Assessing the impact of the EU ETS using firm level data. Jan Abrell, Anta Ndoye Faye, Georg Zachmann

Transcription

1 Assessing the impact of the EU ETS using firm level data Jan Abrell, Anta Ndoye Faye, Georg Zachmann

2 The EU Emission trading scheme (EU ETS) Incentivise the most economic mitigation efforts by Capping carbon emissions Making carbon permits a tradable good 6 years of experience with ETS imply an empirical assessment of Its effectiveness (did it deliver reductions) Its impact on the company s performance (added value, profit margin, employment)

3 Literature Effectiveness: Ellerman and Buchner (2006), Grubb and al. (2009), Ellerman and al. (2010) Results: main reduction of emission took place between Drawback: Short period analysis (2 or 3 years), lack of counterfactual Competitiveness: D ly and Quirion (2008), Anger and Oberndorfer (2008) Results: modest competiveness losses in the first phase Drawback: Static analysis, sector specific or country specific, lack of counterfactual

4 The two phases of the EU ETS : trial phase (first phase) Grandfathering of allowances No banking : second phase Mainly Grandfathering of allowances Banking emissions in the regulated sectors increased by 2 % due to a high cap and the import of offsets Tighter cap, the amount of allowances distributed was reduced from 2007 to 2008 by about 11 percent Daily Closing Price EUA spot

5 ETS emissions by country Phase I Phase II Crisis year

6 ETS emissions by sector

7 Data CITL emission data Allocated emissions Verified emissions Allocation factor: the quotient of free allocation of emissions allocated to the verified emissions AMADEUS Profit Margin, Employment, Turnover, Added Value, Fixed Capital Both sets matched via adresses For CITL installations we got 2101 AMADEUS companies(3696 installations), representing 59% of the emissions Company level data allows us to: Control emission reductions for company performance Analyse whether being included in the ETS had an effect on company performance

8 Meadian allocation factor 1,16 1,14 1,12 1,1 1,08 1,06 Germany 1, Amadeus companies CITL Median allocation factor 1,3 1,25 1,2 1,15 1,1 Poland 1, Amadeus companies CITL Not a perfect match but reasonable!

9 Effectiveness of the EU ETS Our goal is to analyse whether companies changed their emission reduction strategy from to We use a diffs in diffs in diffs approach applied to the following equation y t it d 0 1 it cv 1, it 2005,2006,2007, cv 3 2, it it => thus, we analyse the change in speed in emission reduction between the first and the second phase

10 Results Dependant variable Growth rate of emissions ( ) - Growth rate of emissions ( ) Total Sample(1) AF05 <1.15(2) AF05>1.15(3) Phase dummy **(0.015) *** (0.011) (0.03) changes in 0.191***( 0.028) 0.19***(0.04) 0.21***( 0.04) turnover changes in labour (0.025) ( 0.02) 0.07 (0.05) size Adj R-squared Significance: * at 10%, ** at 5 % and *** at 1%. Standard errors are reported in brackets The tighter cap in Phase II was effective since emissions decrease Under allocated companies in allowances for 2005 reduced their emissions the most

11 Results Paper and paper products 416 firms Non-metallic minerals 806 firms Basic metals 159 firms Electricty heat 660 firms Phase dummy (0.027) ***(0.025) *(0.049) (0.038) changes in turnover changes in labor size 0.154**( 0.077) 0.299***(0.058) 0.089(0.126) 0.136**(0.06) (0.093) (0.044) 0.099(0.208) 0.012(0.042) Adj R-squared Significance: * at 10%, ** at 5 % and *** at 1%. Standard errors are reported in brackets Non metallic minerals and basic metals carried out the main reduction effort No significant effect of moving to a tighter cap for energy and paper sectors

12 Impact of EU ETS on the firm performance Impact of the EU ETS on firm added value, employment (labour size) and profitability (profit margin) With a matching procedure based on a propensity score, we construct a control group in order to assess this counterfactual effect The following two period fixed effect model (before and after treatment) is then estimated in first differences: y it 0 1 d 1,it 2 d 2,it 3 x it 4 cv it it,t 0,1 y i 1 2 d 2,i 3 x i 4 cv i u i ˆ 2 y T y C

