Much to the chagrin of bond underwriters and other market participants, both House and Senate bills aim to take away access to advance refundings. Should the proposals become law, all is not lost for hospitals wanting to lock in today’s low rates to lower their cost of funds –but they must be willing to confront their fears and swim against the tide.

Muni borrowers currently have a couple of options to refund existing tax exempt debt and generate savings.

The first option is a current refunding, which requires the bonds to be callable. Most tax-exempt bonds are callable starting 10 years from their date of issuance, and there is no IRS restriction on how many times bonds can be current refunded with new tax exempt bonds.

The second option is an advance refunding when bonds are not currently callable. In an advance refunding, new bonds are sold today and the proceeds are deposited into an escrow to make debt service payment on the old bonds. Once the escrow is funded, the old bonds are considered defeased and taken off the balance sheet. On first call date, the old bonds are called and retired.

Since 1986, bonds can only be advance refunded once on a tax-exempt basis. 

Both the House and Senate bills propose to eliminate tax-exempt advance refundings, and while borrowers could still refund bonds with taxable debt, the higher cost of funds would likely kill most refunding savings.

If advance refundings go away, borrowers still have a couple of options:

  • MMD rate lock. The hospital buys a rate lock based on today’s MMD yield. If on first call date MMD has gone up, the hospital receives a payment that reduces the refunding par amount. If MMD has gone down, the hospital makes a payment funded from increasing the refunding par amount. Either way, the adjustment to the refunding par keeps the refunding debt service in line to what it would be today, so the hospital locks in savings. MMD rate locks can be expensive.
  • Forward-start, cash-settle swap. This option is similar to a rate lock, but is based on a percentage of LIBOR. It is significantly cheaper than the MMD lock because the LIBOR indexing introduces basis risk (risk that taxable and tax-exempt rates may not move in unison), which may reduce savings.

Properly structured, these options can be effective in locking in most, but not all of the savings that would be expected under an advance refunding, because certain assumptions have to be made about how the refunding bonds will amortize and the shape of the yield curve, both subject to change.

But aside from these risks, the bigger challenge for some hospitals may be overcoming vertophobia, or fear of swaps.

A few years ago, some hospitals were talked by bankers into 20-plus year pay-fixed swaps to hedge floating rate debt with much shorter maturities (auction rate notes or low floaters). Things got ugly when the underlying debt was refinanced with fixed rate bonds and the swaps became orphaned, because in the meanwhile, rates had gone down. This made the swaps a liability, and terminating would require writing a very big check. That didn’t sit well with Boards.

These alternatives involve swaps, albeit with much shorter terms. Hospitals and financial advisors willing to understand how swaps work and the risks involved may decide that properly structured, they can help lower the cost of funds in a world without advance refundings.

For a more extensive discussion, see Alternatives When Advance Refundings Don’t Work.