WACC-y

I don't understand WACC, the Weighted Average Cost of Capital.

No, don't look at me like that - I do understand the mechanics of the thing, and have had my fair share of meetings pinning down appropriate beta comparators, the extent of the market risk premium, Brennan-Lally tax adjustment, and due allowance for the curvature of the earth.

No, what I mean is, I don't understand WACC as a regulatory concept.

Bear with me. I can see WACC as an accounting concept. A firm's balance sheet is made up of  equity and debt, and each comes with a cost: what you have to pay the owner shareholders to invest their risk capital, and what you have to pay lenders to lend. And the overall cost of the whole balance sheet is the weighted average of the two costs, WACC. All good.

But you'll often see a regulated firm described as "earning its WACC" (with subtext, "and no more"). No, it isn't. The firm is earning its return on equity. Its lenders are earning their return on debt. Neither of them is earning WACC.

In fact I can't see why (with one potential exception) regulators take the interest they do in the cost of a firm's debt. The standard regulatory equation in rate of return regulation (certainly as it's been implemented in New Zealand under Part IV of the Commerce Act) is

Allowed revenue minus (efficient) opex minus (efficient) investment minus (economic) depreciation = allowed interest costs plus allowed return on equity

But why do regulators give a fig about the interest costs? They're a cost that the regulated firm has every incentive to minimise: it's paid away to the third-party suppliers of credit, not a return to the firm itself.

The standard formulation makes little sense to me. Why shouldn't the equation be squarely focused on the return to equity:

Allowed revenue minus (efficient) opex minus (efficient) investment minus (economic) depreciation minus actual interest costs = allowed return on equity (ROE)

It's more logical: the only return that matters in a market economy is the ROE. And the rejigged version of the equation is both simpler and less restrictive.

It's simpler, because at the moment rate of return regulation goes through a whole process determining the "appropriate" cost of borrowing for the regulated entity - typically by establishing what a company of similar credit standing in that industry would pay, for debt of a maturity equal to the period of regulation (often five years). It would be simpler just to write down what it actually paid, not mess about guessing what a firm just like it might have paid.

It's less restrictive, because it would not risk imposing unnecessary and possibly inefficient constraints on the maturity of the debt raised. Right now, there'll be many a corporate treasurer who reckons that we are at a cyclical low point in borrowing costs, and will be keen to lock in currently attractive prices for as long as possible. If they're right, paying a bit more than you'd pay today for five year debt in order to issue ten or fifteen year debt could well work out cheaper - maybe a lot cheaper - in the long run, benefiting consumers. Allowing the regulated entity only the five year cost could be counterproductive.

The current formulation also potentially inhibits other efficient approaches to borrowing. There are good rationales, for example, to match the maturity of debt to the working maturity of the assets they finance. In regulated industries, the assets tend to be very long lived indeed. What is the logic of not compensating the regulated entity for the actual cost of doing the sensible thing?

And many treasurers will want to avoid a concentration of debt maturities falling due around the same time: you don't want to be going around the money markets with a large begging bowl if it happens to be in the middle of the next GFC.  Rather, a prudent treasurer will have a mix of maturities: on average they might approximate to the five year maturity the regulator will allow, but equally they mightn't. Why impose a penalty (or subsidy) on what a prudent maturity mix actually costs?

A purely ROE-focused approach, one which drops determining an "appropriate" cost of debt, is a better way to go as a general rule, but I mentioned one possible exception. That's where the debt providers are associated parties. Let's suppose the regulated entity is owned by a private equity company. It could fund its "debt" from a financing vehicle in the group, and stream above-market "interest" payments effectively to itself. But in normal circumstances most companies borrow at arm's length from banks and the capital markets. A quick check that its funders are not interlinked, and in most cases you'd be done.

There is one aspect of debt that regulators should properly monitor, and that is excessive leverage. With an effectively guaranteed income, there could well be a moral hazard risk of the regulatee putting in $10 of equity and a squillion dollars of debt, juicing the ROE if all goes well and lumbering the bondholders and any operator-of-last resort if it doesn't. A maximum leverage ratio, or as a more market-oriented option, a minimum investment grade debt rating, might be a useful regulatory adjunct. But beyond that, leave it to the corporate treasurer to figure out the cheapest financing bundle.

A final benefit of an exclusively ROE-based approach is stopping some game-playing. Quoting the WACC that the regulator has allowed can be deceptive. The regulator can say, "See how effective we've been? We only allowed them 6%!". The regulatee can play the same game: "See how unfair they've been? They only allowed us 6%!". In both cases - particularly at today's interest rates - the WACC is low because the debt component is low. The regulator may not in fact be especially effective; the regulatee may not in fact be hard done by. It's only the ROE that can answer those questions.

