This article was originally supposed to be about the Fair Hydro plan, and in particular the regulatory failures that led to the political need for it.
As so often happens, events overtook my plan. While I was looking at the empirical data explaining that regulatory failure, the Ontario Energy Board did some things that took me in a new direction.
The OEB is more and more exhibiting the signs of regulatory capture.
Note that I said the OEB, by which I mean the regulator as an entity. The adjudicative panels at the OEB, i.e. the Board members, have often shown strong independence. On the other hand, senior management of the OEB – those who are setting policy and regulatory direction – show bias in favour of the views of the industry they regulate.
Further, instead of moving to head off that problem, the OEB is going in the opposite direction, proposing to increase the ability of utilities to influence OEB policies and directions, and decrease the ability of customers – those who pay for everything – to provide balance and perspective.
What is Regulatory Capture?
Now, before you get all hot and bothered, and want to storm the ramparts, let’s look more closely at regulatory capture. “Bias” sounds really bad. (Or maybe really, really bad.) It’s actually not quite as bad as it first appears.
Still bad. Just not AS bad.
Regulatory capture is sometimes compared to Stockholm syndrome, but it’s not quite the same.
Usually attributed to Nobel laureate George Stigler, the term “regulatory capture” refers to the situation in which the regulator’s independence has been compromised by the influence of the industry it regulates. Stigler’s extreme view was that capture was inevitable, because regulation exists primarily to benefit the private interests being regulated, such as utilities. As decades of debate over the subject have evolved the concept, most theorists no longer agree with Stigler that private interests will necessarily crowd out the public interest in every case. However, most also agree that it is a common problem for regulators.
In vastly oversimplified terms, there are three types of regulatory capture.
The classic definition describes a full or complete capture, in which the industry regulated basically dictates what the regulator does and says. This can be because of bribery or other corruption, for example, or just complete subservience. This has happened to many regulators in the past. Some purists consider this the only true regulatory capture. Others, like Scott Hempling, believe that corruption is different from capture, and should not be included under the same umbrella.
There isn’t the remotest evidence that the OEB has been captured in this extreme sense.
The other two types of capture are more ubiquitous, and are more relevant to the OEB situation.
Cultural capture occurs when the regulator and the regulated industry become so close that the independence of the regulator is undermined. Phillip Wallach, in an excellent article reviewing a book of essays, Preventing Regulatory Capture (Carpenter/Moss, ed.), describes cultural capture thus:
“The main mechanisms of this cultural capture will be: group identification, in which regulators come to identify themselves with regulated firms and their employees, sometimes because of the existence of revolving doors; deference to high-status regulated executives; and deference to those with whom one has face-to-face relationships, because of empathy or the desire to avoid conflict.”
This is not really surprising. Regulation, including economic regulation of a monopoly through an administrative tribunal, is based on the concept that experts in a particular field are the best people to understand the intricacies of regulating that field. What that means is that the talent pool the regulator draws from is often the very entities it regulates. What’s more, the social circles of the regulated and the regulator will overlap, sometimes extensively.
None of this is inherently bad. In fact, often former utility staff who work for the regulator are the best people to do the regulating. They really know what it’s like inside a utility. Anyone who has argued a case before an adjudicator with a utility background knows that they can be very effective precisely because they don’t wear rose-coloured glasses. They’ve seen the good, and the bad, up close.
Closeness between the regulator and the regulated is therefore quite normal, even expected. It is only when it morphs into a set of blinders that it becomes a problem.
The third type of capture is informational capture, sometimes called “weak capture”.
Regulators make decisions and develop policies based on information. Regulated industries have an asymmetry of information, in that the utilities or other regulated entities have most of the information. As a result, the information the regulator gets is often from those it regulates. Informational capture arises when the regulator ceases to be sufficiently skeptical about the information it is getting, and reduces its emphasis on obtaining information, verification and input from other sources.
Information capture is sometimes seen as driven by expertise. It is not just that the utilities have control of most of the relevant information. They are also experts in the field, so the information they deliver, and how they characterize it, is taken – consciously or subconsciously – as reliable expert analysis. That is, they not only control the information, but how and when it is delivered. Their lack of independence is ignored.
