Asset Swaps are adding to the Problem
“Asset Swaps” make it more difficult for banks to sell
Banks will often buy a fixed rate bond and enter into an interest rate swap to “effectively” turn it into a floating rate asset. It should come as no surprise that banks that rely the most on rolling over short term debt are the ones most likely to asset swap bonds – yes the “weak” banks are the ones that hold bonds in this form.
The Italian 3% of 11/1/15 is a perfect example. It was a new issue that was done in November 2010. The price at issue was 99.01. It is currently trading at 87.5, so a loss of 11.5% of notional. This was an ideal bond for banks to buy – a new issue with a 5 year original maturity. That loss is likely unrealized at a lot of banks that hold the bonds, as they put it into a hold to maturity account. That is a big enough problem, but the banks are losing on their swaps as well. On November 5th, 2010, the 5 year “mid-swaps” rate was 2.09%. Banks would have agreed to pay that rate and receive a floating rate in exchange. Currently, the 4 year “mid-swaps” rate is 1.66%. On a mark to market basis, they are down 43 bps, with a duration just over 3. That means they are down about 1.25% on their interest rate swap, in addition to being down on their bond purchase.
Banks that participated in the 3.75% of 4/15/2016 are in even worse shape. That bond came at a price of 99.79 and is now at 89.75, so down 10 points on the bond. On April 7th (when the bond was announced) the 5 year “mid-swaps” rate was 3.14%. They would have agreed to pay that rate vs receiving floating. The rate to that date is now about 1.74%. That is down 140 bps, with very close to a 4 year duration, so a mark to market loss on their interest rate swap of about 5.5%. The asset swap package is down 15.5 points and adds the element of counterparty risk.
Yes, there is that ugly term again, counterparty risk. Who is owed money on the swap by the banks that bought bonds via asset swaps? It is another dynamic that is at play that we just don’t know. It also makes it harder for banks to sell their assets because they would have to close out the interest rate swap at the same time.
This problem is much bigger for Italy and Spain than it was for Greece, Ireland, or Portugal. The size of the underlying markets means the amount of unmarked interest rate swap losses is also much larger. The other difference is that the loss on the swap is proportional to the loss on the underlying bonds. On Greece, or even Portugal and Ireland, the losses on the bonds quickly dwarfed the interest rate swap losses. On the second example, the loss on Italian bonds is only 10 points, and the loss on the interest rate swap is about 5.5 points, so it increases the loss on the positions by almost 40% – from down 10 points to down 15.5 points.
Since many of the banks may have entered into the swap agreement without margin requirements, closing the position would not only cause them to recognize the loss, but they would have to put up the cash. In a time where so many of the banks (the weakest ones in particular) are struggling with raising money, this makes it even more difficult to cut losses on positions.
The interest rate or asset swap exposure is still secondary to the losses on the underlying bonds, but is non trivial. The April 2016 bond is 15 billion in size, so if 100% had been asset swapped, that would be an unrealized and unmarked loss of €875 million.
I’m guessing most of the asset swap positions are 5 years and in, but I think the amount at risk is non-trivial but is opaque, concentrated with the weakest banks, and not uniformly well risk managed by their swap counterparties.
Asset swaps work fine so long as the underlying bond never becomes a “credit” problem. So long as it moves more in line with rates than with credit it is a sensible strategy. As Italy and Spain have become credit problems, they are no longer moving with rates, and these positions are adding to the problem. The banks that like asset swapping bonds typically rely the most on short term funding (because they were playing the spread curve), so this problem is amplified as they are the same banks that cannot afford the hits and will have the most difficulty raising new money.
I’ve never understood why banks in particular don’t demand more floating rate new issues, especially when I suspect that Italy went and did swaps with the street to turn some of their fixed rate funding into floating rate funding.