New data on interest rates - regulators, please note

If you head over to the RBNZ's statistics on wholesale interest rates, and you have a look at the spreadsheets with 'Close' in their title - B2 Daily close (2010-current) or B2 Monthly close (2010-current) - you'll find that the RB has compiled some new series on 'swap rates'. There are swap rates for 1, 2, 3, 4, 5, 7, 10, and 15 year maturities, and for convenience a measure of the steepness of the swap rate curve (10 year swap rate less 2 year swap rate).

What we've got, in short, is the local equivalent of the Libor yield curve in the UK - the rates at which the most creditworthy institutions are prepared to lend to each other for different maturities. They're essentially the bedrock rate for corporate borrowing: all other corporates (lesser quality banks, non-bank corporates) will pay the swap rate plus a credit margin on funds they borrow.
It's useful info that hasn't been readily or conveniently available to the public up to now, so well done the RB.

Here's a comparison, by the way, using this new data, of the 10 year government stock yield with the 10 year swap rate. If this seems esoteric, there is a potential regulatory point lurking in here.

You'll see that the swap rate generally lies a little bit above the government stock yield - which it should. There's usually a small element of credit quality difference between the government and the highest quality banks, which interbank lenders need to charge for.

But occasionally the gap widens quite a bit - as it did, for example, in the April-June '13 period. The government stock yield dropped quite sharply, on this occasion because the world's markets were going through one of their intermittent jitters (the MSCI World index of global share prices dropped by over 8% between late May and late June '13). Government bonds were seen as the "safe haven" asset par excellence in times of equity volatility, so consequently their price got bid up and their yield got bid down. Banks' funding costs dropped too, but by nowhere near as much, so the gap between government bond yields and swap rates widened. The difference between the price of government risk and the price of corporate risk, in short, is not a constant. It changes, sometimes at one end, sometimes at the other, sometimes both ends at once. The likes of a GFC, for example, saw government bond yields drop sharply, but corporate credit spreads rise very sharply.

Which brings me to my regulatory point.

When they are figuring out the allowable rate of return on a regulated asset base, regulators set an allowable rate of return on equity, and an allowable cost of debt (combined, they make up 'WACC', the weighted average cost of capital). On the debt side, and for reasons never obvious to me, regulators typically do not look at the actual costs of debt for a regulated entity (which ought to be directly observable in many cases). Instead, they often like to express the cost-of-debt element as a formula, namely the "risk free" rate plus a credit margin appropriate for the regulated entity. Typically, the "risk free" rate is taken to be the government bond yield.

Aha! You see the problem.

Let's say a regulated entity's actual cost of debt is 7%, when the government bond yield for the same maturity at the time is 5% and the swap rate is 5.5%. To a regulator, the cost of debt is the "risk free" 5% plus a credit margin of 2%. All good.

Now along comes a regulatory price reset. In the meantime, as in the chart above, the government bond yield may fortuitously have dropped to an unusually low level, say to 4%. The swap rate may have dropped a bit too, to 5%. But nothing has changed to the corporate's credit rating - it was paying 1.5% over swap before (i.e. over the best-quality corporates), and it is paying 1.5% over swap now. Its cost of debt is 6.5%.

Allowing the regulated entity the cost of debt from the government stock formula, 4% plus 2%, will undercompensate it by 0.5% - which would be a large amount for an entity with a lot of physical capital funded by debt.

So my suggestion would be, if regulators must use a formula, use the swap rate plus a margin.