Making low-carbon technology support smarter

While the impact of increasing concentrations of greenhouse-gases in the atmosphere on the climate system cannot be accurately predicted, there is a non-trivial risk that beyond some ex-ante unknown tipping points – in terms of greenhouse-gas concentration and/or global temperature – irreversible and highly expensive events might unfold. This calls for quick action to reduce the probability that such tipping points will be passed2. Consequently, annual greenhouse gas emissions will have to be reduced dramatically before 2050. In order to stabilize CO2 concentrations at about 450 ppm3 by 2050, global emissions would have to decline by about 40-70% by 2050.

Such aggressive decarbonisation on a global scale will require an international agreement because otherwise fossil fuels not used in some countries will simply be used in other countries4. And we cannot wait until low-carbon technologies become cheaper than expiring fossil fuels because, in particular, the last percentage points of cost advantage that fossil fuels have will be difficult to overtake without a price on carbon5. But an agreement is only feasible and stable if the climatebenefit for each country exceeds the cost. The cost of decarbonisation essentially depend on the cost of low-carbon technologies. Consequently, reducing the cost of these technologies in Europe not only allows for cheaper domestic decarbonisation and for a competitive edge to be gained in selling these technologies overseas, but most importantly it would strongly facilitate an international agreement.
Without public intervention, European companies will under-invest in low-carbon innovation for three reasons: (i) in all sectors, innovators cannot reap the full benefits of their innovation because good ideas might be used to enhance productivity beyond the product made by the original inventor (e.g. by inspiring new innovation or being merely copied by competitors). No company will invest in a project for which the expected return is below the upfront investment, even if the societal benefits exceed the initial investment cost. (ii) The European carbon price is likely to be below the social cost of carbon and there is no sufficient long-term visibility of the carbon price-signal. As companies will
only invest in technologies that mitigate CO2 emissions at a cost below the carbon price, investments in technologies with higher abatement cost (e.g. carbon capture and storage) are not brought forward, even though they might be needed to mitigate climate change. (iii) Low-carbon technologies are most competitive in markets where greenhouse-gas emissions are regulated.

Hence, even though emissions outside the EU are as bad for the planet as emissions within the EU, companies invest at less than socially-optimal levels in developing low-carbon technologies, because they do not receive extra remuneration for the ability of their technology to reduce emissions in markets that do not regulate emissions.

The policy question hence is: how can the EU overcome in the most efficient way these market failures that hold back low-carbon innovation. We first provide some evidence that companies that do ‘green innovation’ appear not to be different from other innovators. Then we discuss what public policies are used to support low-carbon innovation and where we see room for improvement.

PATSTAT and links between innovation and ETS

Patent data are a crucial source of information on innovation. But despite enormous efforts by the European Patent Office the PATSTAT database requires substantial processing to allow sensible research. Expert of WP4 have implemented probabilistic machine-learning algorithms that allow to improve the quality and the information content of the PATSTAT database. In the framework of the SIMPATIC project this unique dataset has been delivered under milestone Nr. 30.

The following papers describe the methodology:

1. A flexible, scaleable approach to the international patent “name game” 

by Mark Huberty, Amma Serwaah, and Georg Zachmann

The inventors in PATSTAT are often duplicates: the same person or company may be split into multiple entries in PATSTAT, each associated to different patents. In this paper, we address this problem with an algorithm that efficiently de-duplicates the data. It needs minimal manual input and works well even on consumer-grade computers. Comparisons between entries are not limited to their names, and thus this algorithm is an improvement over earlier ones that required extensive manual work or overly cautious clean-up of the names.

READ Working Paper

Source Code

Download Data

 

2. A scaleable approach to emissions-innovation record linkage

by Mark Huberty, Amma Serwaah, and Georg Zachmann

PATSTAT has patent applications as its focus. This means it lacks information on the applicants and/or the inventors. In order to have more information on the applicants, we link PATSTAT to the CITL database. This way the patenting behavior can be linked to climate policy. Because of the structure of the data, we can adapt the de-duplication algorithm to use it as a matching tool, retaining all of its advantages.

READ Working Paper

Source Code

Download Data

 

3. Remerge: regression-based record linkage with an application to PATSTAT

by Michele Peruzzi, Georg Zachmann, Reinhilde Veugelers

We further extend the information content in PATSTAT by linking it to Amadeus, a large database of companies that includes financial information. Patent microdata is now linked to financial performance data. This algorithm compares records in PATSTAT with records in Amadeus using multiple variables, and learns their relative weights by asking the user to find the correct links in a small subset of the data. Since it is not limited to comparisons among names, it is an improvement over earlier efforts and is not overly dependent on the name-cleaning procedure in use. It is also relatively easy to adapt the algorithm to other databases, since it uses the familiar concept of regression analysis.

READ Working Paper

Source Code

Download Data

When & how to support renewables? Letting the data speak

This report is co-authored by Amma Serwaah, Research Fellow at WBZ, and Michele Perruzi, Research Assistant at Bruegel.

Low-carbon energy technologies are pivotal for decarbonising our economies up to 2050 while ensuring secure and affordable energy. Consequently, innovation that reduces the cost of low-carbon energy would play an important role in reducing transition costs. We assess the two most prominent innovation policy instruments (i) public research, development and demonstration (RD&D) subsidies and (ii) public deployment policies.

Our results indicate that both deployment and RD&D coincide with increasing knowledge generation and the improved competitiveness of renewable energy technologies. We find that both support schemes together have a greater effect that they would individually, that RD&D support is unsurprisingly more effective in driving patents and that timing matters. Current wind deployment based on past wind RD&D spending coincides best with wind patenting. If we look into competitiveness we find a similar picture, with the greatest effect coming from deployment.

Finally, we find significant cross-border effects, especially for wind deployment.
Increased deployment in one country coincides with increased patenting in nearby countries.

Based on our findings we argue that both deployment and RD&D support are needed to create innovation in renewable energy technologies. However, we worry that current support is unbalanced. Public spending on deployment has been two orders of magnitude larger (in 2010 about €48 billion in the five largest EU countries in 2010) than spending on RD&D support (about €315 million).
Consequently, basing the policy mix more on empirical evidence could increase the efficiency of innovation policy targeted towards renewable energy technologies.