Climate Change: Past and Present Conference

May 30-31, 2009
Gary Libecap and Richard H. Steckel, Organizers

John Landon-Lane, Rutgers University, Hugh Rockoff, Rutgers University and NBER, and Richard H. Steckel, Ohio State and NBER
Droughts, Floods and Financial Markets in the United States

The relationships among the weather, agricultural markets, and financial markets have long been of interest to economic historians, but relatively little empirical work has been done. Landon-Lane, Rockoff, and Steckel push this literature forward by using modern drought indexes, which are available in detail over a wide area and for long periods of time, to perform a battery of tests on the relationship between these indexes and sensitive indicators of financial stress. The drought indexes were devised by climate historians from instrument records and tree rings, and because they are unfamiliar to most economic historians and economists, the authors briefly describe the methodology. The financial literature in the area can be traced to William Stanley Jevons, who connected his sun spot theory to rainfall patterns. The Dust bowl of the 1930s brought the climate-finance link to the attention of the general public. Here the researchers assemble new evidence to test various hypotheses involving the impact of extreme swings in moisture on financial stress.

Karen Clay, Carnegie Mellon and NBER, and Werner Troesken, George Mason University and NBER
Did Frederick Brodie Discover the World's First Environmental Kuznet's Curve, and If So, Why Should Anyone Really Care?

In 1903, Frederick Brodie presented data showing that the incidence of fog in London was rising before 1890 and falling thereafter. Brodie attributed this inverted-U shaped pattern to changes in the production of coal smoke. If Brodie’s data and interpretation prove sound, he identified an archaic Environmental Kuznets Curve. Clay and Troesken subject Brodie’s data and interpretation to a battery of tests using data on hours of sunlight, coal consumption in the metropolis, weekly death rates, and bronchitis death rates. These sources of evidence, for the most part, corroborate Brodie’s original claims.

Price Fishback, University of Arizona and NBER, Trevor Kollman, University of Arizona, Michael Haines, Colgate College and NBER, Paul Rhode, University of Arizona and NBER, and Melissa Thomasson, Miami University and NBER
The Trials of Job: Impact of Climate Change and Weather on Infant and Non-Infant Death Rates During the Great Depression

Fishback, Haines, Kollmann, Rhode, and Thomasson measure the impact of climate on infant mortality and non-infant mortality throughout the Great Depression using a database that combines information on mortality rates, daily high temperatures, and inches of precipitation, and a rich set of socioeconomic correlates for over 3,000 counties in the United States for each year between 1930 and 1940. Variations across the country in climate were associated with differences in infant mortality and non-infant death rates. However, much of the influence of climate is muted a great deal once information correlates are included in the analysis. Both infant mortality and non-infant mortality rates were higher in areas where there was more illiteracy and lower in areas where people had more access to radios and the circulation of news magazines was greater. These effects are more than income effects because we control for urbanization and economic activity in the analysis. Of even more interest is how strongly the influence of climate on mortality changed when we add the information variables to the analysis. Many of the positive temperature coefficients that implied higher mortality with higher temperature were sharply cut in magnitude by the inclusion of the information coefficients. This is consistent with a view that access to better information reduced the impact of climate on mortality.

Anin Aroonruengsawat, UC, Berkeley, and Maximillian Auffhammer, UC, Berkeley
Impacts of Climate Change on Residential Electricity Consumption: Evidence from Billing Data

Aroonruengsawat and Auffhammer simulate the effects of higher temperatures resulting from anthropogenic climate change on residential electricity consumption for California. They estimate flexible temperature response functions by climate zone, which allows for differential effects of days in different temperature bins on households' electricity consumption. The estimation uses a comprehensive household level dataset of billing data for California's three investor-owned utilities (Pacific Gas and Electric, San Diego Gas and Electric, and Southern California Edison). Their results suggest that the temperature response varies greatly across climate zones. Using a downscaled version of the National Center for Atmospheric Research global circulation model, the simulation results suggest that -- holding population constant -- total consumption for the households considered may increase by up to 55 percent by the end of the century. The study further simulates the effects of higher electricity prices and different scenarios of population growth. Finally, simulations consistent with higher adoption of cooling equipment in areas which are not yet saturated, as well as gains in efficiency potentially because of aggressive energy efficiency policies, were conducted.

