basis. If the latter applies, an analysis of the total LIBOR-indexed portfolio may need to be performed in order to estimate the net economic impact of any alleged LIBOR manipulation and to return the plaintiffs to the economic position they would have enjoyed but for the alleged manipulation. Complexity of Securitized Cash Flows Many special purpose vehicles (SPVs) that issue securitized debt (also described as securitized products), such as collateralized debt obligations, have both liabilities and underlying assets that pay coupons tied to LIBOR. In many cases, SPVs have also entered into LIBOR based derivative contracts to match the asset and liability cash flow characteristics and/or protect investors against the impact of large, adverse movements in LIBOR. The universe of securitized products is vast and such vehicles often have complex waterfall structures and unique triggers that affect cash flows to the various investor classes. Estimating the true extent of losses associated with LIBOR manipulation would in many cases require sophisticated financial modelling capabilities.

Forward Interest Rate Curves There are other potential second order effects which are difficult to quantify or estimate. For example, many interest rate derivatives are valued based on projected future interest rates (usually referred to as a forward curve). Artificially low spot LIBOR could have a spill over effect and lead to a depressed forward curve. This could affect the mark-to-market value of many over-the-counter derivative instruments and could have major implications for collateral and margin requirements.