Petrol profits

The Commerce Commission's paper on measuring profitability in the petrol business didn't formally call for submissions but if people had any views they could send them in by close of play today. Here are a few of mine.

The first thing is that, somewhat surprisingly, firms - perhaps many of them - can be earning persistent 'excess' profits even in workably competitive markets. The graph below, which is the absolutely standard 'demand curve crosses the supply curve' picture, shows how it happens.

We know that producer A would be earning its weighted average cost of capital at price PA because it is willing to offer to supply at that price, and it wouldn't if it wasn't. At the higher market price of Pe it is making above-normal-WACC returns.

Or as a very good text book* says, when you have upward sloping supply curves, as in my view you often will, "the market price in equilibrium will normally be determined by the level of cost of the higher-cost producers - the 'marginal producers' - who will make only a 'normal' profit (the market price only just covers their costs) ... At the market price, the lower-cost suppliers will make a healthy margin above cost".

So the ComCom paper is bang on when it says that "Even where competition is effective, the profitability of some suppliers may be above normal or competitive levels" (para 37) and that "Analysis of profitability by itself may not distinguish whether higher than competitive levels of profits are due to cost advantages [as with producer A in the graph], the exercise of market power, or a mix of both".

It follows that the focus of the profitability analysis should be firmly on the returns being earned by the marginal producer at Pe and not on intra-marginal producers like A. This was the approach correctly adopted in last year's first report from our Electricity Price Review (write-up here, with links to the review). It looked at whether prices were in line with the costs of the next (the 'marginal') generator commissioned.

The logic was

Contract prices that were above costs on a sustained basis would suggest weak competition among generators, and that the entry, or threatened entry, of new generators was not restraining prices. On the other hand, prices that were well below costs on a sustained basis would suggest looming problems with reliability of supply because new investment would not be able to keep pace with demand. The comparison suggests competition has been effective in restraining prices. Figure 14 shows how wholesale prices have moved broadly in line with the cost of adding more capacity. Importantly, there is no evidence contract prices have been above costs on a sustained basis in recent years (p32)

The other main point I'd like to make is that the ComCom paper currently places some reliance on where analysis of 'gross margins' might take you in any assessment of profitability. I'd say that the answer is, almost nowhere. They may have accounting or commercial relevance, but for all the reasons mentioned in para 68 of the paper they are indeed "an incomplete measure of performance". From an economic perspective gross margins tell you very little, although they might (in a very homogeneous industry) give some limited insight into productive efficiency. In particular there is no way of telling whether any particular level of gross margins is "too high"

I appreciate that in a world of limited and non-standardised industry data, ComCom is going to have to scrabble for whatever indicators, however indirect, are available to hand. But I'd downplay the gross margins route, and put more reliance on estimates of return on capital employed or return on equity (ROE), which in a market economy are the numbers that matter from an allocative efficiency point of view.

Two final small points.

In para 93 ComCom says that it will consider as an indicator of profitability "The returns being achieved on recent and proposed investment both by new entrants, and by existing participants expanding their operations, in the retail fuel markets ... we would expect returns on more recent investment to approximate the cost of capital if competition is workable and effective", which is very much along the lines of the point I made above about the profit conditions of the marginal producer. The only gloss I'd add is that, as ComCom looks at recent or proposed investments, it should be wary of the 'hurdle' rates companies tend to use to assess the profitability of investments (the projects have to have an internal rate of return that beats some minimum 'hurdle' level).

While generally it's very useful to examine internal company thinking at the time, the evidence is that hurdle rates are not good sightings of what the investing company thinks is its true WACC or ROE. The hurdle rate is typically well north of that, as companies tend to use hurdle rates to filter out overoptimistic managerial gaming of the investment budget.

And if the focus is going to be on ROE (as it ought), Stats already has some estimates of petrol company ROE in its Business Performance Benchmarker tool. Here for example are ROEs by size of petrol station. No idea of the basis of the calculations, but on the adage that if all else fails, read the instructions ...

* Gunnar Niels, Helen Jenkins, James Kavanagh, Economics for Competition Lawyers, 2nd edition, Oxford University Press 2016, p10