The OEB exhibits some of the indicia of the second and third types of regulatory capture.
What is the Evidence of Capture of the OEB? – Part 1: Rates
Start at the empirical level. On the electricity side the regulated utilities are for the most part getting what they want, with some resistance from adjudicators (Board members) in litigated cases, but with minimal resistance from the senior management of the OEB in development of policies and rules.
(Note that this is not as apparently true on the gas side. The reasons for that are complex, and more suited to a separate article.)
The starting point for any economic regulator, whose central job is to regulate rates, is…(wait for it)…rates.
Of course, as has often been noted the actual distribution revenue per customer in Ontario from 2005 to 2016 (the range of years for which we have good, and comparable, information) has increased by 25.6%, from $548.60 per customer in 2005 to $672.26 in 2016. That is a compound annual growth rate (CAGR) of 2.1% per year. Inflation over the same period (GDP IPI FDD) was 23.17%, or 1.9% per year, so the rate increases are apparently a little high, but not outrageous.
There are two reasons why this initial conclusion is misleading.
First, distribution revenue per customer is only a good proxy for rates if the customer mix remains the same, and rates have increased roughly the same for all classes. Neither is true.
Residential customers, the ones with the lowest revenues, increased from 89% to 90% of total customers, meaning the overall increase is slightly understated. (Comparisons by rate class are from 2006 to 2016, because 2005 public data does not include this metric.) On the other hand, the rate of increase of revenues from residential customers increased much more rapidly than for non-residential. For residential, increases in distribution bills were at a rate of 4.66% per year CAGR. This is despite the fact that residential average use dropped by 13.6%, from about 818 kWh per month to 707 kWh per month. On a unit cost basis, the cost to deliver one kWh to someone’s home has increased from 2.78 cents to 5.08 cents, an increase of 82.7%, or 6.22% per year CAGR, more than three times the rate of inflation.
When residential customers say they are using less, and paying more, they are not lying.
For non-residential customers, there was a decrease in distribution bills over that same period of about 1.89% per year CAGR. Average use by business customers has only dropped 2.9% over that period, since use is driven more by economic activity than by efficiency. However, OEB cost allocation policies have resulted in increases in the share of costs allocated to the residential customers, largely because the business customers had been paying more than their fair share (likely a legacy of the former municipal influence on rates).
The other reason the 2.1% overall CAGR is misleading is that it came during a period when external factors were acting to push distributor costs down by substantial amounts:
- Interest rates dropped from an average of 8.57% in 2005 to less than half that in 2016. When factored into both debt and equity costs, the impact of this reduction on the highly capital intensive utilities is around $650 million, or about 19% reduction in overall costs.
- An accounting change meant that most distributors had a drop in their depreciation rates, with an industry-wide impact of about $400 million, i.e. around 10% reduction in costs.
- The distribution sector in 2005 was characterized by a large number of smaller distributors, with many built-in inefficiencies. Mergers and consolidations have resulted in substantial savings in many areas, most notably for the Powerstream, Horizon, Entegrus and Toronto LDCs. This is not possible to estimate properly, but is certainly in the $100s of millions annually.
Some part of these impacts, as well as others like the declining tax rates over the same period, would be captured in the inflation comparison. Some, on the other hand, would not. It is reasonable to guess that, but for these listed impacts, the rate increases allowed by the OEB over that period totalled more than 4% CAGR, i.e. more than twice the rate of inflation.
Protecting the customer is job one for an economic regulator like the OEB. In a decade in which known costs for distributors dropped by 30-40%, more than a billion dollars, it is not credible to say that allowing rate increases in excess of inflation amounted to “protecting the customer”.
Is the same true for transmission and for prescribed generation (OPG), the other areas regulated by the OEB?
Transmission rates are the easiest. I couldn’t find 2005 and 2006 Uniform Transmission Rates (I know, just lazy, really), but since 2007 the rates have increased 45%, i.e. an average of 4.25% per year.
