Thursday, February 25, 2016

Economic Growth and Particulate Pollution Concentrations in China

A new working paper coauthored with Donglan Zha, who is visiting the Crawford School, which will be published in a special issue of Environmental Economics and Policy Studies. Our paper tries to explain recent changes in PM 2.5 and PM 10 particulate pollution in 50 Chinese cities using new measures of ambient air quality that the Chinese government has published only since the beginning of 2013. These data are not comparable to earlier official statistics and we believe are more reliable. We use our recently developed model that relates the rate of change of pollution to the growth of the economy and other factors as well as also estimating the traditional environmental Kuznets curve (EKC) model.

Though the environmental Kuznets curve (EKC) was originally developed to model the ambient concentrations of pollutants, most subsequent applications have focused on pollution emissions. Yet, it would seem more likely that economic growth could eventually reduce the concentrations of local pollutants than emissions. This is the first application of our new model to such concentration data.

The data show that there isn't much correlation between the growth rate of GDP between 2013 and 2014 and the growth rate of PM 2.5 pollution over the same period:



What is obvious is that pollution fell sharply from 2013 to 2014, as almost all the data points have negative pollution growth. We have to be really cautious in interpreting a two year sample. Subsequent events suggest that this trend did not continue in 2015.

In fact, the simple linear relationship between these variables is negative, though statistically insignificant. The traditional EKC model and its growth rate equivalent both have a U shape curve - the effect of growth is negative at lower income per capita levels and positive at high ones. But the (imprecisely estimated, so not statistically significant) turning point fro PM 2.5 is way out of sample at more than RMB 400k.* So, growth has a negative effect on pollution in the relevant range. When we add the initial levels of income per capita and pollution concentrations to the growth rates regression equation the turning point is in-sample and statistically significant. The initial level of pollution has a negative and highly statistically significant effect. So, there is "beta convergence" - cities with initially high pollution concentrations, reduced their level of pollution faster than cleaner cities did.

So what does all this mean? These results are very different than those we found for emissions of CO2, total GHGs, and sulfur dioxide. In all those cases, we found that growth had a positive and quite large effect on emissions. In some cases, the effect was close to 1:1. Of course, we should be cautious about interpreting this small Chinese data set. But our soon to be released research on global PM 2.5 concentrations, will again show that the effect of growth is smaller for these data than it is for the key pollution emissions data. This confirms early research that suggested that pollution concentrations turn down before emissions do, though it doesn't seem to support the traditional EKC interpretation of the data.

BTW, it is really important in this research to use the actual population of cities and not just the registered population (with hukou). If you divide the local GDP by the registered population you can get very inflated estimates of GDP per capita for cities like Shenzhen.

* The turning point is in-sample for PM 10.

Tuesday, February 23, 2016

Mathiness in Climate Change Econometrics

Terence Mills has a "white paper" on the Global Warming Policy Foundation Website. It predicts little future increase in temperature. Not surprisingly, The Australian has published a totally positive article about it. I commented in the comments there:

"Mills assumes that past fluctuations in temperature are purely random and of unknown causes and ignores greenhouse gases, or the sun, or volcanic eruptions, or any other specific factor that might drive climate change. He then fits simple statistical models based on this assumption to the data. Not surprisingly, if you assume that there isn't any specific factor driving the climate, your best forecast for the future is for not much change because you don't know what random shocks will show up to change the climate in the future. A more sensible approach is to test which of the various proposed drivers might actually have an effect and how large that effect has been. There are a lot of refereed academic papers that do just that including some I published myself. It's pretty easy to show that greenhouse gases have an effect on the climate, it's quite big (but fairly uncertain how big), and if emissions continue on a business as usual path there will be a lot of increase in temperature."

More technically: Mills fits univariate ARIMA models to HADCRUT,  RSS global lower troposphere series (only available since 1980) and Central England Temperature series. These include models with no deterministic component (an ARIMA(0,1,3) model of HADCRUT) and a model with a deterministic trend with breakpoints chosen based on "eyeballing" the temperature graph. None of these models predicts any future warming, because there is no trend in the trendless model and because the "hiatus" means there is no recent trend in the segmented trend model. Of course, a model with just a single linear deterministic trend fitted to HADCRUT data would forecast a lot of warming in the 21st Century, though with a very wide forecast error envelope. But that model isn't estimated, for some reason...

This is a prime case of "mathiness" I think - lots of math that will look sophisticated to many people used to build a model on silly assumptions with equally silly conclusions.

In other news, my paper coauthored with Luis Sanchez on drivers of greenhouse gas emissions is now published in Ecological Economics. It is open access till 12th April.

P.S. This post was cited in the Daily Mail.

Saturday, February 13, 2016

Family Portrait

This will slow things down for a while :) Noah was born two days ago. He is a very large baby - 4.71kg and 55cm long. If he was a t-statistic, he'd have 2 or 3 stars. He'll have to make do with one, his surname.*

* Stern = Star in German.