Forward vs. reverse causal questions

Andrew Gelman has a thought-provoking post on asking "Why?" in statistics:
Consider two broad classes of inferential questions: 
1. Forward causal inference. What might happen if we do X? What are the effects of smoking on health, the effects of schooling on knowledge, the effect of campaigns on election outcomes, and so forth? 
2. Reverse causal inference. What causes Y? Why do more attractive people earn more money? Why do many poor people vote for Republicans and rich people vote for Democrats? Why did the economy collapse? [...] 
My question here is: How can we incorporate reverse causal questions into a statistical framework that is centered around forward causal inference. (Even methods such as path analysis or structural modeling, which some feel can be used to determine the direction of causality from data, are still ultimately answering forward casual questions of the sort, What happens to y when we change x?) 
My resolution is as follows: Forward causal inference is about estimation; reverse causal inference is about model checking and hypothesis generation.
Among many gems is this:
A key theme in this discussion is the distinction between causal statements and causal questions. When Rubin dismissed reverse causal reasoning as “cocktail party chatter,” I think it was because you can’t clearly formulate a reverse causal statement. That is, a reverse causal question does not in general have a well-defined answer, even in a setting where all possible data are made available. But I think Rubin made a mistake in his dismissal. The key is that reverse questions are valuable in that they focus on an anomaly—an aspect of the data unlikely to be reproducible by the current (possibly implicit) model—and point toward possible directions of model improvement.
 You can read the rest here.


Pricing the clathrate gun hypothesis

In this week's Nature:
We calculate that the costs of a melting Arctic will be huge, because the region is pivotal to the functioning of Earth systems such as oceans and the climate. The release of methane from thawing permafrost beneath the East Siberian Sea, off northern Russia, alone comes with an average global price tag of $60 trillion in the absence of mitigating action — a figure comparable to the size of the world economy in 2012 (about $70 trillion). The total cost of Arctic change will be much higher. Much of the cost will be borne by developing countries, which will face extreme weather, poorer health and lower agricultural production as Arctic warming affects climate. All nations will be affected, not just those in the far north, and all should be concerned about changes occurring in this region. More modelling is needed to understand which regions and parts of the world economy will be most vulnerable.
Wikipedia on the clathrate gun hypothesis here. For scale, Costanza et al. calculated the annual value of the world's ecosystem services in 1997 at $16-54 trillion, or $23-79 trillion in today's dollars.


Seismic externalities

Injection-Induced Earthquakes
William L. Ellsworth
Abstract: Earthquakes in unusual locations have become an important topic of discussion in both North America and Europe, owing to the concern that industrial activity could cause damaging earthquakes. It has long been understood that earthquakes can be induced by impoundment of reservoirs, surface and underground mining, withdrawal of fluids and gas from the subsurface, and injection of fluids into underground formations. Injection-induced earthquakes have, in particular, become a focus of discussion as the application of hydraulic fracturing to tight shale formations is enabling the production of oil and gas from previously unproductive formations. Earthquakes can be induced as part of the process to stimulate the production from tight shale formations, or by disposal of wastewater associated with stimulation and production. Here, I review recent seismic activity that may be associated with industrial activity, with a focus on the disposal of wastewater by injection in deep wells; assess the scientific understanding of induced earthquakes; and discuss the key scientific challenges to be met for assessing this hazard.
Perhaps an enterprising graduate student can figure out an optimal management strategy for this risk.


Using Weather Data and Climate Model Output in Economic Analyses of Climate Change

After 5 (or 6?) rounds of revisions (a lesson to anyone thinking of writing an interdisciplinary review article...), this is finally published:
Using Weather Data and Climate Model Output in Economic Analyses of Climate ChangeReview of Environmental Economics and PolicyMaximilian Auffhammer, Solomon M. Hsiang, Wolfram Schlenker and Adam Sobel
We tried to write this as a practical and gentle introduction and how-to manual for econometricians and other applied social scientists. I hope it's helpful.