|
Article Excerpt
1. INTRODUCTION
On the European Union Emissions Trading Scheme (EU ETS) launched in 2005 to help Member States (MS) achieve their Kyoto Protocol's (KP) target to reduce 1990 emissions by 8% during 2008-2012, covered installations are only allowed for banking and borrowing allowances within 2005-2007 (Phase I) and 2008-2012 (Phase II). EU allowances (EUAs) are issued annually on industrial accounts, and are valid to cover emissions within the commitment period. This annual issuance of allowances occurs at the end of February, two months before allowances must be surrendered for the preceding year in April. Each year, regulated agents may bank surplus allowances, equal to endowments less of emissions, of a specific vintage for potential later use. Conversely, if regulated agents do not abate enough to cover their emissions level with their actual endowment, they may borrow allowances from future vintages of allocations. Hence, the possibility to transfer allowances to the next years is referred to as banking of emission units. The early use of allowances issued in the present for later years is instead referred to as borrowing of emission units.
A fundamental statement about emissions trading is that it is efficient over time only if banking and borrowing are authorized (Rubin (1996), Schennach (2000), Leiby and Rubin (2001)). With such provisions, allowance prices reflect opportunity costs leading to an efficient choice of abatement measures (Schleich, Ehrhart, Hoppe and Seifert (2006)). The most prominent example of the key role of such provisions is the U.S. Acid Rain Program, where banking has been a major feature in the success (1) of this emissions trading scheme (Ellerman, Joskow, Schmalensee, Montero and Bailey (2000), Ellerman and Montero (2007)).
To our best knowledge, there has been no rigorous empirical analysis of the impacts of intra-period banking coupled with a ban on inter-period banking on EUAs price changes. In a game simulation, Ehrhart, Hoppe, Schleich and Seifert (2005) found that a ban on banking leads to an inefficient adjustment with i) an under-investment in abatement technologies and a low allowance price during 2005-2007, and ii) a more stringent cap with a price peak and over investment in emissions reduction during 2008-2009.
Among the main explanations of low allowance prices towards the end of Phase I, previous literature identifies over-allocation concerns, early abatement efforts in 2005 due to high allowance prices, and possibly decreasing abatement costs in 2006 due to abnormal temperatures and switching from coal- to gasfired electricity in a context of falling natural gas prices compared to coal (Ellerman and Buchner (2008), Mansanet-Bataller, Pardo and Valor (2007), Alberola, Chevallier and Cheze (2008)). Therefore, a thorough analysis of banking and borrowing provisions seems missing. It appears necessary to disentangle those effects on allowance prices, which are expected to develop differently in the following two cases. If inter-period banking is allowed, it is reasonable to expect that allowance price changes do not exceed Hotelling's rule, rising at the market rate of interest. (2) If inter-period banking is restricted, lower Phase I prices and higher Phase II prices are expected. The former result is due to the validity of allowances which is shorter than the time horizon required by market agents. The latter is due to increased allowance scarcity compared to full inter-period banking.
Within 2005-2007, EU ETS participants are allowed to use unrestricted banking and borrowing. Based on this intra-period banking provision, we test whether the allowance price pattern is consistent with a competitive equilibrium in the intertemporal market. The theoretical models by Schennach (2000) and Slade and Thille (1997) of the intertemporal allowance market, applied by Helfand, Moore and Liu (2006) to the U.S. S[O.sub.2] market, guide our Hotelling-CAPM type analysis.
Between 2007 and 2008, participants are not allowed to use banking and borrowing mechanisms. To identify the impacts of the inter-period ban on banking on daily price variations, we study the relationships between the EUA spot price and futures prices of December 2006, 2007 and 2008 contracts. Thus, we aim at evaluating whether the EUA futures price has the power to forecast consistently the future EUA spot price (Fama and French (1987)). Besides, the impact of the late restriction on inter-period banking is evaluated through dummy variables representing official communications between France, Poland and the European Commission (EC).
