Landsat at your fingertips

The USGS has put together a slick GUI that let's you browse (as if it were GoogleEarth) and download Landsat data. The interface is described here.

One of my students found this and showed it to me. (Over the next few weeks, hopefully I'll be able to post much of the material and discoveries from my new course "Spatial Data and Analysis".)

What is identification?

There are relatively few non-academic internet resources on identification and causal inference in the social sciences, especially of the sort that can be consumed by a nonspecialist. To remedy that slightly I decided to tidy up and post some slides I've used to give talks on causal inference a few times in the past year. They're aimed at senior undergrad or graduate students with at least some background in statistics or econometrics, and can be found here:

Causal Inference, Identification, and Identification Strategies

Feel free to drop me a line and give me feedback, especially if somethings seems unclear / incorrect. Thanks!

Year of Reviews in Review: The New Environment and Development Literature

Amir Jina and I were recently discussing the multiple literature reviews that have come out on environment and development topics lately, and realized that there were so many we were starting to lose track. To that end, and as a service to those of you who aren't constantly trawling the working paper and journal lists, here's a quick rundown of the over ten (and counting) recent literature reviews that have come out in the newly emerging environment and development literature:

• Dell, Jones, and Olken's "What Do We Learn from the Weather? The New Climate-Economy Literature". Reviews the growing number of papers focusing on climate and weather impacts, and provides conceptual guidelines for using these results in IAMs and similar models.
• Greenstone and Jack's "Envirodevonomics". Explores low valuation of environmental quality in developing contexts, proposes four basic mechanisms through which this might occur, and reviews the environment and development literature in search of relevant evidence.
• Zivin and Neidell's "Environment, Health, and Human Capital". Highlights economists' main contributions to our understanding of the relationship between the environment and human health, and reviews the applied micro literature on pollution exposure.
• Kousky's "Informing climate adaptation: A review of the economic costs of natural disasters" (ungated RFF working paper here). Reviews the empirical literature on disaster impacts with an eye towards informing the adaptation literature.
• Hsiang, Burke, and Miguel's "Quantifying the Influence of Climate on Human Conflict" (ungated version here). Metanalysis of the climate and conflict literature documenting a general relationship between climate fluctuations and conflict across both spatial and temporal scales. Accompanying review article is Hsiang and Burke's "Climate, Conflict and Social Stability: what does the evidence say?"
• Currie, Zivin, Mullins, and Neidell's "What Do We Know About Short and Long Term Effects of Early Life Exposure to Pollution?". Provides a conceptual model and reviews the empirical literature on the effects of early life pollution exposure.
• Auffhammer and Mansur's "Measuring Climatic Impacts on Energy Expenditures: A Review of the Empirical Literature". Overview of the empirical literature on climate fluctuations as a driver of energy demand.
• Auffhammer and Schlenker's "Empirical Studies on Agricultural Impacts and Adaptation" (forthcoming). Reviews the growing ag-climate literature and the empirical adaptation literature that has come from it.
• Deschênes' "Temperature, Human Health, and Adaptation: A Review of the Empirical Literature". Outlines conceptual issues related to deducing empirical relationships between temperature and health and overviews relevant literature with an eye towards informing IAMs.
• Almond and Currie's "Killing Me Softly: The Fetal Origins Hypothesis". Provides an overview of the explosion of literature on fetal origins / in-utero impacts on latter life outcomes that has emerged in the last decade. 
• Auffhammer, Hsiang, Schlenker and Sobel's "Using Weather Data and Climate Model Output in Economic Analyses of Climate Change"provides a review of methods and issues associated with researching climate and its impact on social systems.
• Pattanyak and Pfaff's "Behavior, Environment, and Health in Developing Countries: Evaluation and  Valuation" surveys the literature on the interaction between household behavior and environmental health problems and comes up with four generalizable policy pathways

Weekend Links

1) A general audience-accessible explanation of why was Typhoon Haiyan so damaging

2) Statistically derived contributions of diverse human influences to twentieth-century temperature changes

3) Dani Rodrik on ideas in political economy models (NBER)

4) Andrew Gelman and Guido Imbens defend the search for "causes of effects" (NBER, and previously on FE)

5) ODI has a new report on "The geography of poverty, disasters and climate extremes in 2030"

6) Guy Grossman and Walker Hanlon's paper on leadership quality and monitoring is out in AJPS

7) "The Climate Corporation sells weather insurance, but it is an insurance company the way Google is an encyclopedia company"

8) "Climate models" (via Colin Schultz)

9) Maternal health externalities of youtube

10) "Learning how to die in the Anthropocene"

Destruction, Disinvestment, and Death: Economic and Human Losses Following Environmental Disaster

Typhoon Haiyan as seen from space, Copyright 2013 JMA/EUMETSAT

Last spring Sol and I finished up the working paper version of our paper "Destruction, Disinvestment, and Death: Economic and Human Losses Following Environmental Disaster." Since the paper is long and fairly technical, we decided it would be worthwhile to do a shorter, more general-audience-appropriate piece for the blog, something that seems especially relevant given Typhoon Haiyan's devastating landfall this past weekend. If you'd like to take a look at the paper itself, you can find a copy of it here on SSRN; a copy of the supplemental appendix can be found here.

