Imagine you get up extra-early to make a nice homemade lunch for yourself, including a extra-tasty BLT with fresh-cooked bacon. However, while on the bus, you experience an existential crisis for which, you conclude, the only plausible resolution was to become a vegetarian. Come lunchtime, you open your brown bag, only to discover a sandwich you now can't eat. You might be annoyed that you'd wasted time making the sandwich instead of sleeping in, but you shouldn't really care since there's no way to undo your decision and get the time back. As economists would put it, your effort to make the sandwich is a sunk cost. So what to do? You're extremely unlikely to go back on your new philosophy just because of a BLT (no matter how tasty), and you have to eat, so you'll probably buy a new sandwich. Or, if you're clever, you'll take the bacon off the sandwich and put it in the bag - a crucial breakthrough in BCS (bacon capture and storage) technology. In any case, you'll adapt. Having a BLT doesn't lock you in to eating it, no matter how much effort you put in to making it.
If this is true for BLTs, why isn't it true for power plants? Many people seem to believe that it's very important to stop building new fossil fuel plants (especially coal) not just because of their GHG emissions and other associated pollution, but because once these plants are built, they'll be "locked in". As Michael Levi puts it in a recent Foreign Policy piece, "new coal-fired power plants are expensive and thus tough to displace later." (note - I'm not picking on Levi here. The comment is a tiny part of an excellent piece, and many others have made the same argument).
I don't understand this argument. It shouldn't matter how expensive it was to build something. If you later decide it has to go, the costs of building it are sunk. It doesn't matter how hard you worked on that sandwich if you're not going to eat it, or how much you paid for that ticket if the movie is terrible. The past shouldn't be an anchor - and even if it is human to sometimes value sunk costs, why would rational, motivated companies do so? When and if we do put a price on carbon, some plants will shut down - and it won't matter how expensive they were when they were built. This has been happening for decades in reaction to other environmental regulations.
So why does this argument persist? (or, alternatively, what am I missing?)
One explanation is that it's just a descriptive statement. It looks like power plant technology gets locked in because power plants tend to stick around for a long time. Therefore, one assumes, anything built now will similarly be there for decades into the future. But this is a circular argument. A big reason coal plants have stuck around so long is that we haven't been very good at making them pay for the damage they cause. Put good policies to do that in place, and they will close - and it will be the worst offenders that close first, not the most expensive plants.
Another explanation is that "lock-in" isn't really what is meant - instead, the argument is really that its just expensive to build polluting plants now, then shut them down in the near future. We'd be better off building cleaner plants now, so any policy that fails to promote that is more expensive in the long run. This is correct, of course, but it's not lock-in. And it's not all that compelling, either. If policymakers think future energy policies will make coal uneconomic, they should say so. Even if they say nothing, power companies are smart enough to see the writing on the wall. Most already include estimates of future regulatory costs in their models, and are less likely to build coal plants as a result. It's undoubtedly true that we shouldn't use policies that push investments we think will be wasted in the long term. But this isn't lock-in, it's just avoiding waste.
Finally, lock-in might be about politics. Even if firms react rationally to policy change, they probably won't like it. They'll therefore complain to politicians, with their lobbyists using the superficially compelling (but incorrect) argument that carbon policy shouldn't be enacted because it would require the shutdown of "expensive" facilities. If you're an advocate of stronger carbon policy, therefore, you worry that more coal generation just means more rhetorical ammunition for opponents. This metagame lock-in argument is very clever, but I'm not sure if it's true, and it needs to be spelled out.
A final possibility is that the way the power sector is organized makes lock-in possible, maybe due to the complexity of regulated electricity markets. If this is true, I'm happy to be corrected.
Until and unless that happens, I'm deeply skeptical of whether lock-in is a problem. That doesn't mean I think building new coal is a great idea - it's bad enough on its own merits, and likely to be wasted investment under almost any future energy policy. But that's less true for gas. Lock-in gets used as a counterargument to the idea that gas can be a bridge fuel to a zero-carbon future. There are good reasons to be critical of that scenario, but I don't think lock-in is one of them. When the economics of energy change, whether because of policy or just underlying conditions, the mix of generating technologies will change with it. How much it cost past generations to build those plants won't - or at least shouldn't - matter.