Revisiting regulation

In a later-career bit of diversification, I've been lecturing, last week delivering an intensive three-day course at the University of Auckland Law School - "Economic regulation: principles and practice", a master's level programme also available for some undergraduate study paths.

It's been stimulating: the class was high calibre, motivated, and ready with questions for me and the three visiting speakers I'd lined up. Big, big hat tips to Andrew Riseley, General Counsel on the regulation side of the ComCom house, Diego Villalobos, Principal Economist of the same parish, and former Telco Commissioner and the big honcho on regulation and competition at Minter Ellison, Dr Ross Patterson.

I don't know whether bringing in visiting firemen is standard in academia. All I can report is that, having tried it earlier at Victoria on a business cycles course run with colleagues Adrian Slack and Viv Hall, it seems to go down well with the students. They get to see that the stuff the lecturer has been going on about is actually what is happening and being used out there in the real world, and hearing it said in another voice probably helps it all sink in. Plus it also gives them some feel for whether they'd like to get into that line of work themselves later on.

Along the way I discovered a newish (2017) book that I recommended to the students as their first go-to resource. It's Thomas Lambert's How to regulate:  a guide for policymakers, Cambridge University Press. If you haven't come across it, it's excellent. Lambert is a full professor at the University of Missouri Law School, but evidently caught the economics virus as part of his undergraduate philosophy degree, and is very comfortable in the crossover badlands between economics and law. He contributes to the interesting Truth on the Market competition/regulation blog (I sympathised with their somewhat plaintive 'About us' description, where they say "We hope you find some of our posts insightful, thought-provoking, or at least mildly interesting").

His book had the structure I wanted for the course - an explanation of why workably competitive markets are the ideal, followed by all the instances where they won't necessarily work as well as you'd want (externalities, market power, asymmetries of information and the like), with good examples of how they can crop up and what you might do about them. You can currently get the paperback at the ever reliable Book Depository for NZ$51.16 (postage included) but if the pennies are tight or you prefer e-books you'll find Amazon does a Kindle version for US$20.79.

As you assemble your thoughts for a course like this, you wonder what the big takeaways for the students ought to be. Mine? The primacy of workably competitive markets; hence and otherwise the need to make sure any diagnosis of "market failure" is well founded; matching problems with the appropriate regulatory responses and, within that, prioritising more market-friendly and lighter-handed solutions; and regularly reviewing the need for regulation.

On which latter score we look to be doing reasonably well. I was encouraged by the latest rollback from the telco folk. The Telecommunications Commissioner Stephen Gale and his team are recommending that resale of Spark's copper-based voice services doesn't need its collar felt any more: "competition has been established, is increasingly effective, and is no longer dependent on access to these services". Right on, lads.

Though I'm less encouraged by the proposed 'building blocks model' (BBM) regulation of Chorus's fibre lines. One of Ross Patterson's points was that wireless broadband will serve as an effective competitive discipline on fibre prices, and I'm inclined to his view. The case for regulation no longer looks compelling, let alone regulation along the heavy duty BBM model that seems to have become our default. Fortunately, as Diego explained, we have had the wit to introduce an element of incentive regulation into the BBM at least as it applies to electricity lines businesses.

We also went through the history of the regulatory pendulum: right out to the pro-regulation side through to at least the late Seventies (France was still nationalising banks as late as 1982 and New Zealand was in regulatory lockdown until 1984); right back to the pro-market side up to the GFC; and the more recent swing to reregulation.

One thing that occurred to me is that, while the zeitgeist is now pro-regulation, and we may not now get a chance to fix them, the high water mark of the deregulation decades still left many areas overregulated when the tide started to retreat again. This past weekend's social and mainstream media, for example, are full of the follies of Easter trading restrictions, and (as I've said before) in an era when government funds are tight and we apparently can't find the funds for needed infrastructure in Auckland and elsewhere, successive governments have elected to go on owning a bunch of dairy farms, a policy which has not the slightest shred of public policy rationale.

Finally, we had a bit of fun in the class with the Weighted Average Cost of Capital (not a sentence you ever thought you'd read). We played "guess the beta", beta (for those who aren't regulation tragics) being a parameter in WACC which attempts to capture whether a share is more volatile than the average share and which might therefore need to offer a higher return. Beta is defined as how much a share price goes up relative to changes in the share market as a whole: beta greater than one, it goes up (and down) more than the market, beta less than one, it doesn't do as well (or as badly) as the rest of the market.

