This one is going to dive deep into regulatory arcana, so run for the exits (or the back button now).

Verizon is none too happy about a recent NH PUC Order setting its cost of capital in the state. Indeed, Verizon is "shocked and troubled," emotions not easy for a corporate entity to muster.

The New Hampshire Order goes through a discounted cash flow analysis of Verizon New Hampshire to arrive at an average weighted cost of capital of 8.2% (see page 70 of the Order). This figure is the product of Verizon's imputed capital structure, estimating the cost of long- and short-term debt, along with the cost of equity. The pressing question now: Who cares?

First of all, Verizon cares. The decision on its cost of capital forms the kernel for all its wholesale and retail rates in New Hampshire. A higher cost of capital leads to higher rates; conversely, a lower cost of capital leads to lower rates. For Verizon, a lower cost of capital makes them less likely to invest in New Hampshire, particularly if the regulator-set rate is below its actual cost of capital. CLECs care, but in the opposite direction. A lower cost of capital makes for lower wholesale rates. Paradoxically though, the CLECs' presence and competitive threat to Verizon is the very thing that would tend to inflate Verizon's cost of capital. A low cost of capital is associated with Verizon's former status as a regulated monopoly.

Consumers, meanwhile, want the right cost of capital. Call it the Goldilock's principle: a too high cost of capital assumption leads to artificially high rates; a too low cost of capital assumption leads to low rates, but also reduced investment -- and the short-term pleasure of artificially low rates will be more than offset by the long-term effects of reduced investment.

But what is the right answer for the cost of capital? Like many regulatory determinations, the process pretends to a rigor and precision that is just not there. You can express DCF analysis mathematically (this page has a good discussion of DCF analysis, which has much wider application than just the regulatory sphere). But the big decisions arrive in estimating the discount rate and other forward-looking assumptions that predict the risk that a given firm faces. These numbers are by no means clear. Indeed, they are particularly difficult to estimate in a dynamic market like communications where the technology, market definitions and players are in great flux.

DCF analysis is therefore so plastic that it allows illegitimate concerns to drive the analysis. Thus, if the desire is "low rates" then arriving at a low average weighted cost of capital is a way to do it. And this is the problem. The inability to make any analytically rigorous input assumptions allows secondary concerns to predominate. Do I want to induce entry --even artificial entry -- into the market? Then derive low wholesale rates that will make entry attractive. Cost of capital assumptions are just the ticket, too, for they plug into the rate formulas at an elemental level and hence cascade through the rate structure. However, it is very difficult to say definitively that the NH PUC decision is wrong. It looks low. It uses some questionable assumptions, to be sure, but there is enough indeterminacy in the analysis that it is on the outer bounds of defensible-ness.

DCF analysis, like so much rate setting, is an exercise in analytic futility. And I know of what I speak. In 2002, I set a cost of capital for Qwest of 9.55% here. I am not denying it must be done, but you don't want to do it very long, and regulators need to be conscious of the effects and incentives they face when making these exceedingly difficult decisions.

This brings me to my final point about the difference between regulatory versus market supervision of an industry. When regulators get something wrong, those errors will be durable and difficult to reverse. Decisions will be made in New Hampshire -- by Verizon and by CLECs -- based on this remarkably low cost of capital. Reliance interests will congeal around this and -- if it was erroneously low -- it will be difficult for the NH PUC to back out of it, for they will be raising rates if they do.

In contrast, markets use DCF analysis to value companies and decide whether they are worthy of investing in or not. Erroneous DCF analysis in a market leads to an investor losing money, a disciplining effect that tends to concentrate the mind. Furthermore, investors' incentives to get DCF analysis "right" are strong. Do it right, you make money; do it wrong, you lose money. Contrast that to regulators' incentives where low-balling the cost of capital yields lower rates, but the negative effects are foisted off on the company and its shareholders. [Mind you, low rates issuing from an accurate assumption of the cost of capital are a good thing. Mistakenly low estimates of the cost of capital slowly desiccate the company you are regulating.]

Verizon, of course, has choices where it can allocate its capital, and this will provide the long-term discipline to the NH PUC and other regulators tempted to reach unrealistic cost of capital assumptions. Verizon will invest where it gets the best return. NH may not be that place.