SWAP, revisited

An 8-1 vote should be explained – especially if you’re the “1”. In the spirit of “a mathematician does not understand his work until he can explain it to the first man he meets in the street”, it’s clear that I need to take another run at this.

The Board’s intent, with Monday’s resolution to terminate our Interest Rate Swap agreement if it the cost drops to $5.5 million, was an attempt to minimize financial loss to the district.

The reason a resolution could be valuable is that interest rates tend to be volatile. (In particular, 10-year Treasury rates, which appear to have an inverse relationship to the cost of ending the SWAP agreement)  Since the board only meets twice a month, having the resolution in place allows the administration to act quickly should a predetermined trigger point be reached.

The question is: at what point should that trigger be set?

Everything is a “gamble”. If we do nothing between now and the end of the year, we’re “gambling” that interest rates won’t decrease, which would add to the cost of ending the agreement, perhaps significantly. On the other hand, were we to decide to end the agreement right now, we would be “gambling” that rates won’t increase – with the result that we would have spent more than necessary. The resolution is an attempt to balance those risks.

What’s going to happen? Of course, nobody knows. What’s likely to happen? Well, again, nobody really knows. But we have some educated guesses from the 60 financial firms that were surveyed by Bloomburg. Of course, these educated guesses are all over the map, but on average they project that the 10-year Treasury rate will rise to 4.15% by the end of the year. Under that assumption - not very solid, but that’s the information we have - the cost of ending the SWAP agreement at that point would be about $5 million..

So how does one decide where to set the trigger? From a strictly mathematical perspective, you would start with the $5 million figure – the point at which the risk of “spending more than necessary to get out early” exactly balances the probability of “waiting too long”, and then pick a number that’s slightly lower.

(Why, lower? Volatility. If the expectation is that rates will reach a certain number, there’s a good chance that you’ll reach a slightly higher peak at some point along the way, and you want to catch that if you can. Keep in mind that without volatility – if the swap termination value moved smoothly - there would be no reason to have the resolution at all; you could just wait until interest rates plateaued.) *

But here’s the problem with this analysis – we don’t make decisions from a strictly mathematical perspective, and for good reason. Here’s an example.

Say that a person you know to be trustworthy (and rich) offers you the following deal: flip a coin, heads – you pay him $10, tails – he pays you $15. A pretty good deal, and one that you’re likely to take.  But suppose the numbers are $10,000 and $15,000, respectively. Unless you’re pretty well off – or desperate - you’re not going to take that deal. The reason is quite rational: a $10,000 loss has a greater “value” to most of us than a $15,000 gain. (This is why you should never play high-stakes poker with a wealthy person; they value risk differently.)

There have been a number of studies demonstrating that human beings place a higher “value” on what they stand to lose than on what they might gain, even if when the extrinsic values are exactly the same.

The danger lies in applying this thinking to an organization as large as the school district, where the value of “losing $500,000” and “failing to save $500,000” are exactly the same, particularly when the entire amount will be amortized over ten or twenty years.

I said I’d take another run at it.  Persuasive or not, there it is.

*p.s.  Particularly astute readers will recognize that termination agreement such as this is useful primarily in a volatile, flat market. If interest rates are projected to go down, you should get out now. If they're projected to go up (as is the present case) you should hold tight for a while longer.