Long Lag Problems  

Jamais at Open The Future has a post on the difference between "long run" problems and "long lag" problems - Long-Run vs. Long-Lag.

All distant problems are not created equally.

By definition, distant (long-term) problems are those that show their real impact at some point in the not-near future; arbitrarily, we can say five or more years, but many of them won't have significant effects for decades. Our habit, and the institutions we've built, tend to look at long-term problems as more-or-less identical: Something big will happen later. For the most part, we simply wait until the long-term becomes the near-term before we act.

This practice can be effective for some distant problems: Let's call them "long-run problems." With a long-run problem, a solution can be applied any time between now and when the problem manifests; the "solution window," if you will, is open up to the moment of the problem. While the costs will vary, it's possible for a solution applied at any time to work. It doesn't hurt to plan ahead, but taking action now instead of waiting until the problem looms closer isn't necessarily the best strategy. Sometimes, the environment changes enough that the problem is moot; sometimes, a new solution (costing much less) becomes available. By and large, long-run problems can be addressed with common-sense solutions.

Here's a simple example of a long-run problem: You're driving a car in a straight line, and the map indicates a cliff in the distance. You can change direction now, or you can change direction as the cliff looms, and either way you avoid the cliff. If you know that there's a turn-off ahead, you may keep driving towards the cliff until you get to your preferred exit.

The practice of waiting until the long-term becomes the near-term is less effective, however, for the other kind of distant problem: Let's call them "long-lag problems." With long-lag problems, there's a significant distance between cause and effect, for both the problem and any attempted solution. The available time to head-off the problem doesn't stretch from now until when the problem manifests; the "solution window" may be considerably briefer. Such problems can be harder to comprehend, since the connection between cause and effect may be subtle, or the lag time simply too enormous. Common-sense answers won't likely work.

A simple, generic example of a long-lag problem is difficult to construct, since we don't tend to recognize them in our day-to-day lives. Events that may have been set in motion years ago can simply seem like accidents or coincidences, or even assigned a false proximate trigger in order for them to "make sense."

But a real-world example of a long-lag problem should make the concept clear.

Global warming is, for me, the canonical example of a long-lag problem, as geophysical systems don't operate on human cause-and-effect time frames. Because of atmospheric and ocean heat cycles (the "thermal inertia" I keep going on about), we're now facing the climate impacts of carbon pumped into the atmosphere decades ago. Similarly, if we were to stop emitting any greenhouse gases right this very second, we'd still see another two to three decades of warming, with all of the corresponding problems. If we're still three degrees below a climate disaster point, but have another two degrees of warming left because of thermal inertia regardless of what we do, we can't wait until we've increased to just below three degrees to act. If we do, we're hosed.

With long-lag problems, you simply can't wait until the problem is imminent before you act. You have to act long in advance in order to solve the problem. In other words, the solution window closes long before the problem hits.

We have a number of institutions, from government to religions to community organizations, with the potential to deal with long-run problems. We may not do well with them individually, but as a civilization, we've developed decent tools. However, we don't have many -- perhaps any -- institutions with the inherent potential to deal with long-lag problems. Moreover, too many people think all long-term problems are long-run problems.