Hey, as someone who used to work for an energy utility looking at setting up a solar market place, there's some info you're missing:

- The wholesale price is the cost of generating electricity. The retail price includes the cost of serving the customer, i.e. billing. Sure, there are less transmission costs, but in New Zealand they only play about 20% of the retail cost.

- Also, just because midday rates of electricity are high, those high rates tend to subsidise the cost of running the network when rates are low (i.e. other times). Power Stations don't come for free. People want security of supply, but then to be able to poach from the most profitable parts of the day?

I don't know why your first point is relevant, the marginal cost of billing is incredibly minute. The second point is interesting but calling it poaching is tendentious. No one is saying power stations, transmission lines, distribution networks, voltage support, frequency regulation, or any of the other myriad features of modern electricity grids are free. The point is how do you distribute the costs of those goods fairly (i.e. in proportion to use). Utilities are saying netmetering is unfair because it results in stranded costs that must be born by non-solar customers and I am saying that they are overestimating their stranded costs to the extent that they don't count the increase in the efficiency of their ability to serve load (because solar gen is basically already made and delivered electricity they bill for but don't have to buy). It is a really really hard question to know by how much they are overestimating stranded costs.

IMO this is about aggregate demand and the stranded cost thing is a red-herring.

Sources please! There's more to retail than just billing. Off the top of my head there's advertising, acquisition of customers, creation of plans/pricing/campaigns, getting consumption information (which in New Zealand is from 50+ local distribution companies, each with their own formats), there's purchasing from the wholesale market in such a way that you're not losing money all the time, there's setting up a call centre to service those customers, there's meeting regulatory requirements such as a government disputes process, those are just the ones I can think of sitting here.

And I'm not saying that those costs should be massive, I'm just saying those costs exist. Someone must pay for them. Considering I've observed that the margin on retail electricity customers is not much (like $2K of customer lifetime value over 10 years), it's not like in New Zealand electricity companies are rolling around in cash from profit per customer, it's really just a volume game (the company I worked for had 430,000 retail customers).

BTW, the value chain in energy from generation to transmission to meter to cash is full of lots of participants. In New Zealand, the generator, wholesaler, retailer, national distributor, and local distributor are all separate entities. To say "Utilities ... are overestimating their stranded costs to the extent that they don't count the increase in the efficiency of their ability to serve load" ignores the reality of this value chain. At best, this really only benefits the local distributor.

I was coming at it from a US perspective. In the US retail is still dominated by natural monopolies with service areas and captured customers so those costs are low. There is a huge push toward the kind of retail competition you are talking about of course, however, while the growth in the sector is huge the vast majority of customers still behave as captives. https://ei.haas.berkeley.edu/research/papers/WP252.pdf

As for the value chain you are right that I am simplifying but we are on an internet comment board not writing a white paper. However, the simplification doesn't matter. Whether it is vertically integrated utilities or deregulated and unbundled electricity service, the point still stands that stranded costs are overestimated to the extent that they don't account for the fact the netmetered kWhs tend to be high value, are resold at retail, and do not have to be purchased or delivered by the seller. Adding more middleman or more costs just adds accounting wrinkles (and which I think are inappropriate for the forum).

If your point is that the harm is caused by a reduction in aggregate demand then that is a completely different argument.

>Sources please!

Asking for sources for a rebuttal to your opinion is a bit underhanded given that you provided none to backup your claim.

And no, your comment isn't more than an opinion because you've worked in the industry. An informed opinion sure but asking for someone to provide sources to their argument would be a bit hypocritical unless you are going to step up as well.

I enjoy a good internet discussion!

So the topic of debate is whether its fair to charge a retail price for solar buybacks, or a wholesale price. I argued that a retail price is unfair, because energy utilities have a bunch of retail costs, of which billing is one of. The counter argument is that billing is a minute component of the retail cost. You've suggested that I have an informed opinion, but need sources to back up my argument.

