The Treasury Department should alter proposed arbitrage regulations that include a two-pronged test to determine whether the floating rate on certain interest rate swaps and the variable rate on their underlying bonds are substantially the same, according to two industry groups. Recent comment letters submitted by the National Association of Bond Lawyers and the Securities Industry and Financial Markets Association praised the proposed regulations, while encouraging federal officials to make the technical changes to them. A third letter, by the Government Finance Officers Association, expressed support for the NABL and SIFMA comments.Treasury officials released the proposed regulations in September and solicited comments on them through last week. It plans to hold a public hearing on them Jan. 30, at which representatives of NABL and GFOA plan to speak.The investment of an issuer’s tax-exempt debt is subject to a set of rules commonly referred to as the arbitrage regulations, with “arbitrage” referring to earnings on bond proceeds that materially exceed the yield on the bonds. The existing regulations provide detailed rules for determining and restricting yield on the bond issue and investments, and for computing and paying arbitrage rebate to the federal government. If an issuer does not comply with the regulations, it risks the Internal Revenue Service declaring its bonds taxable.In general, the proposed rules would make several discrete changes to the existing rules, including accommodating interest rate swaps with floating payments based on taxable rates, such as the London Interbank Offered Rate.Libor swaps, which have become commonplace in the muni market, would be eligible for simple integration, under which variable-rate bonds subject to an integrated floating-to-fixed rate swap are treated as variable-yield bonds. But such swaps would not be eligible for super integration, which has a separate set of requirements that allow the bonds to be treated as fixed-yield bonds.NABL and SIFMA both focused on a proposed two-prong “snapshot” and “historical” test, which would provide that the floating rate on a taxable-index hedge and the variable rate on the hedged bonds be treated as “substantially the same” for the purposes of the hedging rules if the rates are no more than 0.25% apart, and if, for a three-year period ending on the date the issuer enters into the swap, the average difference between the rates does not exceed 0.25%.NABL argued that the tests are unnecessary because the existing rules already require relatively close correlation between the floating rate on the hedge and the variable rate on the hedged bonds. For instance, the limitations explicitly prohibit correlation with equity indexes, such as the S&P 500 index, which don’t move closely with tax-exempt markets, NABL said in its letter.If scrapping both tests is not possible, NABL suggested that Treasury allow issuers to meet a modified version of the proposed historical test, which the group said would provide more meaningful comparison of rates, better allow issuers to consider and manage the potential basis differential between the floating rates on their bonds and the floating rates on related hedges, as well as provide the IRS and Treasury a reasonable assurance of correlation.NABL suggested the historical test be conducted over “at least” three years, rather than exactly three years.“That would enable issuers to examine rates over a longer period if they feel it gives them a better measure of their potential basis risk,” said Douglas A. Bird, a partner at McKee Nelson LLP and primary author of the NABL letter.SIFMA recommended that issuers be allowed to meet either the snapshot test or the historical test because its research showed that a significant number of transactions executed over the past 15 years would fail one or both tests. SIFMA also recommended that the snapshot test be modified to reflect a 30-day average rate for both the floating-swap rate and the “expected” variable interest rate on the hedged bonds, to smooth out market abnormalities that may exist on any particular day. Alternatively, SIFMA suggested that Treasury eliminate the snapshot test, which it projected issuers may have failed over 40% of the time during the past 15 years, while maintaining a three-year computation period for the historic test and allowing an issuer to make appropriate adjustments for changes in federal and state tax law, credit quality, and other market issues.