Morgan Kelly, University College, Dublin, and Cormac O’Grada, University College, Dublin
Did Climate Matter? The Little Ice Age and European Growth

Kelly and Ó Gráda focus on the claim that a Little Ice Age constrained demographic and output growth in northern Europe between the Middle Ages and the nineteenth century. They analyze several long-range climate estimates to see which square best with measured temperature and rainfall data when such data became available. This leads them to argue that summer and winter temperature and rainfall each year were effectively independent draws from an unchanging distribution c. 1300-1900. In other words, climate was constant, so whatever variations in economic growth occurred during this time were not attributable to climatic swings. They also show, however, that short-term fluctuations in the most reliable of their weather series had significant effects on variations in crop yields in the fourteenth and fifteenth centuries and on variations in wages and living standards up to 1800. Finally, they invoke data on demographic and agricultural trends that are more consistent with climate stasis during the “Little Ice Age” than with deterioration.

Wolfram Schlenker, Columbia University and NBER, and Michael Roberts, North Carolina State University
The Evolution of Heat Tolerance of Corn: Implications for Climate Change

Roberts and Shlenker examine how precipitation, moderate temperatures, and extreme temperatures influenced corn yields in Indiana between 1901 and 2005. Using a fine-scale weather dataset of daily weather records, they find that the effects of precipitation and extreme heat evolved over time. While the detrimental effect of either too much or too little water seems to have steadily diminished over time, the evolution of tolerance to extreme heat is highly nonlinear, growing with the adoption of hybrid corn in the 1940s, peaking around 1960, and then declining. Corn in Indiana is most sensitive to extreme temperatures at the end of their sample. Since climate change models predict an increase in extreme temperatures, the big question is whether the next breeding cycles can increase both average yields and heat tolerance simultaneously as in the period 1940-60, or whether a continued increase in average yields can only be achieved at the expense of more sensitivity to extreme heat, as in the period from 1960 onwards.

Price Fishback, University of Arizona, NBER, Janathan Fox, University of Arizona, and Paul Rhode, University of Arizona, NBER
The Economic Response to Climate Change in the Farm Sector: The United States, 1895-1969

Fishback, Fox, and Rhode use a 75-year panel of states to identify the effect of weather fluctuations on different types of agricultural commodities. Using information from the USDA and National Climatic Data Center, they estimate these effects for corn, a crop with high transport costs and used in other local productive activities, and cotton, a crop with relatively low transport costs and few local uses. Results indicate that for crops such as cotton that are sold primarily in international markets, changes in local weather have little effect on farmgate prices. However, crops with strong local markets such as corn are much more sensitive to changes in state-level temperature, precipitation, and drought conditions. Attempts to control for farmer expectations did not substantially affect the coefficient estimates, suggesting that while long-term crop mix adjustment may still affect the plausibility of the estimates, short-term crop mix adjustments do not.

Alan L. Olmstead, UC, Davis and NBER, and Paul W. Rhode, University of Arizona and NBER
Adjusting to Climatic Variation: Historical Perspectives from North American Agricultural Development

Providing greater historical perspective would enlighten current discussions about future human responses to climatic variation. During the nineteenth and twentieth centuries, new biological technologies allowed North American farmers to push cropping into environments previously thought too arid, too variable, and too harsh to cultivate. Olmstead and Rhode document these changes for three major staple crops, noting that the climatic challenges that previous generations of farmers overcame often rivaled the climatic changes predicted for the next hundred years in North America.

Zeynep K. Hansen, Boise State University and NBER,Gary D. Libecap, UC, Santa Barbara and NBER, and Scott E. Lowe, Boise State University
Climate Variability and Water Infrastructure: Historical Experience in the Western United States

Hansen, Libecap, and Lowe analyze the impact of climatic conditions and variability on agricultural production and flood control in the western states, which are characterized by the continent’s driest and most variable climate. They have assembled county-level data on dams and other major water infrastructure; agricultural crop mixes and yields; precipitation and temperature; soil quality, and topography. Using this extensive dataset, they analyze the impact of water infrastructure investments on crop mix and yields and the incidence of floods in affected counties relative to similarly-endowed counties that lack such infrastructure. They anticipate that water infrastructure will smooth agricultural crop mixes and output, and reduce flooding. In addition, they explore the political economy of the Reclamation Act of 1902 to cast light on how politics and climatic factors may influence contemporary investment decisions.