So I showed them the betas for a few of the listed utility-style companies, based on the very useful financial data you can find at Yahoo! New Zealand's finance site. You have Chorus on 0.61, and the gentailers not far away: Meridian 0.71, Genesis 0.79. And I pointed out that the current beta in the default price/quality paths for the electricity lines businesses is 0.72. Same diff.

All good, and then I showed them some companies and asked them to guess their betas. The class generally made a good fist of the likes of Auckland Airport (1.17), Fletcher Building (1.26), and Sky City (1.41). The surprises, for them and for me when I was devising the mental exercise, were Port of Tauranga (an unusually low 0.48) and - for a company down the higher-tech end and, with its assembly line robotics, you'd imagine would be facing some leveraged exposure to world trade - Scott Technology's oddly low 0.67.

ComCom had a go a while back at pulling together the literature on what drives betas - it's here, on pp35-8 - but I can't help feeling that it's still a work in progress. You'd wonder if the betas are sometimes more driven by the fads of investors than by the inherent volatility of the firm's line of business. 'Value' stocks for example can have extended periods in the sun - right now, for example, surveys of fund managers show that steady-Eddie utility shares are all the rage, partly because of the current 'hunt for yield' - only to languish later when 'growth' stocks are the in thing. And we regularly see 'sectoral rotation', where you can't give tech shares away one day and can't get them for love or money the next.

So despite the WACC cost of capital equations that look cut and dried, there's still a greyness around appropriate rates of return. Even if it wasn't a good idea anyway for dynamic efficiency reasons, ComCom's practice of using things like the 67th percentile of an estimated WACC range is exactly the right thing to do as a guard against faux precision.

Speaking of rates of return, ComCom has just put out the latest couple of papers as part of its petrol market study, one on what they're minded to zero in on and another on measuring profitability. If you want to respond to either of these (and I'll likely rise to the bait on the profitability one) you've only got till May 7 to do it, so skates on. And if market studies in general are of interest, don't forget to sign up to their update mailing list at marketstudies@comcom.govt.nz.

Great expectations

The latest monthly business opinion survey from National Australia Bank had this interesting item.

The interesting bit, for me, was that top right hand box: the high "hurdle" rates of return that businesses typically require new investment projects to meet. There is a big puzzle here: the hurdle rates are well above the weighted average cost of capital (WACC) that finance theorists, and regulators, say should be adequate for companies to earn. The sectoral distribution of the hurdle rates makes more sense, and looks to be broadly in line with back-of-an-envelope guesstimates of the relative riskiness of the different sectors, but the levels of the hurdles looks remarkably large.

Turns out that this isn't peculiar to the sample of companies the NAB surveyed: it seems to be well nigh universal. In the latest issue of the Reserve Bank of Australia Bulletin, there's an article, 'Firms' Investment Decisions and Interest Rates', which confirms NAB's findings for Australia...

...and which also summarises international research that found exactly the same thing overseas. For example

Studies of firms overseas have found that they also use hurdle rates that are above their cost of capital. Jagannathan, Meier and Tarhan (2011) surveyed firms in the United States in 2003 and found that a typical firm used a hurdle rate several percentage points above its WACC. Brunzell, Liljeblom and Vaihekoski (2013) found a similar result for Nordic firms. Similarly, firms in other countries also appear to use hurdle rates that are not sensitive to the cost of capital

I think we can safely assume that New Zealand businesses are in the same boat (anyone aware of specific research on topic?)

We don't know exactly (or even approximately) why this happens. I've always thought that part of the explanation was a principal-agent problem: the CFO is bombarded with potential projects from ambitious executives with self-aggrandising projects, and needs some device that might help sort out the viable from the vanity (though a high hurdle rate will also have the downside of encouraging hitting fours and sixes at the expense of lower-risk steady accumulation of runs). But I'm also rather attracted to two other possible explanations canvassed in the RBA article: this one...

the level of the hurdle rate may be greater than the WACC if the potential investment has greater non-diversifiable risk than the overall operations of the firm

...and this one

managers might value the option to defer an investment until its expected net present value is greater. In the absence of more sophisticated analysis, using a hurdle rate in excess of the WACC may be a reasonable approach to account for this option value of waiting (McDonald 2000)

If I were still a regulator - and particularly a regulator looking at a sequence of projects rather than a company's overall rate of return -  I think I'd be somewhat perturbed about these results. For good reason, more enlightened regulators tend to err a little on the side of generosity when it comes to regulated WACCs, since for dynamic efficiency it is far better to slightly overcompensate than undercompensate. But when you see the prevalence of these high hurdle rates, well in excess of WACC, you wonder if there's something that the standard regulator's WACC calculation is missing.