In intro to comparative, we have generally compared spatially across countries (or states). There is a lot of explanatory power, however, that can be achieved by modeling and comparing political phenomena at the sub-national level. Maximilian Auffhammer and Richard Carson–two economists–have done this by modeling the rise in Chinese CO2 emissions, using a panel data set at the provincial level in China. Their data set includes 30 provincial-level entities (or provinces) analyzed between 1985 and 2004. See a post I made on China’s pollution problems and economic growth here. I’ve reproduced a slide show from the post below. Here’s a snippet from their paper, which can be accessed here:
The People’s Republic of China (PRC) has long been seen as the key future participant to an effective agreement limiting the adverse impacts of climate change. It is currently the number two emitter of carbon dioxide (CO2) and is about to overtake the United States, who has held this position since 1890, as the leading emitter. Further, the United States has long preconditioned its adherence to any international agreement such as the Kyoto Protocol on China’s formal concurrence that it would also undertake substantial CO2 reductions. Efforts to reach such an agreement failed in the late 1990’s during the Clinton administration and the Bush administration decided not to pursue policies that would allow it to sign the treaty and have it rati¯ed by the U.S. Senate.
This paper presents econometric forecasts that strongly suggest that the short to medium term path of Chinese CO2 emissions has increased by a factor of two or more since that time. Our best forecast has China’s CO2 emissions surpassing the United States before the year 2010 rather than 2020 as previously anticipated (Intergovernmental Panel on Climate Change, 2000; Siddiqi, Streets, Wu and He, 1994; Panayotou, Sachs and Zwane, 2002). Our focus in this paper is on exploring alternative econometric specifications for forecasting China’s CO2 emissions using a rich new panel dataset from 1985 to 2004 at the provincial level. The prediction of a dramatic recent increase in the predicted path of China’s CO2 emissions over the short to medium term horizon is shown to be robust to a wide range of alternative specifications. We show, however, that it is possible to strongly reject both the standard engineering specifications that appear in the Intergovernmental Panel on Climate Change (2000), and the recent Stern Report (2006) as well as the popular environmental Kuznets curve specification. All of the “best” models are dynamic in nature employing some type of lag structure, which is consistent with the nature of an installed durable capital stock.