Of course, the decline in interest rates benefitted the transmission companies in the same manner as the distribution companies, but the benefits of changing accounting rules for depreciation, and M&A activity, did not affect them significantly. Since the interest rate increase is partially captured in inflation, it is no surprise that the 4.25% per year rate increase is similar to the real underlying distribution rate increase. The adjusted rate increase for transmission is likely somewhat higher, perhaps as much as 5% CAGR, but even without adjusting for the interest rate decline it would still be very high.
Now turn to OPG. In 2007, before its rates were regulated by the OEB, the nuclear rate was $49.50 per MWh, and the hydroelectric rate was $33.00 per MWh. The rates in 2016 were $72.30 per MWh and $44.55 per MWh respectively, increases of 46% and 35%. On a CAGR basis, they are 4.3% and 3.4% per year. This is despite sometimes aggressive adjudicated results from Board panels, including one cutback that went all the way to the Supreme Court of Canada for a decision.
This does not include the future increases currently proposed by OPG, which would see the nuclear rate increase by +10% per year for the next ten years, and the hydroelectric rate by 1.5% (although they would be blended).
Like transmission, OPG has benefitted from declining interest rates, but not really from M&A or depreciation changes.
All of this is to say that, for the more than two-thirds of the typical electricity bill that the OEB regulates, it has allowed increases that are effectively more than double the rate of inflation for the last decade.
Thus, on the most basic test – rates – the OEB likely fails.
This is not direct evidence of regulatory capture, of course, but it does appear that “someone’s got some ‘splainin’ to do”, as they say. We certainly have to dig a little deeper.
Unfortunately, digging doesn’t produce better results.
What is the Evidence of Capture of the OEB? – Part 2: The Big Build
I’ve written earlier this year about the Big Build, in which the OEB has allowed wires companies to expand their capital infrastructure by massive amounts over the last decade, and continues to do so today. Capital spending in excess of apparent requirements is at least $6 billion in distribution alone, and transmission is also substantial. In fairness, the capital spend of OPG probably cannot be laid at the OEB’s doorstep. Their hands have been tied on that one.
Now, this is not another opportunity for me to rant about capital spending, and the huge future debt being piled on customers over the last decade.
I’m tempted. But no.
What this is about is regulatory capture, so the Big Build is interesting in this context mostly for demonstrating to whom the OEB listens when it is developing and implementing policies.
The Big Build started with the utilities saying they needed more money for capital. The utilities didn’t come to the Board with empirical evidence that they needed to ramp up their capital spending. Oh, no. They, and their trade associations, and their favourite consultants, just kept repeating it as a mantra, over and over again. We need more. No evidence. Just repetition.
The OEB is chock full of economists, so they are very familiar with the Averch-Johnson effect, first introduced in 1962, and today a still debated but generally accepted basic paradigm in utility regulation. What Professors Averch and Johnson posited, and demonstrated, was that companies that are regulated based on a rate of return tend to over-build. Sometimes it is conscious, and sometimes subconscious, but in fact they will chase profits. Since profits come from rate base, they will build as much rate base as they possibly can.
Further, as subsequent research has shown, if the return on capital for the regulated entity is high relative to what the shareholder can earn elsewhere (as is the case in the Ontario wires sector, where the shareholders are mostly municipalities), the tendency to over-build is increased.
Since this built-in bias was well-known to the OEB when they were hearing all this sturm und drang (what, would you have preferred “weeping and gnashing of teeth”?) about capital needs, they should have been skeptical. When the customer groups asked to see empirical evidence of increasing needs, a truly independent regulator would have been leading the charge to get that evidence. “Why do you need more money now?” they should have said.
Instead, the OEB asked a different question: “How do you want to spend this extra money?” Rather than look analytically at overall capital needs, the OEB asked for Distribution System Plans, otherwise known as wish lists.
Now, in principle there is nothing wrong with Distribution System Plans. Every distributor should have one. (In fact, they shouldn’t have to be told, but that’s a whole other story.) What’s wrong is that the OEB started from the premise that if the utilities said they had to boost their capital spending, that must be correct. They are the ones that know their systems, right?