Compared to the literature on efficient banking in the U.S. S[O.sub.2] Program, our results are twofold. Concerning intra-period banking provisions, we show the Hotelling rule does not hold during 2005-2007, which confirms the first commitment period of three years did not meet successfully the necessary conditions for an efficient intertemporal price development. Concerning inter-period banking provisions, we observe a divorce between the EUA spot price, which is steadily decreasing towards zero, and the EUA price of the December 2008 futures contract, which is stabilizing around 20[euro]. Our statistical analysis reveals the cost-of-carry relationship between spot and futures prices for delivery in Phase II does not hold from October 2006 until the end of Phase I. This result suggests that the restriction to transfer non cancelled allowances between Phases I and II undermines the ability of the EU ETS to provide an efficient price signal for emissions abatement. Besides, the submissions and final decisions of the French and Polish National Allocation Plans (NAPs) for Phase II confirm this statistically significant effect of banning banking in explaining low allowance prices until the end of Phase I. These results are robust to the interaction with energy markets and extreme temperatures events previously identified as being the main determinants of EUA prices changes in the literature.
The remainder of the article is organized as follows. Section 2 details the banking and borrowing provisions in the EU ETS. Section 3 introduces our econometric analysis. Section 4 presents the data. Section 5 contains the results and a discussion. Section 6 concludes.
2. BACKGROUND
This section reviews the environmental and economic effects of banking and borrowing, the motives that led EU Member States to ban banking and borrowing between 2007 and 2008, and the allowance prices development along with its characteristic structural break in April 2006.
2.1 Environmental and Economic Effects of Banking
A fully intertemporally flexible emissions trading scheme presents a superior environmental and economic efficiency to a scheme in which the transfer of allowances is restricted.
In terms of environmental effects, allowance banking and borrowing may change the temporal path of emissions and aggregate emissions. While banking reduces social damages in presence of a convex damage function coming from emissions and stricter future standards (Kling & Rubin (1997)), unrestricted borrowing may have negative consequences with a concentration of emissions on early periods by delaying abatement decisions. To correct these unwanted allowance paths, the regulator may introduce a non unitary Intertemporal Trading Ratio (ITR) (3) including interests on banking and discouraging borrowing: if firms borrow a lot of allowances during early periods, they will reimburse more allowances than actually used during the next periods. Furthermore, banking provisions could affect the rate of non-compliance and the resulting excess emissions. (4)
In terms of economic effects, the theoretical literature suggests allowance banking and borrowing may improve economic efficiency under specific assumptions. (5) First, banking links futures allowance prices to spot prices as stated by Maeda (2004). Second, banking and borrowing improve price stability (Ellerman and Montero (2002)). If inter-period banking is not allowed, allowance prices are likely to be more volatile at the end of each compliance period. If firms are left with a surplus of allowances with respect to their emissions cap, allowances are worthless, and their price should fall to zero. If firms are left with a deficit of allowances with respect to their emissions cap, the allowance price should rise sharply at the end of the period. Third, banking and borrowing improve the liquidity of the allowance market by increasing the quantity of allowances available for trading, (6) and the volume of allowances traded (Godby, Mestelman, Muller and Welland (1997)). Fourth, allowance banking and borrowing should facilitate adjustment to changes in the emissions cap, especially if the regulator enforces stricter targets, as for Phases II and III of the EU ETS.
By using a regional model of the world economy (7) with endogenous technical change, (8) Bosetti, Carraro and Massetti (2009) validate empirically some insights on the implications of intertemporal allowances transfer. Banking not only increases the economic efficiency, but also the environmental effectiveness of climate policy, particularly in the short term. Indeed, banking increases the amount of emissions abated in the first decades, thus reducing the risk of irreversible environmental impacts on climate. Besides, banking plays an important role in accelerating the adoption and diffusion of energy efficient and carbon free technologies. Overall, the authors find evidence that significant cost savings (9) are indeed possible by allowing a full intertemporal flexibility.
Following this review of the various effects of banking and borrowing that need to be accounted for when tailoring environmental regulation, we describe in the next section the reasons that led EU Member States to the interperiod ban on banking.
2.2 Reasons to Ban Banking Between 2007 and 2008
In the Directive 2003/87/EC, (10) the EC initially left to Member States (MS) the decision to allow or not the transfer of banked and borrowed allowances from Phase I to the next compliance periods. Later on, the EC statement of its assessment methodology for the review of NAPs II effectively negated these provisions, by requiring that any banked EUAs must be deducted from the Phase II cap. (11)
Discussions concerning the banking and borrowing provisions were indeed characterized by sudden decision changes. The EU ETS was launched on January 1, 2005 while some NAPs I were still under negotiation. After the validation of all NAPs I, all MS decided, with the exception of Poland and France, against the inter-period transfer of allowances: all allowances not surrendered by the end of 2007 will be cancelled, and not transferred to...
|