Robert S. Pindyck, MIT and NBER
Uncertainty, Extreme Outcomes, and Climate Change Policy

Focusing on tail effects — low probability but very adverse outcomes — Pindyck incorporates distributions for temperature change and its economic impact in an analysis of climate change policy. He estimates the fraction of consumption that society would be willing to sacrifice to ensure that any increase in temperature at a future point is limited. Using information on distributions for temperature change and economic impact from studies assembled by the IPCC and from “integrated assessment models” (IAMs), he fits displaced gamma distributions for these variables. Unlike existing IAMs, his model takes economic impact as a relationship between temperature change and the growth rate of GDP, not its level, so that warming has a permanent impact on future GDP. The fitted distributions for temperature change and economic impact generally yield values of the fraction of consumption that society would be willing to sacrifice below 2 percent, even for small values of any limit of increase in temperature at a future point -- unless one assumes extreme parameter values and/or substantial shifts in the temperature distribution. These results support adoption of only moderate abatement policies.

Richard Sutch, UC, Riverside and NBER
The Impact of the 1936 Corn-Belt Drought on American Farmers’ Adoption of Hybrid Corn

Sutch notes that the severe drought in 1936 revealed an advantage of hybrid corn not previously recognized – its drought tolerance. This revealed ecological resilience motivated some farmers to adopt hybrids despite their commercial unattractiveness in normal years. But that response to climate change had a tipping effect. The increase in sales of hybrid seed in 1937 and 1938 financed research at private seed companies that led to new varieties with significantly improved yields in normal years. This development provided the economic incentive for late adopters to follow suit. Because post-1936 hybrid varieties conferred advantages beyond improved drought resistance, the negative ecological impact of the devastating 1936 drought had the surprising, but beneficial, consequence of moving more farmers to superior corn seed selection.

Raghav Gaiha, University of Delhi, India, Kenneth Hill, Harvard University, Antanu Mathur, International Fund for Agricultural Developmentm, and Vani S. Kulkani, Harvard University
On Devastating Droughts

Gaiha and his co-authors note that while the frequency of droughts has risen, their deadliness has declined. Their analysis throws light on the underlying geographical, institutional, and development indicators in explaining inter-country differences in mortality. They also confirm the favorable effects of openness in saving human lives: they illustrate that much of this devastation is avoidable -- through a timely and speedy entitlement protection strategy. Their simulations yield additional insights: for example, even moderate learning has the potential to avert a large fraction of deaths. But capacity-building, which is synonymous with availability of more resources for disaster prevention, has considerable potential too in averting deaths. In fact, these findings are broadly consistent with the view that fatalities are greater in countries with weak governments and pervasive poverty. How do democratic regimes help to avert mortalities? Government responsiveness is greater when the severity of the crisis is greater. Also, voters punish incumbent politicians for crises beyond their control. But voters also reward politicians for responding well to climatic events, but not sufficiently to compensate them for their “bad luck.” Even within a democratic regime, there are marked differences in the ability to prevent starvation deaths. Competitive local politics and decentralized structures of governance are crucial. Specifically, local political parties and vigilant village councils act not just as conduits of information on distress but also pressure district administration to take appropriate action. If the goal of development is security of livelihoods and human lives, a broader strategy is called for, a strategy that goes well beyond protection of food entitlements of the vulnerable. Some key elements include higher agricultural research outlays, public-private partnerships in promoting pro-poor technologies, a compatible incentive structure, and more effective extension systems. Specifically, soil and water conservation technologies with effective community participation deserve high priority in arid, semiarid, and sub-humid regions/areas. As large sections of the rural population in developing countries will continue to be vulnerable to various shocks -- droughts, pests, famines, floods, among others -- insurance also has a potentially important role in mitigating the hardships. In conclusion, while building resilience against natural disasters, such as droughts, is a challenge for developing countries, the prospects are far from bleak.

Martin Weitzman, Harvard University and NBER
Additive Damages, Fat Tailed Climate Dynamics, and Uncertain Discounting

Weitzman argues that there is a loose chain of reasoning connecting the following three basic links in the economics of climate change: 1) additive damages may be more appropriate for analyzing the impacts of global warming than multiplicative damages; 2) an uncertain feedback-forcing coefficient, which might be near one with infinitesimal probability, can cause the distribution of the future time trajectory of global temperatures to have fat tails and a high variance; 3) when high-variance additive damages are discounted at an uncertain rate of pure time preference, which might be near zero with infinitesimal probability, it can make expected present discounted disutility very large. Some possible implications for welfare and climate-change policy are briefly noted.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us