The Washington Post editorial board recently scolded President Obama for putting off a decision on building the Keystone XL pipeline, which is a striking position given the paper’s usual support for actions that address climate change.

If you’re not familiar with the pipeline debate, here’s some quick background: the Keystone XL pipeline is designed to bring oil extracted from Canadian oil sands — a process that emits a larger amount of greenhouse gases than other sources of oil — to refineries and markets in the United States. Apart from the concern about an oil spill along the pipeline’s proposed route, which passes through the Nebraska Sandhills, environmentalists are concerned that the extra demand for oil from Canadian oil sands will quicken that resource’s exploitation and increase the U.S. greenhouse gas footprint at a time when we should be trying to reduce it.

The central argument of the Post was that if the United States doesn’t build the Keystone XL pipeline and burn this oil from the oil sands of Canada, then China or someone else will — therefore, the United States should. It’s a fascinating argument, a line of reasoning often repeated in political punditry in Washington.

Without getting into the merits of the Keystone XL pipeline, I want to focus in on the befuddling logic of the Post’s editorial, which seems to go something like this:

  1. Action X (in this case, not building the pipeline) will contribute to solving problem Y (in this case, our country’s large greenhouse gas footprint).
  2. However, action X by itself will not solve Y problem.
  3. Therefore, don’t do action X.

Stated this way, it’s pretty clear that the leap from 2 to 3 is questionable. Otherwise an overweight person should never go on a diet, since any particular healthy meal is unlikely to solve the obesity problem.

Here’s a personal example of why the logic of what I’ll call “the Keystone argument” is questionable: I often have to go from Washington, D.C., to New York for work and I will usually take the train, in part because it is just about as quick as flying once you factor in the time getting through security, in part because I find it a nice stretch of time to do some work on my laptop, and in part because I like that the train emits less greenhouse gas emissions per capita compared with flying (how much less is a matter of debate). It is true that, on any particular day, my decision to take a train won’t affect the number of flights flown between Washington and New York. But the free market works — if enough people choose not to fly consistently, eventually fewer flights get scheduled in response to that reduced demand.

Why, then, do pundits so often use the Keystone argument, and why do we sometimes find it convincing? In some cases, it’s because there is no articulated theory of how action X (along with other actions) can add up to a solution of problem Y. The problem of global climate change is indeed daunting, and everyone agrees that an ultimate decision not to build the Keystone XL pipeline will not come close to fully solving that problem. Maybe the Post feels that absent a full-fledged theory of how the Keystone XL pipeline’s denial will help solve the entire climate change problem, it’s not worth doing.

So…When Can Small Actions Matter?

Nevertheless, we should recognize that there are cases where lots of voluntary actions add up to a significant total change. My hometown used to have bad droughts almost every summer, and the first action the town council announced was voluntary water-use restrictions. And enough of the town’s residents (although certainly not everybody) would follow those voluntary restriction, often reducing the town’s overall water demand enough to make it through the summer.

So, the only appropriate reason to believe the Keystone logic would be if action X actively got in the way of solving problem Y. Perhaps spending energy on action X distracts from other, more important potential solutions, or action X will make it harder to negotiate a bigger solution to problem Y. But barring that, there seems no reason not to do action X just because it will not, by itself, solve problem Y.

It’s likely that the Post editorial board feels that denying the Keystone XL Pipeline will cause some economic harm to Americans, and that we have no guarantee that other nations will take similar actions to reduce their greenhouse gas footprints. This is a version of the classic collective action problem, with China standing accused of being a potential future free-rider.

These kinds of concerns are why my hometown often had to turn to mandatory water restrictions on water use when the drought got severe enough — it was hard to get enough people to actually reduce their water use voluntarily, because lots of citizens would choose to ignore the voluntary restrictions. But in many contexts, notably international politics, there is no organization with the power to enforce a mandate. In situations without an overarching organization, trust is needed among the different actors to get collaboration. If all parties are always skeptical of other parties’ motives, no one will ever take any action, and the problem will never be solved!

Enter the Prisoner’s Dilemma

The good news is that there’s a body of research that initial small positive acts by one actor can spur collaboration by others. Maybe the most famous example is the Prisoner’s Dilemma game, where two prisoners who both committed a crime together and were caught by police, each face a choice of defecting (confessing to the police what they did) or cooperating (remaining silent during interrogation). If they both cooperate, the police will have no case against them and they will both receive a light sentence (3 months). If one of them defects but the other cooperates, the one who ratted on his friend is released by police but the other one gets a long prison sentence (1 year). However, if they both defect they both will get a moderate prison sentence (6 months).

If this game is only played once, and there is no mutual trust between the two players, it is rational for each player to defect. To see why, picture yourself playing this game: if the other player cooperates, you will get off free if you defect, and if the other player defects, you will get a shorter sentence if you defect. Either way, you receive the greatest benefit by defecting.

However, there’s an enormous body of research into iterated Prisoner’s Dilemma games, where the same players play against each other over and over again. Here conditional cooperation will yield more benefits than always defecting. The best strategy known is tit-for-tat, where in the first turn you cooperate, and then after that you do whatever the other player did in the last round.

This all sounds abstract, but there are applications to real life. One of the biggest concerns when managing natural resource commons is to avoid what is called “the tragedy of the commons”: each person should rationally use more of a resource because he reaps all the gains from this action, but society at large shares the harm from the action. However, if everyone uses more of the resource it will be degraded or destroyed. Eleanor Ostrom’s research, which won her the Nobel Prize in 2009, shows that this tragedy needn’t happen, but that there are many human societies in which conditional cooperation, in the context of a stable management structure for the resource, can lead to long-term persistence of natural resource commons.

Cooperation between nations is arguably harder than cooperation between people in the same community who know and trust each other. But conditional cooperation still ends up being very important. It was unilateral action by the United States to essentially ban ozone-depleting CFCs that helped pass the Montreal Protocol, the global agreement of many nations to reduce CFC use. Unilateral actions, especially by powerful actors, help the formation of global agreements.

How does all this logic relate to the Washington Post’s commentary on the Keystone XL pipeline? An ultimate decision to not build the pipeline could, however slightly, increase the odds of reaching a broader climate agreement because it is a step toward conditional cooperation between nations. It is an action that, by itself won’t solve the climate change problem, but could help spur larger change.

So think about this the next time someone tells you that you shouldn’t bother to recycle, or bike to work, or round up to donate to a good cause, or…build a pipeline.

(Image: Protesting the Keystone XL pipeline at an October rally in Washington DC. Source: Flickr user ElvertBarnes via a Creative Commons license.)

If you believe in the work we’re doing, please lend a hand.


  1. The problem with the prisoner’s dilemma is that it ignores the ethical dimension and focuses on only self interest. While many only care about the latter, it is simply dissonant to make affirmative acts contrary to one’s values. See

    If X contributes to solving problem Y without side effects that outweigh problem Y, then we should do X. You lead by example, not by lying in wait.

  2. The assumption is, of course, that the pipeline’s energy isn’t needed or would otherwise be available on less onerous terms. Neither of these are true.

    I support the Nature Conservancy because, until now, I thought it shared my sense of balance between conservation and economic good sense. I also support the Keystone XL pipeline.

    There is no shortage of extremist environmental organizations who insist on the perfect instead of accepting the good. Does the Conservancy consider itself different from the WWF or Greenpeace?

Add a Comment