The OEB’s policy response to the utility “we need more capital” message has consistently been to believe the message, and implement policies (incremental rates to fund capital spending, anyone?) that provide more money for that purpose.
When a regulator believes what the regulated companies say, despite no supporting evidence, that is a classic indicator of regulatory capture. That is particularly true in a subject area where, according to well-accepted regulatory principles, it is likely that the utilities will have an underlying bias, and thus may not be objective.
The likeliest reason for the Big Build was regulatory capture.
What is the Evidence of Capture of the OEB? – Part 3: Policy Development
Less empirical, but still illuminating, is how the OEB goes about developing new policies. You would expect that a regulator that believes “outcomes” are the most important goal (they are) would start by talking to customers. The opposite is in fact true. They start by talking to the utilities.
Just a couple of examples will demonstrate this.
At a recent OEB meeting, a senior executive of the OEB told us that residential customers were not concerned about the Board’s recent change to all fixed distribution rates.
I was surprised, so I asked where that information came from. After some hesitation, we were told “We talked to the EDA [Electricity Distributors Association], who speak for their customers.”
When I noted that speaking to the distributors – who asked for this change in the first place – is not the same as speaking to the customers, the executive said “Oh, and we noted that there were no complaints at the Community Days around the province.”
I said “Did you go to any of the Community Days?” When the response was no, I explained that I went to several of them, and the most common complaint of customers was the impact of the new all fixed rates on tenants and seniors. You couldn’t miss it. Some of those customers were very upset about it.
Deer caught in the headlights. They simply didn’t know what the customers thought, and they assumed that the utilities were communicating on behalf of the customers.
Another example arises out of a similar situation. The Board is proposing to change the structure of distribution rates for business customers, in order to deal with behind the meter generation and the potential for erosion of distribution revenues. To develop the new policy, they consulted with everyone, but they didn’t like the answers proposed. Then, they spoke to the distributors, reached a consensus with them, and approved the new policy internally.
Then they went to talk to the customers to tell them what they planned.
Turns out, what they had already decided (before talking to the customers) was in fact a bad idea. It completely missed the fact that solar behind the meter – really the main point of the change – operates at a 15-20% capacity factor, so that their proposed new charges would make rooftop solar completely uneconomic for businesses. (Sorry, Mr. Energy Minister.) It also completely missed the fact that it would catch (and penalize) peak-shaving by larger users with gas cogeneration, something that is specifically incented by the province under the Industrial Conservation Initiative, and desperately needed. (Sorry again, Mr. Energy Minister.)
How do you establish a new policy on rates, involving hundreds of millions of dollars a year of additional charges, without at least investigating how customers will be affected by the policy? The answer is, if you are too eager to believe what the regulated utilities tell you, that’s what can happen.
And that sounds a lot like regulatory capture.
A much bigger example is changes in the last couple of years to the Board’s M&A policy (called MAADs, an increasingly appropriate acronym).
Under the old policy, the Board tried to encourage consolidation amongst electricity distributors by offering a five year grace period (a rate freeze, in effect) after a merger. The theory was that mergers should produce substantial savings, and for the first five years the savings would go to cover transaction costs, and then provide a financial reward to the shareholders. Customers would benefit right away from the freeze, but would not participate in the main benefits from the merger until after five years. A number of successful mergers took place on this basis.
The utilities, however, complained that their incentive was too small to promote consolidation, even with pressure from the province on the utilities to merge. The utilities wanted more financial incentives to go to them, and less benefits to their customers.
The OEB complied. With limited discussion (except with the utilities, of course), the OEB said that the period over which the benefits would go to the utility shareholders would be increased to ten years. Not only that, but there would no longer be a rate freeze. Now, the utilities would continue to get rate increases based on a formula, so the customers would get nothing at all in that period. Further, if the utilities wanted to spend extra money on capital, they could get that from the customers too, including the additional profits that come from investing in more capital. All of this, without any reference to the actual costs to run the merged utility, which were expected to decline.