Contact Energy have a management paper that discusses energy costs. A definition of Netback from Contact Energy (2015) is:

Netback is calculated by deducting the network, meter, levy and cost to serve costs from customer tariffs. This enables the performance of the retail channels to be measured without using an energy cost. The netback is meant to cover, inter alia, the cost of energy, capital return, risk margin and a retail margin.

Cost of Energy was $994M, while Cost of Transmission and Metering was $621M, and Cost to Serve (such as billing) was $118M for Contact Energy for the year ending 30 June 2015 (Contact Energy, 2015). The remainder, as EBITDAF was $484M.

As you can see, $118M for Cost to Serve is 11.8% of the Cost of Energy. This doesn't change with solar buybacks. The cost of energy is from a different source (your personal solar panels), but there's still costs of transmission and metering, cost to serve, and depending on the utility you connect to, some sort of EBITDAF (unless they're a not for profit). So that's $0 (your solar panels) + $621M (transmission/metering) + $118M (retail/billing) + $484M (earnings before income, tax, depreciation, amortisation, and changes in fair value of financial instruments).

Do check out McNicol (2013) for an interesting article on this.

tl;dr - Electricity networks are expensive, and the cost of energy is about half the total costs of running the system. Who pays for that, if the utility has to pay the retail rate to you because you're using solar panels?

References: Contact Energy. (2015). Management discussion of financial results for the year ended 30 June 2015. Retrieved 9 February 2016 from https://www.contact.co.nz/cenergymedia/contactenergy/files/p...

McNicol, H. (2013). Are solar energy systems worth the cost? Retrieved 9 February 2016 from http://www.stuff.co.nz/the-press/business/the-rebuild/932970...

Here is my point, suppose a kWh costs 10 cents retail. This represents an average cost of maybe 1 cent of profit, 1 cent of admin costs, 3 might be transmission costs, and 5 might be generation costs.

In a 1-1 net metering scheme, I generate 1 kWh in the distribution grid, it goes to my neighbor, the utility bills my neighbor 10, and credits me 1 kWh. When I draw that 1 kWh, the utility loses 2 cents on average. But they also saved 8 cents by selling my kWh to the neighbor. The summing works like this:

10 from my neighbor to the utility 0 from me to the utility -10 for the utility to provide me that kWh 8 to the utility for avoiding the cost of generating and distributing the 1 kWh that went from my system to the neighbor.

The net to the utility is 8 cents. Ouch. But we are talking about the average kWh. It might cost the utility 3 cents to buy an off-peak kWh and 8 cents to buy an on-peak. If that is the case, they are already losing 2 cents for every on-peak kWh they are selling me but making up for it with off-peak (I am likely to use more off-peak anyhow).

In this case, netting might look like the following:

10 from my neighbor to the utility 0 from me to the utility -8 for the utility to provide me an off-peak kWh 11 to the utility for avoiding the cost of generating and distributing the 1 kWh that went from my system to the neighbor.

The net to the utility in this case is 13 cents.

13 cents is better than 8 cents but it is worse than the 20 cents they would have made if I just consumed without generating.

Whether you credit at wholesale or do a .75-1 netmetering or whatever depends on evaluating the stranded costs and whether they are reasonable. Insofar as you look at average kWh prices and not the value of actual production from solar, you inevitably overstate the stranded costs.

Sounds like a case for splitting the delivery cost from the generation cost. Here in Norway I can buy my electricity from an generating company but I also pay the local company for the cost of transmission plus a fixed fee for the connection. This means that the basic infrastructure gets paid for even if I use no electricity at all.

This is true in many states actually. (This model is definitely true in the state I live in)

Nevada specifically seems a bit messed up right now. Politics will have to duke it out... I am not actually on the Solar side of things.

You seem to be missing the fact that utilities charge a "grid connection fee" to cover the cost of overhead like billing. In my case, that fee accounts for half of my bill. You're also failing to consider that solar installations remove load from the grid, allowing utilities to put off upgrades that would have otherwise been inevitable.

