Wednesday, March 31, 2010

The German Experience with Promoting Renewable Energy

RWI, a German institute of economic research, recently published an article called Economic Impacts of Renewable Energy Technologies - The German Experience. Although I have no problem with the authors' arithmetic, per say, I do not quite understand their reasoning. 

The authors accept the astounding success of Germany's feed-in tariffs in expanding renewable energy capacity:
With a share of about 15% of total electricity production in 2008 (Schiffer 2009:58), Germany has more than doubled its renewable electricity production since 2000 and has already significantly exceeded its minimum target of 12.5% set for 2010.
However, they question the efficiency of feed-in tariffs for wind:
Although wind energy receives considerably less feed-in tariffs than PV, it is by no means a cost-effective way of CO2 abatement. Assuming the same emission factor of 0.584 kg CO2/kWh as above, and given the net cost for wind of 3.10 Cents/kWh in 2008 (Table 6), the abatement cost approximate 54 € per tonne. While cheaper than PV, this cost is still more than threefold the current price of certificates in the ETS. In short, from an environmental perspective, it would be economically much more efficient if greenhouse gas emissions were to be curbed via the ETS, rather than by subsidizing renewable energy technologies such as PV and wind power. After all, it is for efficiency reasons that emissions trading is among the most preferred policy instruments for the abatement of greenhouse gases in the economic literature (Bonus 1998:7).
This logic is false for at least two reasons. First, the emissions trading system (ETS) does not automatically and magically turn into emissions reductions. These reductions take place either through lower production levels and higher prices of conventional fossil fuel sources, or through a shift to sources such as wind, which the FIT system is supporting. In the absence of feed-in tariffs, the emissions permits (and hence electricity tariffs on average) would simply be higher. The authors see the low price of permits in the ETS as some sort of merit, but the lower the price, the slower the shift to clean energy sources. Second, the price of emissions permits, and therefore the cost of conventional power generation, is dependent on the amount and allocation of permits in the market (in addition to the opportunity cost). The authors continue:
With respect to climate impacts, the prevailing coexistence of the EEG and the ETS means that the increased use of renewable energy technologies attains no additional emission reductions beyond those achieved by ETS alone. In fact, the promotion of renewable energy technologies ceteris paribus reduces the emissions of the electricity
sector so that obsolete certificates can be sold to other industry sectors that are involved in the ETS. As a result of the establishment of the ETS in 2005, the EEG’s true effect is merely a shift, rather than a reduction, in the volume of emissions: Other sectors that are also involved in the ETS emit more than otherwise, thereby outweighing those emission savings in the electricity sector that are induced by the EEG (BMWA 2004:8).

In the end, cheaper alternative abatement options are not realized that would have been pursued in the counterfactual situation without EEG: Very expensive abatement options such as the generation of solar electricity simply lead to the crowding out of cheaper alternatives. In other words, since the establishment of the ETS in 2005, the EEG’s net climate effect has been equal to zero4.
The authors are certainly right to point out that feed-in tariffs in the range of 40-50 c/kWh for photovoltaics are not the most efficient way to expand clean energy capacity, but that does not mean that feed-in tariffs for wind have been inefficient. The fact that there have been "obsolete certificates" only means just that. There have been too many of them. In essence, the paper builds the whole case against the cost-efficiency of feed-in tariffs for wind by comparing the system to the current or historical prices of ETS permits. Moreover, the argument is based on two assumptions, both of which are false. 1) the current/historical prices of ETS permits reflect the social cost of carbon, and 2) the price of emissions permits would not change in the absence of feed-in tariffs - Wait, let me take that last one back:
These theoretical arguments are substantiated by the numerical analysis of Traber and Kemfert (2009:155), who find that while the CO2 emissions in Germany’s electricity sector are reduced substantially, the emissions are hardly altered at the European scale by Germany’s EEG. This is due to the fact that Germany’s electricity production from renewable technologies mitigates the need for emission reductions in other countries that participate in the ETS regime, thereby significantly lowering CO2 certificate prices by 15% relative to the situation without EEG (Traber, Kemfert 2009:169). In essence, this permit price effect would lead to an emission level that would be higher than otherwise if it were not outweighed by the substitution effect, that is, the crowding out of conventional electricity production through CO2-free green technologies.
By this point, one is inclined to think that the authors can't decide whether low ETS prices are a good or a bad thing, and which is driving the price of which. They essentially argue that the German feed-in tariff system is a subsidy for other countries in the ETS system. This is actually true and not just in the way described above, but also because feed-in tariffs are a form of R&D subsidy for technology companies in renewable energy (and not just for countries in the ETS system, either). However, is this a valid argument against feed-in tariffs? I don't think so. First of all, the ETS system itself, in particular phase 1, has been poorly designed as the market has been flooded with too many permits. Second, the same logic could be used to argue against fiscal stimulus in that it is not worth it because some of the benefits leak abroad. However, that just calls for more coordinated stimulus, not rejecting stimulus as an effective policy.

I am not arguing that the German feed-in tariffs are necessarily at the right level. They most certainly are not. The feed-in tariff for photovoltaics, in particular, is a grossly inefficient R&D subsidy for an immature technology. However, that does not mean that the FIT mechanism, in general, is not well suited to mitigate climate change. We must realize that any system has to be both effective as well as efficient. Efficiency alone is not enough. The ETS and feed-in tariffs are more complements than substitutes. Both are required. FITs have been a tremendous success in terms of expanding clean energy capacity, and in offering investors a safe and stable return on investment. The ETS system, while having lots of potential, needs much more work to induce a similar enthusiasm for clean energy investments. Not only have the permits been priced too low, but the volatility has also been very high, and those who know finance know what that means: higher required returns and hence higher costs of emissions abatement.