The motivation for "Destruction, Disinvestment, and Death" stems from the fact that we actually know surprisingly little about how people fare in the wake of natural disasters.

Climate, conflict, and social stability: what does the evidence say?

The "sister paper" to our recent Science article on climate and conflict has come out in Climatic Change. This new article is a traditional review article that walks readers through individual studies in the literature and discusses some of the debates in less technical terms than the Science article. The sample of studies included is slightly different since the submission timeline for the two studies was different.

Climate, conflict, and social stability: what does the evidence say?
Solomon Hsiang and Marshall Burke

Abstract: Are violent conflict and socio-political stability associated with changes in climatological variables? We examine 50 rigorous quantitative studies on this question and find consistent support for a causal association between climatological changes and various conflict outcomes, at spatial scales ranging from individual buildings to the entire globe and at temporal scales ranging from an anomalous hour to an anomalous millennium. Multiple mechanisms that could explain this association have been proposed and are sometimes supported by findings, but the literature is currently unable to decisively exclude any proposed pathway. Several mechanisms likely contribute to the outcomes that we observe.

Weekend Links

1) Remote sensing evidence of differing institutional history

2) Monasteries and snow leopard conservation

3) RealClimate on the Marcott et al. Holocene temperature reconstruction (previously on FE)

4) Papers from the 1st International Workshop on Econometric Applications in Climatology are up

5) Was Stalin Necessary for Russia's Economic Development? (NBER)

6) Hi-res paleogeography

7) On hours worked vs. productivity (via Yaniz Stopnitzky)

8) Massive groundwater discovery in Kenya (via Emily McPartlon)

9) "[W]e present a new index of the year when the projected mean climate of a given location moves to a state continuously outside the bounds of historical variability" Estimate range: 2047-2069. (Nature, via Sarah Dwyer)

10) Almost half of public school students in the United States are low income (via Dave Pell)

Envirodevonomics

There's a new working paper by Michael Greenstone and Kelsey Jack that's of obvious interest to FE readers:

Envirodevonomics: A Research Agenda for a Young Field
Environmental quality in many developing countries is poor and generates substantial health and productivity costs. However, existing measures of willingness to pay for environmental quality improvements indicate low valuations by affected households. This paper argues that this seeming paradox is the central puzzle at the intersection of environmental and development economics: Given poor environmental quality and high health burdens in developing countries, why is WTP so low? We develop a conceptual framework for understanding this puzzle and propose four potential explanations: (1) due to low income levels, individuals value increases in income more than marginal improvements in environmental quality, (2) the marginal costs of environmental quality improvements are high, (3) political economy factors undermine efficient policy-making, and (4) market failures such as weak property rights and missing capital markets drive a wedge between true and revealed willingness to pay for environmental quality. We review the available literature on each explanation and discuss how the framework also applies to climate change, which is perhaps the most important issue at the intersection of environment and development economics. The paper concludes with a list of promising and unanswered research questions for the emerging sub-field of “envirodevonomics.”

New GIS Data for the Demographic and Health Surveys

USAID's Measure Demographic and Health Surveys (DHS) are an extraordinary (free) data set on maternal and child health from around the world. They've just released their new spatial data repository which, among other things, adds crucial shape files of subnational region borders over time. From the official announcement:

We are pleased to announce the launch of a new open data GIS resource from MEASURE DHS.                                                                          

The Spatial Data Repository provides geographically-linked health and demographic data from the MEASURE Demographic and Health Surveys (DHS) project and the U.S. Census Bureau for mapping in a geographic information system (GIS).

-Boundaries of DHS regions can be explored to visualize change over time.

-Data from DHS indicators and U.S. Census population estimates can be downloaded in GIS format.

Please share with your colleagues and friends.

Go explore!  http://spatialdata.measuredhs.com/

Weekend Links

1) Sol's work on climate and conflict as covered by America's Finest News Source

2) Air pollution elasticity of tourism, China edition (via Sunny Wong)

3) Research idea generation propensity by location (via Alex McQuoid)

4) Weather and consumer behavior (via Kyle Meng)

5) "Twitter mood predicts Hunter College High School start date", if you will (via Luke Stein)

6) "Our undiscounted estimates indicate that the cost of the five main NCDs will total USD 27.8 trillion for China and USD 6.2 trillion for India (in 2010 USD)" (NBER)

The value of forecasting

Mark Rosenzweig and Chris Udry have a pretty nice new working paper on the value of weather forecasts in India:

Forecasting Profitability
We use newly-available Indian panel data to estimate how the returns to planting-stage investments vary by rainfall realizations. We show that the forecasts significantly affect farmer investment decisions and that these responses account for a substantial fraction of the inter-annual variability in planting-stage investments, that the skill of the forecasts varies across areas of India, and that farmers respond more strongly to the forecast where there is more forecast skill and not at all when there is no skill. We show, using an IV strategy in which the Indian government forecast of monsoon rainfall serves as the main instrument, that the return to agricultural investment depends substantially on the conditions under which it is estimated. Using the full rainfall distribution and our profit function estimates, we find that Indian farmers on average under-invest, by a factor of three, when we compare actual levels of investments to the optimal investment level that maximizes expected profits. Farmers who use skilled forecasts have increased average profit levels but also have more variable profits compared with farmers without access to forecasts. Even modest improvements in forecast skill would substantially increase average profits.