> just because midday rates of electricity are high, those high rates tend to subsidise the cost of running the network when rates are low

Prices should be quoted appropriately. Frankly, grid operators should be market makers, continuously quoting bids and asks they'd be happy to buy and sell power at. If the mid-day rate subsidizes the evening, maybe the latter needs to be higher.

They kind of are like this: there are real-time electricity markets in the US. One issue is that a lot of demand is inelastic because price signal transmission is very hard. Grandmas don't want to be adjusting their Air Conditioner on a second-by-second basis due to price fluctuations. Utilities effectively buffer this movement by aggregating demand and purchasing enough to meet that demand.

> Grandmas don't want to be adjusting their Air Conditioner on a second-by-second basis due to price fluctuations. Utilities effectively buffer this movement by aggregating demand and purchasing enough to meet that demand.

This is how Nest makes money. They serve as demand response. You sign up to get a credit on your electric bill, and when demand is extremely high, a utility can signal to Nest to reduce demand for 15 minute increments; its cheaper than firing up a peaking natural gas plant.

https://nest.com/energy-solutions/#demand-response

The high-value kWH logic only holds true with a negligible amount of running solar capacity, however. If we realize the vision of a massive restructuring of energy provision towards photovoltaics, we also have to remodel how to determine the fair price for solar production. In Germany, for example, wholesale rates on sunny days are severely depressed since a few years (which doesn't hurt producers, however, since they were guaranteed fixed feed-in rates, but electricity consumers in general who have to pay for the support programs.)

This is a good point though one thing is that we can pretty much always use more electricity (Alcoa can start smelting more or something) and since we don't have feed-in tariffs I am not sure we will see those same market signals.

Right. In addition, flattening intermittent feed-ins as a business model hasn't really taken up (via large scale storage solutions, for example). Of course FITs strengthen the drive towards market-ignorant feed-ins. Still, as most renewables have very low variable costs, one might expect similar problems in more market-like regimes too.

<Every unit I send back to the grid is a unit that doesn't need to be delivered by the power company but the power company can sell it to my neighbor for retail price.>

Your entire premise assumes that there is an immediate, local, retail consumer who will buy all of your production the instant you produce it. Under the net-metering model, the utility has to pay you retail for the power even if it is surplus and unsellable.

I don't understand your logic for "more valuable" - there's more electricity at midday (because sun) and less usage (because sun and people are at work), so more supply & less demand => price drops, not rises!

Utilities who bill for "time of use" generally have more punitive rates during the day time because of higher demand (and also send these flyers suggesting you use electric washers and dryers closer to midnight).

The fact that people are at work doesn't imply a drop in usage - they just start consuming electricity at the office, retail location, etc.

I suspect that the increased demand such as air conditioning or business-related electrical usage peaks during the daytime (business hours). In hotter climates, air conditioning is a substantial user of electricity, even to the point where some places have to ration their usage of it. Solar puts electricity back on the grid at the time when it's most in demand.

Well in general demand peaks midday while household usage is u-shaped (morning and evening peaks). This makes sense, since when people go to work they are not at home and their work energy use tends to be higher than home energy use. But if you don't like that explanation, simply look at any price curve for any of the organized electricity markets (they are public) and you will see demand and price peaks mid-day.

Aha... so do you have a typo in your answer?

> Solar production peaks midday

should have been

> Solar consumption peaks midday

?

Electricity consumption peaks mid-day, solar or not. Solar production happens to also peak mid-day, which is great because it coincides with the peak consumption period, which makes it more valuable.

No. Solar consumption does peak midday but that is trivial because electricity is nearly always used when it is produced. Electricity demand, which is basically the same thing as consumption, peaks midday (as do prices). And solar production peaks midday. The fact that solar production matches demand is a nice feature of solar that I think militates in favor of net-metering policies.

Ok, I get it... I misread and misunderstood your comment in a few different ways :)

Industrial machinery for facilities that don't run 24 hrs/day increase the draw energy during business hours.

air conditioners in the desert