Understanding the Link Between Aggregated Industrial Production and the Carbon Price - Green Energy and Efficiency: An Economic Perspective - page 121

framework. This econometric strategy will allow us to decompose the joint varia-
tion of carbon prices and industrial production in two (high and low) regimes.
As such, it provides a useful and updated extension of the studies by Chevallier
[ 9 ], SETAR and STAR models), Chevallier [ 10 ], two-regime Threshold cointe-
gration), and Chevallier [ 11 ], two-regime Markov-switching VAR model).

4.1 Data

Let us rst present the data used in this study. The dataset contains CO 2 futures

prices and the EU 27 Industrial production index, which were obtained from the
European Climate Exchange (ECX), Thomson Financial Datastream, and Eurostat.
The data sample goes from the opening of ECX on April 22, 2005 to January 25,
2013 (i.e. a sample of 2,008 daily observations).
The carbon price is the ECX EUA Futures price series in EUR/ton of CO 2 ,
rolled-over using front months contracts. In addition, concerning the macroeco-
nomic variable of interest, we follow thoroughly the approach by [ 10 ], who selected
the EU 27 Industrial Production Index by Eurostat as the variable of interest to be
used as a proxy for the in

uence of economic activity (to cope with the limitation of
being unable to observe actual CO 2 emissions at the plant level). 10
Both series are pictured in Fig. 2 . Concerning the EU industrial production (on
the right Y-axis), we may distinguish three distinct phases during our study period.
First, the period going from January 2005 to May 2008 may be viewed as a phase
of economic growth. Second, we notice after May 2008 an abrupt decline in the
industrial production, characterizing the entry of EU economies into the recession.
These events follow with some delay the developments of the US economy, fol-
lowing the rst interest rate cut by the Federal Reserve in July 2007. This event is

mostly viewed as the start of the economic downturn, as the nancial

distress in the housing sector met the headlines. Third, from April 2009 until July
2010, we may observe a timid uptake in the industrial production. Therefore, our
study period contains an interesting mix of economic growth, recession and
recovery that we aim at analyzing jointly with the behavior of EUA Futures prices
(on the left Y-axis). The latter time series seems to follow the same pattern, with the
presence of shocks during 2005

rst signs of

2007 originating from institutional features of the
EU ETS (see [ 16 ] for an exhaustive coverage of this topic).
Descriptive statistics are presented in Table 3 . They provide useful information on
the distributional characteristics of the time series considered, and more especially

-

10 The EU 27 industrial production index has a base 100 in 2000, and is seasonally adjusted. The
index is converted from monthly to daily frequency by using the Matlab function by L. Shure,
which performs linear interpolation so that the mean square error between the original data and
their ideal values is minimized.