19 Jun ISDAfix Benchmark Swap Rate Manipulation
ISDAfix Swap Rate Manipulation – Q&A
The ISDAfix Benchmark Swap Rate Manipulation lawsuit involves 13 banks, all sued by Alaska Electrical Pension Board (AEPB) and other plaintiffs. The banks named are: Bank of America, Barclays, BNP Paribas, Citibank, Credit Suisse, Deutschebank, Goldman Sachs, Royal Bank of Scotland, J.P. Morgan, Chase Bank, Nomura Group, HSBC, UBS, and Wells Fargo & Co, Inc. Also named – and in bold print – is ICAP, PLC, the interdealer broker who was responsible for collecting the benchmark rates from the panel banks.
How are ISDAfix Benchmark Swap Rates fixed?
- The ISDAfix Benchmark rates are submitted to the Administrator (ICAP) at 11:00 AM New York Time.
- They are based on actual trades and the midpoint of the submitting bank’s quote.
- Benchmark rates are submitted for 2 to 30 years.
- ISDAfix refers to the fixed rate on an interest rate swap.
Who uses ISDAfix Benchmark Swap Rates, and how do they use them?
- ISDAfix Benchmark Swap Rates are used by roughly 6,000 corporate treasurers.
- Rates are used to value cash-settled swaptions on the expiration day.
- They are tied to commercial and residential real estate securities.
- Rate changes are used to calculate the performance of structured notes.
How did the submitting banks manipulate ISDAfix Swap Rates?
The manipulation seems surprisingly simplistic.
- Each business day (before 11:00 AM NY Time), just prior to the actual fixing time, ICAP would call the submitting banks (see list above) and provide an indicative ISDAfix swap rate.
- If the submitting bank agreed, their name was added to the list of institutions supporting that rate.
- If the submitting bank felt ISDAfix should be a different rate, their preferred rate was noted by ICAP.
What we have learned from court documents:
- Submitting banks with large expiring swaption positions would communicate the rate they wanted for the Benchmark Rate.
- The banks manipulated the rate in one of three ways:
- They were trading swaps to push the market to the rate needed.
- ICAP would wait to process the submission rates that were used to calculate the rate.
- If several banks had swaptions expiring, they would all report the same rate to ICAP. (From 2009 – 2012, the rates from all submitting banks were almost the same.)
It didn’t seem to matter whether the bank was long or short, payers or receivers – they rigged rates for extra profitability regardless.
Why did the administrator (ICAP) help the banks rig rates?
As the administrator, ICAP was supposed to follow this procedure:
- ICAP posts a beginning poll rate for all maturities (11:00 AM NY time).
- Collected transaction rates (based on trades) from the submitting banks
- Finally, they were to get a market from each submitting bank just prior to fixing and use the midpoint of each quote.
Conflict of Interest?
ICAP was also an inter-dealer broker, making a commission on every transaction with the banks. ICAP reaped a commission for every trade, particularly on those days when submitting banks would execute rapid-fire transaction just prior to fixing.
Proof of Collusion
Beyond the 1 million emails, phone calls, and other internal communications, the most obvious proof of collusion is found by viewing the range of swap rates submitted to ICAP.
- ISDAfix rates were identical out to five decimal places 90% of the time.
- Once the investigation began in December 2012, ISDAfix rates were identical only 70% of the time.
Changes Made to ISDAfix Benchmark, 2014
The following changes have been announced:
- Clarification of ISDAfix now emphasizes that contributing banks use executable bid/offer rates
- Suspension of ISDAfix maturities or currencies with insufficient liquidity in the underlying swap market
- Stronger checks on submission rates for validation
The term, “stronger checks,” seems a bit ambiguous, and this blogger hopes the language is strengthened to afford market participants greater confidence as to the course and criteria for these all-important benchmark rates.
ISDAfix Benchmark Rates & The Impact on Cash-Settled Swaptions
Cash-Settled Payer Swaption:
- The buyer of a cash-settled payer swaption has the right, but not the obligation, to exercise their swaption on the expiration date. The buyer of a swaption pays a premium.
- The seller receives that premium. In return for receiving the premium, the seller is obligated to pay the buyer if the option is in-the-money at expiration.
- The payer swaption will be exercised if the swap rate is HIGHER THAN the strike rate on the expiration date.
- If the option is in-the-money, the seller will pay the Present Value of the difference between the Benchmark rate – the Strike Rate.
Cash-Settled Receiver Swaption:
- The buyer of a cash-settled receiver swaption has the right, but not the obligation, to exercise their swaption on the expiration date. The buyer of a swaption pays a premium.
- The seller receives that premium. In return for receiving the premium, the seller is obligated to pay the buyer if the option is in-the-money at expiration.
- The receiver swaption will be exercised if the swap rate is LOWER THAN the strike rate on the expiration date.
- If the option is in-the-money, the seller will pay the Present Value of the difference between the Strike Rate – the Benchmark Rate.
Cash-Settled Swaption Example:
The client is long a 1-year PAYER into a 10-year swap, the strike rate is 2%. On the expiration date, the ISDAfix 10-yr Benchmark swap rate is 2.50%. The buyer receives 50 bps per year, discounted back to Present Value.
But if the “true” rate was 3.51%, the client would have made 51 bps, strike rate 3% – 3.51%.
The client should have made 1bps more than they did.
The table below shows the result if ISDAfix Swap rates were benchmarked 1 bps higher. On $1 million notional amount on 10 yr swap rates, the buyer has left $902.54 on the table per $1 million notional amount.
Assuming the average maturity of all swaptions outstanding ($29 trillion) were 10-year swaps, all long contracts would be leaving $26.17 billion on the table due to the ISDAfix Benchmark rate manipulation.
The real impact of incorrect ISDAfix Swap Reset Rates was much larger than the above diagram shows. Note that this diagram only shows the impact of 1bp difference on swaptions alone. Many other products use the ISDAfix Benchmark Swap Rates.