Draft Response To Regulator's Consultation on TOU Pricing

2010-11 Time-of-Use Consultation (EB-2010-0364)

I’ve prepared some remarks as I consider a response to the Ontario Energy Board (OEB) review of the Retail Pricing Plan (RPP), and specifically the Time Of Use (TOU) pricing. 
The Board notes, in Kirsten Walli’s letter of October 18,2010, that now that the board has directed all residential customers be moved to TOU pricing, and many customers already have been moved to TOU pricing, it’s a good time to review TOU pricing.   The December 6th, 2010 letter provides details on the consultation.  That letter notes that there is an intention “to ensure that the design of TOU prices is fair and meets the objective of ultimately reducing overall power system costs,” and Appendix A of the document identifies a number of specific issues that give some direction for the arguments.

I think the time would either be before they installed 4 million smart meters (or whatever the figure is), or after they have some real-world data to mine on the existing 750,000 customers.  But now is the time the Board is asking, so I’ll try to respond if only to utilize my amateur’s knowledge of smart meters, and dynamic pricing schemes.  Comments, either on this site or by e-mail, would be appreciated.

Before addressing the specific issues identified by the Board staff, there are some issues that should be first addressed related to framing the discussion.  The Board has contracted the Brattle Group to facilitate the discussion, and that group prepared both”Assessing Ontario’s Regulated Price Plan: A White Paper,” and a related Powerpoint presentation.   The Brattle Group is fully aware of dynamic pricing alternatives to TOU that deliver far greater reductions in peak demand (such as Super Peak TOU, CPP, VPP, ant RTP charted on page 6 of this The Brattle Group document for the Smart Grid Latin America Forum 2010). 

Specific to Ontario, there are other statements from the consultant I would reject off hand.  While the theory appears broadly accepted that managing peak demand can save the costs of constructing, or maintaining, additional generation capacity, there is little indication that would lead to fewer emissions – of green-house gases or anything else – in Ontario.  The OEB would not be protecting the broad consumer interests of Ontarians if they judged the value of a reduction in peak demand against any value other than the cost of generating capacity.    Regardless, the OEB should be aware, as it’s consultant is, that TOU is not the best Demand Side Management (DSM) mechanism to reduce peak consumption.

The alternative rates discussed by the Brattle Group are of no use, because all build off of the first which is “The existing TOU with the addition and reallocation of expected wind and solar GA costs to the peak period (price ratio = 2.7-to-1).”  That violates the OEB’s starting point of ensuring that the TOU prices are fair.

Wind output in Ontario is not only well known, due to the excellent data maintained at the IESO, but it really hasn’t varied all that much from initial studies on the potential of wind supply in Ontario back in 2005.  Wind is expected to produce at about an annual capacity factor of a little under 30%, with a summer (June, July, August) capacity factor around 17%.  My interpretation of the IESO data is that it has been doing that, and this graph shows wind capacity factor by month are as they were expected to be:

The latest 18-month outlook from the IESO lists the forecast capability at summer peak as 178MW for an installed capacity of 1235 MW of wind, which is realistically optimistic.  Last years’ peak had 1085MW of capacity producing 72MW of output (half the IESO’s hoped for capacity factor this year).    It is not fair to put the cost of wind onto summer peak hours.  I would point out how ‘fair’ in an economic sense is impacted through increased wind and solar penetration.  Because a functioning market requires price to fluctuate as a signal both to producers to change production levels, and consumers to adjust consumption levels, the ‘fair’ price of electricity will be increasingly dependent on how windy, and sunny, it is.  The price being dependent on supply fluctuations was made clear both in Texas and Alberta this winter as unexpected plant outages (coal, and gas) led to huge spikes in market prices, both during cold spells.  Gord Miller noted, on the Agenda last fall, an Ontario record HOEP of $1891.14/MWh – in arguing for TOU pricing.  That price occurred Wednesday, February 18, 2009, at noon – when Ontario demand was an unremarkable 19413MW.  That is not now an on-peak time, nor is it proposed to be.   The economic ‘fair’ price cannot be scheduled.

If fair is not presented in an economic sense, it needs to be defined.  That should be the first step of the OEB’s pursuit – and I think it might become the last as the groundwork of the current consultation seem a step back from the integrity of the RPP process that established the current prices.

Specific Comments to Appendix A issues

Are the current three price periods still appropriate?

