Beware the Fat Boys Race

https://www.linkedin.com/feed/update/urn:li:activity:6357860046705086464

The RIIO2 Incentive Structure.

The next step in the RIIO2 process will be the publication by Ofgem in February of its proposals for the RIIO price control framework – setting out the overall approach to be adopted in RIIO going forwards. While it will be a consultation the expectation must be that some elements will be pretty firm. On stakeholder engagement the companies need clarity to start recruiting panels which will take some time. And on other issues there will be another level of detail to be worked through before the final framework is nailed in the summer which means that Ofgem will need to go nap on its broad approach sooner rather than later.

For some issues that doesn’t matter too much but others go to the heart of the regulatory regime and are vital to get right. One of these and perhaps the most contentious of the framework issues is what to do about perceived high rates of return.

Ofgem has been embarrassed by the high returns many of the network companies are earning and consumer groups are demanding that average returns be capped. The ideas that Ofgem has floated in working groups on the topic include “anchoring returns” so the average ex-post return equals the ex-ante allowance; a soft cap and floor so high returns can only be earned by companies performing high on customer service or innovation; or a reopener to attempt to make any necessary adjustments mid period.

As a way to avoid embarrassing headline figures these may all seem attractive but Ofgem needs to ensure that it thinks through the perverse incentives that each of these approaches might create. Price control regulation is all about placing incentives on companies to behave in ways that are aligned with the consumer interest and so it’s vital to look at the different models through that lens.

The current RIIO model provides strong incentives for companies to cut costs but seeks to ensure that is through genuine efficiency (rather than letting the network go to rack and ruin) by requiring the companies to deliver certain outputs and providing incentives around reliability and customer service, for example.

The idea of “anchored return” – setting an average return with individual performance determining the spread – or a shift to focus on “relative” performance - would undermine those incentives. Companies would no longer know what they could expect to earn as a return from actions to improve efficiency or customer experience – which means they don’t know if it’s worth investing. Their performance depends not on what they do but what everyone else does. That may happen in competitive markets but it’s not a good basis for regulatory incentives.

With a small number of players (and several in common ownership) it’s not hard for them to reach an understanding that it’s not worth trying too hard because they are guaranteed between them the average return.

One of the shifts made in RIIO was to set rewards for customer satisfaction based on performance against a fixed externally determined (cross sector) benchmark rather than relative performance which had been the basis before. Simply doing better than other networks should not be a basis for getting a reward – there should be no prizes for winning the fat boys race. This element of the RIIO ED1 price control was challenged by Centrica at the CMA and was upheld.

The idea of “anchoring returns” is much closer to what happens in the US with rate of return regulation (albeit with a chimera of competition). Rate of return regulation is well known for having failed to deliver the efficiency gains and customer service improvements that we have seen here.

The alternative, as now, is that if all companies are performing well (against some absolute benchmark) then they should all be able to earn high returns. That really shouldn’t be a problem if they are all genuinely high performing. In practice they probably won’t be – but they could be.

The problem as Ofgem highlights is where an element of that outperformance is simply the result of assumptions made at the time of the price control being wrong. This can be external economic factors such as real price effects where Ofgem will probably want to look at what more can be done with indexation to avoid windfall gains and losses. The other factor is the intractable issue across price controls in all sectors of information asymmetry and the ability of companies to over-egg what they say they need. It’s hard even with hindsight to tell what was over-egging and what was the effects of the incentives working. But Ofgem, not unreasonably, take the view that the bigger the level of outperformance the more likely it is that some of it came from over-egging.

That’s what leads them to ideas around a cap and collar approach. If individual companies have an upper and lower bound on what they can earn (and it does need to be symmetric to be fair) then that will dull incentives slightly but they will at least be able to make rational calculations of what it’s worth investing in and what it’s not, which they can’t do with a relative regime.

The other challenge around legitimacy, brought into sharp relief by the latest RIIO annual reports on GD1, is the fact that it is currently possible to fail to meet targets on customer satisfaction and still earn high returns as Cadent have done (albeit not as high as other GDNs). The incentives for customer service yield only a small return compared to what can be achieved through efficiency savings. That’s not necessarily wrong in itself – the incentives for customer service should reflect the value that customers place on them to avoid gold plating – but the end result runs contrary to what RIIO is trying to achieve. One solution as Ofgem have suggested would be to set certain service standards that the companies must meet if they are to earn “super-rewards” (ie above a certain level).

The other route into this whole area and which Ofgem don’t seem to be exploring is to look at the sharing factor (the proportion of any outperformance that gets returned to consumers). Consumers would be much less concerned about the companies making high returns if they were getting a good share of any efficiency savings themselves. Again one of the features of RIIO was always intended to be that consumers saw a benefit straight away from efficiency savings by the company – rather than having to wait for that to be reflected in the next price control as was the case under RPI-X. However, over the years, the share that consumers get has fallen and is further muddied by adjustments for corporation tax.

There clearly is a case for resetting the sharing factor generally to give consumers a larger and immediate benefit (but without going so far that it undermines any incentive for the companies to push for such savings). However more could be done including having tiered sharing factors with higher sharing factors for higher returns. This would be an improvement on a straight cap on returns as it would avoid a cliff edge for companies and provide some continuing incentive for efficiency and customer service even at higher levels of returns.

Time is running out for Ofgem to clarify its thinking in this important area. It’s vital that it avoids a return to the fat boys race, that it ensures the regime it comes up with provides the right incentives for companies and delivers benefits for consumers. A focus on the share of benefits that consumers gets would seem a better way of meeting those goals than any of the approaches that Ofgem currently seem to have on the table.