An ungated copy can be found here. 

Weekend Links

1) Time lapse photos of the north pole this July

2) Fungicides further implicated in Colony Collapse Disorder (via Yaniv Stopnitzky)

3) Evidence that a cosmic airburst may have triggered the Younger Dryas (PNAS)

4) Bob Pindyck on "Climate Change Policy: What Do The Models Tell US?" (NBER)

5) "We re-interviewed the study participants 20 years after the intervention. [Psychosocial
] stimulation increased the average earnings of participants by 42 percent." (NBER, and not unrelated on FE)

6) Myopic misery

7) Hackathons and public goods provision (via Emmanuel Letouzé)

Please read our paper on climate and human conflict carefully

Edward Miguel, Marshall Burke and I have a new paper quantifying the link between climate and conflict.

There has already been a lot of public criticism of this paper. Marshall has written detailed replies to many of these comments, explaining the why many of these comments are misguided or simply inaccurate.  His reply is on G-FEED here.

I recommend that researchers and journalists read these replies before they further promote inaccurate statements to the public.

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.

Weekend Links

1) "Exposure of children to toxic lead, and the subsequent declines in IQ and earning potential, costs the developing world nearly $1 trillion annually" (via Yaniv Stopnitzky)

2) More evidence on the long term health impacts of in-utero radiation exposure

3) On sustainable (or at least resilient...) concrete

4) Global forced displacement / migration is at an 18 year high (via Thomas Dreesen)

5) Deadline to submit to Scientific American's editorial intern program is July 8th (via John Matson)

6) "We suggest that cultivated flowers are rewarding because they have evolved to rapidly induce positive emotion in humans" (via Michael Balter)

7) Gender Identity and Relative Income within Households (NBER via Amir Jina)

8) The Anti-Science Climate Denier Caucus (via Oceans at MIT)

9) Using quant text analysis to advance the "geek" vs. "nerd" debate

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.

Souped-up Watercolor Regression

I introduced "watercolor regression" here on FE several months ago, after some helpful discussions with Andrew Gelman and our readers. Over the last few months, I've made a few upgrades that I think significantly increase the utility of this approach for people doing work similar to my own.

First, the original paper is now on SSRN and documents the watercolor approach, explaining its relationship to the more general idea of visual-weighting.

Visually-Weighted Regression 

Abstract: Uncertainty in regression can be efficiently and effectively communicated using the visual properties of statistical objects in a regression display. Altering the “visual weight” of lines and shapes to depict the quality of information represented clearly communicates statistical confidence even when readers are unfamiliar with the formal and abstract definitions of statistical uncertainty. Here we present examples where the color-saturation and contrast of regression lines and confidence intervals are parametrized by local measures of an estimate’s variance. The results are simple, visually intuitive and graphically compact displays of statistical uncertainty. This approach is generalizable to almost all forms of regression.

Second, the Matlab code I've posted to do watercolor regression is now parallelized. If you have Matlab running on multiple processors, the code automatically detects this and runs the bootstrap procedure in parallel.  This is helpful because a large number of resamples (>500) is important for getting the distribution of estimates (the watercolored part of the plot) to converge but serial resampling gets very slow for large data sets (eg. >1M obs), especially when block-boostrapping (see below).

Third, the code now has an option to run a block bootstrap. This is important if you have data with serial or spatial autocorrelation (eg. models of crop yields that change in response to weather).  To see this at work, suppose we have some data where there is a weak dependance of Y on X, but all observations within a block (eg. maybe obs within a single year) have a uniform level-shift induced by some unobservable process.

e = randn(1000,1);
block = repmat([1:10]',100,1);
x = 2*randn(1000,1);
y = x+10*block+e;

The scatter of this data looks like:

where each one of stripes of data is block of obs with correlated residuals. Running watercolor_reg without block-bootrapping

 watercolor_reg(x,y,100,1.25,500)

we get an exaggerated sense of precision in the relationship between Y and X:

If we try to account for the fact that residuals within a block are not independent by using the block bootstrap

watercolor_reg(x,y,100,1.25,500,block)

we get a very different result:

Finally, the last addition to the code is a simple option to clip the watercoloring at the edge of a specified confidence interval (default is 95%), an idea suggested by Ted Miguel. This allows us to have a watercolor plot which also allows us to conduct some traditional hypothesis tests visually, without violating the principles of visual weighting. Applying this option to the example above

blue = [0 0 .3]
watercolor_reg(x,y,100,1.25,500,block, blue,'CLIPCI')

we obtain a plot with a clear 95% CI, where the likelihoods within the CI are indicated by watercoloring:

Code is here. Enjoy!

Global Poverty and Practice Postdoctoral Fellows at Berkeley

Blum Center for Developing Economies has a now program for Global Poverty and Practice Post Doctoral Fellows.

(Applications deadline = July 15)