3 prices are not appropriate.  The OEB pricing, if it must be done on a TOU basis, should be aligned with the simplified On-Peak and Off-Peak pricing currently in use in the IESO reporting.  The OEB is currently paying a private consulting company for worse data than could be mined from both the IESO and Hydro One and other utilities (particularly Hydro One).  Should political optics require maintaining TOU pricing, co-ordination with other public bodies to simplify data exchange should be a priority.

There is also no discernable difference, in Ontario consumption, between on-peak and mid-peak hours in the winter – the IESO methodology seems more tested, and reflective of market demands.

Is the current seasonal structure appropriate on a go forward basis?

The current seasonal structure is inappropriate.  On the demand side, peak usage periods are related to temperature, and meal times.  On the supply side the supply mix would, if not overbuilt, cause periods of over-supply (increasingly true as illustrated by Surplus Baseload Generation), and shortages of supply.  The supply argument is for RTP (as noted by Direct Energy), not TOU.  There need not be any great cost associated with changing the approach.  The OEB has consistently set RPP levels that have coincided very closely with the wholesale rates during the same period.  The problem in the retailer market has been the global adjustment – which is also an enormous impediment to RTP being meaningful to the wholesale market – which I believe is now larger than the residential market being discussed.  There is an implication that the destruction of the HOEP pricing mechanism through the excessive contracted supply can be rectified through consumer pricing tricks.  It cannot. 

Regardless, Direct Energy has submitted comments they can provide tools to facilitate RTP for their customers.

Given that the Ontario electricity system is summer peaking, would it make sense to adopt a structure which specifically addresses the summer peak?

Yes – and no. Yes, it would make sense to allow a functioning market to set a higher price when calling for reduced demand – and in that the highest peak being reduced should, in theory, reduce generation capacity needs, the restriction to the summer peak seems sensible.  But we only have higher summer peaks now because we have fewer households heating with electricity.  A spike in natural gas prices could change that entire dynamic.  Market mechanisms can adjust to such changes.

or

No, it would not make sense because the total system peak is only one issue the variable pricing might address.  The other is smoothing of the usage pattern over the course of a day.  I believe there are emissions implications in the amount of ramping up, and down, of supply – and there are definitely supply mix issues.  I have graphed the average daily variance in Ontario Demand (the maximum less the minimum hourly use reported in the IESO data), by month.  The desire to level off the consumption pattern in November and December is similar to the need to do so for the summer months (although hot summers see the greatest variance as well as the overall greatest demand):

Are the RPP Manual target ratios of 1:2:3 still appropriate?

The Board is aware the 1:2:3 rations are not appropriate.  They started with those prices, and have moved away from them.  The flattening of TOU prices is matched at the wholesale level reflected in the HOEP, and, as stated previously, the existing RPP planning used by the Board has produced realistic rates.  The narrowness in the rates fairly reflects a dysfunctional market (resulting from both demand destruction and supply mix/over-procurement issues).  This chart illustrates the averaged HOEP rates for the different peak periods (as per my calculations from base data at the IESO site):

The Board should continue emphasizing RPP supply cost recovery as the primary objective, and ignore the ratios.

The Board should not use GA cost assignments to enhance the TOU ratios regardless of “cost causality”

I conclude with a reminder of what fairness has, historically, meant to us.  It has meant that the little guy should not be singled out to pay the brunt of costs for our mistakes.  It has meant a progressive pricing element (such as the lower price for the first 600/1000kWh - that does mean switching to TOU is a rate hike for those with usage under 900kWh/month).  The OPA should be able to locate studies by the Conservation Office (prior to it being rolled into the general OPA site) which studied the entire market at peak use times.  I recall that document stating at those times demand was about 1/3^rd commercial, 1/3^rd industrial, and 1/3^rd residential.  The claims of The Brattle Group that a 4% reduction would be 1064MW doesn’t appear to account for this – The RPP group may only account for 1/3^rd of Ontario demand at the peak, and that reduction would be about 350MW based on last year’s 25000MW peak.  This seems like a lot of work to discourage one coal generation unit from staying on standby when it’s hot.  I end with a quote from Severin Borenstein:   “...: in an electric system that must always stand ready to meet all demand at the retail price, the cost of a constant-price structure is the need to hold substantial capacity that is hardly ever used. But utilities optimize by building "peaker plants" for this purpose, capacity that has low capital cost and high operating cost. The social cost of holding idle capacity of this form turns out to be not as great as one might think.”