Natural Resources Committee on June 5th, 2018

[Nick Whalen, Liberal, St. John's East, NL]

Do you have any opinion on whether the availability of energy data in the U.S. makes their market more or less competitive than Canada's? Do the entities that need the data have the data they need, the people who are trading at the levels that you're talking about?

[Greg Abbott]

That's a good question.

[Steve Lappin]

My sense would be the latter. Those who need that data and trade around it have a ton of detail. They're either subscribing or getting it naturally through their trading activity. That's just based on our view in the markets we deal with. They have access to all of that data, so I would be highly surprised if any of those involved in the marketing or trading were having to go to the government website to access that information. They'd have their own sources.

[Greg Abbott]

I also think there's a really competitive business in the aggregation part of the data world, so there's a ton of desktop providers and they really are trying to build that perfectly aggregated collection of data. There are a lot of people in our business for sure, both in Canada and the United States, and almost every single one of them has some sort of subscription to an aggregator. I do think they would access EIA or AGA types of information, but I also think they're using an aggregator to collect price indices, storage information, and everything into one desktop.

[Nick Whalen, Liberal, St. John's East, NL]

How has the EIA affected the U.S. market for aggregators? Has it reduced it, helped it? Do they have a more competitive marketplace for that in the U.S. or a less competitive one compared with Canada?

[Steve Lappin]

That's probably tough for us to comment on.

[Nick Whalen, Liberal, St. John's East, NL]

That's fine. I have another question to move to, because I have very little time.

End-user data is something people have spoken to us about as being something desirable by their particular group, but the manner in which end-user data is collected has really varied. The U.S. does a survey, and I'm not sure if any of your clients on the buy side have expressed any desire for data products that you aren't able to provide because you can't get the data, even though it would be nice if the government provided end-use data in Canada for energy users.

Should that be aggregated across all vendors so that everyone has a clearer sense of the end-use natural gas market, the end-use electrical market, or without a smart grid, the self-generated energy market? From your side, is demand coming up for any of those things?

[Steve Lappin]

We really don't.... We do publish—one of them is the Alberta power market—so we host auctions for that, so that provides the regulated rate option. So if you're in Alberta and you're dealing with that generator provider, that provides the end-user with a variable price that comes with it. We've seen that through markets. Certainly our natural gas pricing represents.... I guess it wouldn't be the retail, more the wholesale—

Are you talking both about the retail and commercial actual end-user? It's a bit of grey area.

[Nick Whalen, Liberal, St. John's East, NL]

No, I think at the commercial level it's known, but it's more how the adoption of clean energy may affect or disrupt other aspects of the energy market. It's unclear when people transition to clean energy what the reduced demand will be on the natural gas market or the electrical grid, and how people can forecast that so they can invest.

[Steve Lappin]

Yes, so again, at the wholesale level, I think that's well covered and quite available, and I think you can get at it, but in fairness, as Greg said, you have to kind of piece it together, in terms of its being centralized and brought together. I think you could go to the various exchanges, clearing houses, other aggregators, brokers. There are just a lot of third party entities that have this information, so you'd have to cobble that together, I think.

In terms of actual end-user information, yes, other than these regulated rates stuff, I'm not sure there's the same level of transparency on that front.

[The Chair, Liberal, James Maloney]

Thanks, Mr. Whalen. That was almost right on time. That was great.

Mr. Schmale, you have as much time as you want.

[Jamie Schmale, Conservative, Haliburton—Kawartha Lakes—Brock, ON]

“Almost right on time”...? I thought Nick was going into Mr. Canning's time.

Thank you, gentlemen, for being here. I appreciate your coming and sharing your opinions. Building on what Nick was talking about, if Stats Canada or an EIA-type agency started to not only publish the data, which they have now, but do analysis on that data, how would that affect your operation or that of businesses like yours?

[Steve Lappin]

I'll take a shot at that, and let Greg finish.

Obviously that's something we monetize and sell, in terms of providing that information. I suppose to the extent it almost became competitive with that, then perhaps that wouldn't be a positive thing. I think, though, there's probably a difference. With exchanges and clearing houses like ours, which use actual transactional data that is electronically captured, and you're talking thousands of transactions per day, rather than surveys and that sort of thing, I think it's just a different type of information.

