Unretiring: Ars Technica's Almost Good Look At Renewable Economics

So, unretiring this blog for one post; don't think it'll be a habit, people. Ars Technica has a nearly good article on the economics of renewables that is worth reading, mostly because it actually does some thinking about the realities on the ground instead of engaging in airy wishful thinking. For all the headline happy talk, at least the article is clear-eyed enough to recognize (but dramatically downplay) the real drawbacks of renewables: the capital and operational costs needed to keep 24/7 power going. They identify two "apocalyptic" situations for utilities involving renewables, the first being the

Consumer grade death spiral: Right now, given the right incentives, consumer photovoltaic installations can pay off in the long run. But there's a big up-front investment and a long payoff time. Though various leasing agreements have been developed that take some of the burden off consumers, the high costs and long payback act as a large barrier to adoption.

But if the costs of electricity rise a bit—something that could be catalyzed by the push for solar at the utility level—then these barriers get lower, potentially driving more consumers to install their own hardware. That would force utilities to try to make the same amount of money from fewer customers, which means raising rates. And so the cycle begins, as the higher rates drive even more consumers to install their own hardware.

This is mostly a problem where net metering is involved. If you expect the retail price to equal the wholesale price of electricity, this is a demand for frictionless markets. This does not now nor will it ever exist; the real costs of operating a grid do not vanish simply because some people use a more virtuous energy source. (Side lesson: if you wish to truly "go off the grid", do so, but be prepared to pay for backup storage when you do.) The greens howling about an extra $5/month as a separate maintenance fee most likely haven't seen nothin' yet.

Part the next:

Grid-scale death spiral: The other potential apocalyptic situation that utilities face is caused by things like California's push for renewables. They take place against the backdrop of a very slow growth in demand in the US; the US Energy Information Agency estimates that demand will grow at a rate of less than one percent for decades. While that still entails a regular construction of new electricity sources as old hardware is retired, those requirements are pretty easily met. "Even with plant turnover," Kannberg told Ars, "you're looking at maybe another two percent a year of additional generation requirements. So you only have three or four percent required to meet load and plant replacement. Part of the challenge with some of the renewables is that we're adding them at a rate that is far in excess of four percent or five percent in some regions. And that's already initiating a reduction in the utilization of conventional generation."

That will only get worse as the push for 50 percent kicks into gear, which means that the electricity generated by new renewable hardware ends up competing for customers with existing fossil fuel based hardware. The renewables have a key advantage in this competition. Fossil fuel plants must have the electricity they sell cover the cost of the fuel burned to create it. Once built, renewables get their electrons nearly for free. As a result, they can always undercut the price of electricity generated by fossil fuel plants.

Here, they fall into a trap that is a frequent pitfall for renewable proponents: the cost of fuel is not the same thing as the cost of operation of the grid. That is, to be honest, either the cost of storage (and its operation) has to be factored in to the cost of renewables, or the capital/operation cost of the standby power needed when the sun is down or the wind is calm.

The net result is that fossil fuel plants are going to be sitting idle more of the time. This isn't a huge problem from an engineering perspective. Natural gas plants can be brought back online as needed (although coal can't), and time spent idle means significantly lower maintenance costs, which largely offsets the lack of business.

But those facilities were built with expectations in mind (and ledger books) about when they'd return a set amount of cash to their owners. To meet those expectations on fewer hours of operation, the only real option is to charge more when power from these plants is needed. Which makes renewable alternatives more appealing, and that will drive further construction.

This is true only if we draw a line around renewables and absolve them of responsibility for operating 24/7. That renewable advocates pull this stunt silently — something we also frequently see when power and energy get conflated — shows they are more animated by considerations other than the engineering aims of the electric grid, and in fact are at times openly hostile to them. There are several points worth raising here:

1. The cost of renewables has to include operational and capital costs of other sources when the sun goes down or the wind is still. Or,
2. Include the cost of storage for surpluses generated during peak availability.
3. Even if you don't build new capacity, existing generating capacity has capital return costs that need to be met, meaning the price those sources charge the grid have to go up to make up for it. (Ars does get this one right.)
4. Anyone telling you renewables harness "free" energy conveniently forget there are real maintenance and operational costs associated with renewables — and green advocates (and even engineers) are terrible at estimating how much these will be. For instance, the Ivanpah solar plant will require five times as much natural gas (yes, it's dependent on fossil fuels!) as was initially designed.

But at least Ars is looking at these issues. I all too frequently read terrible stories on these topics that pull the "enough energy to power X homes" dodge, which confuse power and energy, and are frequently PR efforts from frauds. It is, indeed, the economics, stupid. While I haven't done anything like a systemic study on the matter (and it would be too huge to really attempt, I fear), it seems to me dollar costs can be used as a rough proxy for energy costs. If someone tells you they need to throw those costs elsewhere, run.