13 Added Value Dependent variable Added value Total sample (4202 firms) Period (1)= (2)= Impact of EU ETS (0.08) (0.08) Changes in fixed capital 0.08***(0.01) 0.06***(0.01) Changes in employment 0.11***(0.01) 0.10***(0.02) Adj R-squared Under allocated firms (AF<1) (1) 1436 firms (2) 1538 firms Impact of EU ETS (0.04) (0.06) Changes in fixed capital 0.08***(0.01) 0.11***(0.01) Changes in employment 0.16***(0.02) 0.17***(0.02) Adj R-squared Over allocated firms (AF>1) (1) 2766 firms (2) 2664 firms Impact of EU ETS (0.07) -0.12(0.10) Changes in fixed capital 0.05**(0.02) 0.07***(0.02) Changes in employment 0.08***(0.02) 0.09***(0.02) Adj R-squared No impact on added value for the two periods

14 Profit Margin Dependent variable Profit margin Total Sample Period (1)= (2)= Impact of EU ETS (0.45) *(0.37) Changes in employment -0.59*(0.32) -0.52(0.32) Changes in turnover 3.91***(0.21) 3.67***(0.21) Adj R-squared Under allocated firms (AF<1) Impact of EU ETS (0.31) *(1.11) Changes in employment -0.42(0.43) -0.34(0.43) Changes in turnover 2.61***(0.27) 2.54(0.27) Adj R-squared Over allocated firms (AF>1) Impact of EU ETS 2.14* (1.25) 2.32 *(1.29) Changes in employment -0.95**(0.50) -0.87*(0.49) Changes in turnover 5.29***(0.35) 5.07***(0.34) Adj R-squared Overallocated firms have benefited significantly from the EU ETS in terms of profitability Underallocated Firms did lose Overall effect surprisingly negative

15 Employment Dependent variable Employment Total sample Period (1)= (2)= Impact of EU ETS (0.002) **(0.004) Changes in employment 0.50***(0.002) 0.52***(0.02) Changes in turnover 0.04***(0.02) 0.05***(0.02) Adj R-squared Under allocated firms (AF<1) Period (1) (2) Impact of EU ETS (0.003) (0.095) Changes in employment 0.49***(0.002) 0.50***(0.002) Changes in turnover 0.04***0.003) 0.03***(0.003) Adj R-squared Over allocated firms (AF>1) Period (1) (2) Impact of EU ETS **(0.004) (0.002) Changes in employment 0.52***(0.002) 0.51***(0.003) Changes in turnover 0.05***(0.004) 0.06 ***(0.005) Adj R-squared Overallocated firms have increased slightly their employment level after the launch of EU ETS.

16 Conclusion Phase II is promising compared to Phase I The magnitude of the impact appears small Limited impact on economic performance Extension 1: Include more years of the ETS to capture longer-term effects such as investments Extension 2: Effectiveness regression performed on a susbsample division that take into account the reduction of allowances distributed in 2008 compared to 2007

17 APPENDIX

18 Excess allocation by sector and country

19 Median Allocation factor SAMPLE REPRESENTATIVENESS France 1,5 1 0,5 Median allocation factor Median allocation factor UK 1,2 1,15 1,1 1,05 1 0, Poland 1,3 1,25 1,2 1,15 1,1 1, Amadeus companies CITL Amadeus companies CITL Amadeus companies CITL

20 Median Allocation Factor 1,25 1,2 1,15 1,1 1,05 Spain Amadeus companies CITL

21 Sectoral and regional distribution (Amadeus Companies) Sectors Number of firms Frequency (%) Other non-metallic mineral products Electricity and heat Paper and paper products Basic metals Coke and refined petroleum products Countries Spain Germany Portugal France Czech Republic Poland Italy Finland UK-Ireland Bulgaria- Romania Sweden Austria Belgium-Luxembourg Slovakia Netherlands Denmark Slovenia-Hungary Estonia-Latvia-Lithuania

22 Descriptive statistics Added value Employees Fixed Capital Profit Margin Allocation Factor 1% % % Median Mean % % Std