If that were assimilated by Stats Canada or anybody else, I'm not sure that would be competitive. I think there are other publishers out there who do use surveys and gather data through informal sources, etc., and they may be more impacted by it.

I think the overall transparency is good for all businesses—the more known they are, the more they're out there, the more available they are. It would probably be a fine line at what point it starts to compete with businesses that, as I say, charge folks for that data.

[Greg Abbott]

I think also what's really important is that, while I'm the data guy and it's kind of weird to say this, we do have a core business that really is trading in clearing commodities as well, so I think providing more tools for our clients to be informed and to make more informed decisions around transactions is actually a good thing for our business. From that perspective, it may be competitive with our data business, which obviously we'd want to be aware of, but at the same time, providing that information to clients and allowing them to make better decisions around transactions is important.

NGX is a physical clearing house, a physical marketplace, which makes us sort of unique in the world and unique in North America. Having that kind of depth of information around what's happening in the actual marketplace is important. It's not just other markets where you really...a financial market where it's just buy low and sell high. You really are taking a look at the actual physical underlying fundamentals of a product before you trade it, and I think having that in one spot could be beneficial to our clients.

[Jamie Schmale, Conservative, Haliburton—Kawartha Lakes—Brock, ON]

We heard from that last group of witnesses that the data is there but it's kind of all over the place. How do we put it in a place where it can be accessible? Whether or not the analysis done on that information is done by Statistics Canada or a different organization, whether it be individual businesses or...? I threw out the idea of a user-pay system for that information, where it's separate from government.

Judging by what Statistics Canada is doing now, where do you see that they can improve, going forward, that would help you and others?

[Steve Lappin]

I don't know, certainly for our purposes—and it's tough to say on behalf of our trading counterparties. You just don't hear much, in terms of use of Statistics Canada information. There may be some. I'm not sure. In fairness, on the other side, EIA is more talked about and referenced in terms of being used in modelling and decision-making in all of those sorts of things.

I think it's about credibility. It's about the amount of resources that are going into ensuring that what's there is good, credible data, and that it's well-sourced, it's fulsome, and it's complete. I do know that in the past, for example, we have tried to use some Statistics Canada stuff, and you weren't sure of the sourcing and how accurate it was. It was a bit spotty, in terms of information. Whereas, EIA in the U.S. is much more comprehensive.

I think that's really it. If there's a more concerted effort to get it right and ensure that it actually does put out very comprehensive information, it would be much more useful to the market.

[Jamie Schmale, Conservative, Haliburton—Kawartha Lakes—Brock, ON]

Looking ahead at what Statistics Canada or EIA is doing, obviously protecting data is essential to all facets of information. As you see government getting more into the collection of various pieces of data.... Previously in committee, we heard that companies and government organizations are able to track data in real time. You can talk about the Nest system, where they can get information on the temperature of your house, etc., or about the smart meters for hydro where they can get real-time energy consumption.

Do either of you have any thoughts or concerns about the security of that data?

[Steve Lappin]

Particularly with our organization and having that data, you can just imagine the provisions that are made for cybersecurity and that sort of thing. It's an incredible amount of resources and efforts to ensure that the data, even at a wholesale level—we're not talking individual retail, that sort of thing—is maintained and put out there.

To the extent that it's handled and treated in the same manner and with that same level of care by the government at a federal level, I don't see any issues with it.

You get into that overall invasion. I guess it depends how far you're talking about going with it. If it's general usage—volume, pricing, that sort of thing—then I think that stuff is fairly innocuous to most and I'm not sure I would view that as highly sensitive. If you get into particular individual habits and that sort of thing, then maybe it becomes a bit more touchy.

The short answer is that I don't see any major concerns with it. It just has to be managed well. It's done around the globe on a very careful basis.

[Jamie Schmale, Conservative, Haliburton—Kawartha Lakes—Brock, ON]

We saw data hacks with the Pentagon and with the FBI, etc., so we know it can happen. If a single government agency continues to collect more and more information on every single citizen, on a wide range of things, that's where, potentially, I would like to dig deeper, if we do go down that path.

I will wrap up. I'm done.

[The Chair, Liberal, James Maloney]

Thank you.

Mr. Cannings.

[Richard Cannings, NDP, South Okanagan—West Kootenay, BC]

Thank you for getting up early and appearing before us.