The result? Mergers like the recent Alectra combination, which created the second largest distributor in the province. The shareholders of the merging utilities are expected to be rewarded with an extra $425 million, starting essentially right away. The customers will only start to get benefits from the merger ten years later (so they say). Meanwhile, they will suffer through more rate increases in excess of inflation, year after year.
The new MAADs policy is just another example of the OEB believing the utilities, and fashioning policies to respond to what the utilities want. At the expense of the customers.
There are lots of other examples, some of them obvious, some of them more esoteric. A pattern emerges. The OEB either responds to complaints of utilities with a new benefit for them, or it thinks up something on its own and checks with the utilities to make sure it works for them.
It would be easy to overstate this, and that would not be fair. The OEB carries out many consultations, including many with customer groups. Customer groups often – usually, in fact – have sufficient resources to participate effectively, and do so. In many of those processes, the OEB takes note of customer concerns and adjusts policies accordingly.
But there are a disconcerting number of examples where it is apparent that what the utilities want in a policy is what will happen.
Preventing Regulatory Capture
There has been an enormous amount written, by academics and others, on the ways to prevent regulatory capture. This was especially the case after the financial crisis of 2008-9, which many ascribed to regulatory failures in the financial industry rooted in regulatory capture.
The ways to fight regulatory capture can be loosely divided into two categories: structure and process.
Scott Hempling has written extensively about methods to avoid regulatory capture by shoring up the structure of the regulator. For example, he talks about the importance of empowering the personnel of the regulator, including in particular ongoing professional development, so that professionalism and personal responsibility generate a natural barrier to regulatory capture.
Hempling also talks about the regulator ensuring that it frames the agenda, and that its regulatory stance is proactive rather than simply responsive to utility applications. He notes that regulators who are the most independent are the ones that get out in front of issues, and then allocate sufficient internal resources that they can control the agenda right through policy development and implementation.
For the most part, the OEB is already doing the things that Hempling suggests. While some have complained that OEB staff are not sufficiently empowered (and that is probably true to a certain extent), on balance the structural defences are in place at the OEB.
Others who talk about structural solutions emphasize clarity in the legislation. Still others have suggested that the legislature should actually appoint an officer to oversee the independence of regulatory bodies, much like an Auditor-General or an Ombudsman, but directed to administrative tribunals.
These are of only marginal relevance to the OEB.
On the process side, the primary theme is systems or approaches to ensure customer voices are heard more clearly and more often. Professor Daniel Schwarcz of the University of Minnesota, an insurance regulation specialist, believes that the key ways that a regulator can fight regulatory capture are a) encouraging and resourcing independent customer representatives, and/or b) appointing and empowering independent consumer advocates. Many other commentators and regulatory specialists, in a diverse range of regulatory areas, echo these suggestions.
Professor Elizabeth Madill of Stanford suggests something similar in the realm of judicial review. If access to the courts by customers and other interest groups is expanded and facilitated, she says, that can act as a check on regulatory capture of agencies.
Sadly, the OEB is moving in the opposite direction. In a new policy, just announced, the OEB will limit the voices of customers even more than in the past. Further, those voices will be focused on the bigger, litigated cases, just the ones where regulatory capture has been less of a problem (in part because of the participation of customer groups).
(This new policy, which has not been the subject of any consultation with customers, will be dealt with in a future article.)
Thus, it can be expected that regulatory capture will, if anything, get worse at the OEB.
The OEB, like any other regulator, has good and bad aspects. On some key metrics, the OEB is not delivering as it should for the customers, and the primary reason for that is probably undue influence by the regulated utilities, i.e. regulatory capture.
The OEB and the provincial government may want to take note of what has happened to the National Energy Board (NEB) over the past decade. As the Federal Government’s NEB Modernization Panel recently explained in their Final Report, many Canadians have lost confidence in the NEB, precisely because they feel it has been captured by the industry it regulates. The road back for the NEB to rebuilding the trust of the public, particularly those who are most directly impacted by its decisions, could be a long uphill struggle. Realistically, it may never happen.
Better the OEB fix the nascent problem it has now, while it is still not too late, as opposed to exacerbating it, perhaps beyond repair.
- Jay Shepherd, October 4, 2017