This is a mysterious world for me, the world of energy exchange and all that. I just want to get into these price settlement curves. You mentioned that you do projections over the coming days, and I think you said five years. I assume that this is some type of modelling or some types of projections. I wonder where the data for that comes from, whether you get it externally. I assume that if you're talking five years, it's more than just what was sold yesterday. Through your exchange, it's looking at the world.

Could you maybe expand on that? I'm assuming that a better data centre in Canada might be very informative for those models. Could you comment on that and what you would like to see in a new system?

[Greg Abbott]

Sure.

With the price settlement process, effectively a couple of different pieces happen. NGX has built our own relationship engine or settlement engine. Effectively it's really just the way to price gas, or any commodity, across multiple different locations.

First, we have relationships between the different locations. For example, we'd look at gas trading in Alberta and figure out what the difference would be for gas to get shipped out to Ontario, or Dawn for delivery. We may use that as a base market, and then for a lag market we'd say, “Okay, we know the differential is $1, so if gas is $2 in Alberta, it's $3 at Dawn.” You build all these relationships across the market.

The liquidity on NGX, though, on our exchange, is quite high. As you mentioned, a lot of our forward curves do have actual transactions or bid offer spreads that go fairly far out the curve as well. We can use that to extrapolate our settlements. For example, you may not see a transaction on a day, but because our market is live, you may see a bid and an offer spread that's a few years out the curve.

If in Alberta somebody said, for 2022, they were willing to buy for $3 and somebody was willing to sell for $3.25, we're able to pick the midpoint there and help settle our curve and smooth that curve out.

First, we'll use all the transactions and all the postings on the NGX screen, and then, as you said, we go external. We will talk to market participants who were doing transactions off screen if there were bilateral transactions happening that day. We have a group of staff members each afternoon who are talking to everyone in the market, traders in the market, and figuring out what the pricing was for that day, really just modelling out that forward curve and then using that relationship table that I referenced to settle products that were less posted.

Something might be a little less liquid, so we'll take all the liquidity that was in the Alberta market, for example, which is a very liquid market, figure out what the differential is between Alberta and another market nearby, or close by, and apply that differential to it to settle the other one.

We do look at other different data sources. To your original question as to whether it could help, it could. Having a new centralized location for data could help, but again, there's some real-time functionality to this as well. It's being done every single day. Those numbers are really important. Not only do they mark the end of the market and the start of the next day's market, but NGX also runs a clearing house so that everybody who has positions gets marked against those settlement price curves at the end of the day, too. If you have a position, we'll settle that, and then that number gets actually applied to the margin we hold for those clients.

It all ties together. It's a bit of a science and a bit of an art to do the settlement curves at the end of each day, but we are definitely accessing more sources than just what's on our screen. To do our settlement curves each day, there are a lot of third party data sources, other clients, traders, and brokers in the market who we talk to.

[Richard Cannings, NDP, South Okanagan—West Kootenay, BC]

Just to be clear, you may be using these other settlements, bids, and offers that go into the future, but obviously there are people outside your company who are using some data to make those future bids and they might be interested in a new energy information agency, or whatever, that could give them more data.

Is there data, then, that is not available or not readily available in Canada? I think there was a comment about Statistics Canada data being spotty and not as useful as the EIA's. Are there any specific recommendations you might have for what would make those future projections more useful and more accurate?

[Greg Abbott]

Not really. I guess I would come back to.... For us—and again, we're a little bit on a different footing than some of our clients—it comes down to the aggregation of that data. Most of the datasets that clients would need to develop the forward curve are out there somewhere, but it's multiple different sources you're going to have to go to to find that stuff. I do think that happens, and I think different participants in the market who are calculating forward curves will have different datasets they use. I think it's almost more of an aggregation than an accessibility problem, as opposed to the actual product being there. There's a lot of data out there right now. You really can find transactional data. You can find receipt data. You can find storage information. You can find that stuff, but it's not usually collected in one location.

[Richard Cannings, NDP, South Okanagan—West Kootenay, BC]

We'll just finish by going back to Mr. Whalen's question that he abandoned because he wanted to ask something else. That was around that potential competition for aggregators if we had a new energy information agency that was aggregating data, perhaps in real time, or aggregating recent data.

Could you just comment on the possibility of competition with those aggregators and whether that would be a good thing or a bad thing?