 So, today I'm going to speak a bit broadly about the determinants of energy innovation, specifically the political economic factors. This paper has been co-authored with two of my senior researchers at FIM. Let's dig into this. Okay, in the IPCC AR5 report, there is a throwaway line. Like, someone just threw it out there, which basically says, it will take unprecedented levels of improvement in institutional quality to limit the temperature rise below two degrees Celsius. Now, you might read this and move along, but when I read this, I thought, what are our institutions right now? In many of the countries that will have to achieve very high reductions in emissions, institutional quality is rather lacking, to be politically correct. As a result, we felt it was extremely critical to examine the determinants of energy innovation. Specifically, we look into environmental and R&D policies, governance quality, institutional quality, and lobbying to understand how this impacts energy innovation. We use two indicators of innovation, namely the industrial energy R&D, which we consider energy inputs, innovation inputs, and energy patents, which are considered as the outputs from innovation. So, there are various gaps in the literature. Without going into this, we are trying to look into all these factors simultaneously and study their impacts. Okay, measurements. So, measuring innovation has been difficult, and it remains difficult because of a lack of clear definition. What we decided to do was look at the amber dataset by OECD. It's a continuous dataset. It has clear definitions, and at the same time, it has clear sectoral boundaries so that there are not a lot of overlapping in the innovation R&D. So, what we look into is, again, one is power R&D, which is downstream R&D in electricity, water, and the gas distribution industry. Secondly, for R&D expenditures, we look into the electricity, water, and gas distribution along with mining. So, this is upstream plus downstream for the energy supply sector. And mathematically, this is arguably a lower bound because we do not consider the embedded capital R&D into the supply sector. Moving on to patents, what we do is we look into three types, basically. So, the energy power patents, which we define as those related to energy generation, green patents, that is power patents, plus patents related to environmental management, climate change mitigation, energy efficiency in buildings and lighting, emissions mitigation as an abatement, and, of course, full efficiency in transportation. Finally, we also sum this up, and we classify them as environmental patents. Again, because this is a rather econometric study, I need to clarify that we scale all the innovation proxies relative to the total value added to account for the heterogeneity among the various countries. So, I'm coming to the number of countries we look into and the data stuff. So, the sources of data, energy innovation, and birds OECD dataset, it's publicly available, it's very good, it's got time series, it's across countries. I encourage anyone who's interested in R&D innovation to look into this dataset. There are a lot of good insights to be gained out of this. Patents, we look into the OECD's patent statistics and patent cooperation treaty, two datasets, they have similar definitions and their data are well matched. Okay, so for the data, we looked into 20 OECD countries. Now, we've spoken a lot about the institutional quality in developing countries, but we decided to look into OECD because of data issues. So, the datasets we used only provided data on OECD countries. But again, this analysis can be expanded to the different developing and non-OECD countries. Essentially, we study four hypothesis, number one, and each of them have to deal with the factors that we look into. Number one, what we do is we hypothesize that environmental policy stringency results in dynamic efficiency gains. And at the same time, stringent regulations provide incentives for both energy saving and pollution reducing technologies. In institutional quality, which we measure in terms of good governance from the world governance indicators by the World Bank, also increases incentives to invest in energy related innovations. That's the second hypothesis. The third one, political orientation of the government influences investments in energy on innovation, but we do not make a priori assumption on this. The impact can be both ways. At the same time, high share of energy intensive sectors, which induces market size effects and can be classified as lobbying, also has impacts or can act as a determinant of energy innovation. The next slide, this could be very boring. This is an econometric study. If you have specific questions, find me outside and ask questions, or you can also ask questions later. I'm not going to spend too much time on this, but what we do is we run a standard fixed effects model with both location and year fixed effects to include or to control for unobserved heterogeneities. Along with the four factors that I spoke about in details, we also control for other factors such as energy prices and trade openness of various countries. This environmental policy stringency is a central theme of our research. As a result, let's look into this very carefully. This indicator has been developed by OECD to researchers Bota and Kozlux in their 2014 paper. If you look at the top, that's a composite index. The composite index is broken down into two subcomponents. First, the market-based policies, which includes taxes on carbon dioxide, nitrogen dioxide, and sulfur dioxide. But at the same time, they also include trading schemes on carbon dioxide, renewable energy certificates, energy efficiency certificates. The third subcomponent of the market-based policies are feeding tax on both solar and wind. The second subcomponent of the EPS or the environmental policy stringency has to deal with the non-market policies, which includes both standards, that is emission limit values, and R&D subsidies provided by the government. Now, each of the EPS and the two subcomponents are scored on a scale of zero to six, depending on the policy stringency. The higher the score, the more stringent the environmental policies are of a country. Finally, the scores are then weighted to aggregate the EPS total. This has been widely used and is very reliable. This is a chart that we try to come up with. As the indicator bar shows at the below, the redder you are, the more stringent your environmental policies are. Since we looked at mostly OECD countries, this map is that of Europe. This index shows that of the entire world or all the countries that we took into. Here is Australia, which has very lax environmental policy stringency. But if you look at some of the European countries, they have pretty high stringency of environmental policy, which is good. I was in Ireland recently and the finance minister came up and gave a talk about how stringent environmental policies are. But if you look at Ireland, it has actually very good, very lax environmental stringency in policies. He was not very happy that I pointed that out to him. Moving on, again, coming back to the political economic factors. The foreign institution on political economic factors that we look into are the stringency of government support to energy innovation, which we have just discussed, the EPS indicators, the quality of governance. These are measured in terms of government effectiveness, the rule of law, and control of corruption from the world governance indicators of the World Bank. These are standardized scores between negative 2.5 and positive 2.5. So the higher you are, the better the quality of your governance. The third factor we look into is the political orientation of the government. And this also comes from a dataset which basically classifies governance policies into left-leaning versus right-leaning. And finally, the distribution of resources across interest groups, which we call lobbying, the subcomponents are market size effects and the power of the energy lobbying. And finally, also the share of the energy intensive industries. I'll come into this. So this is a table, which basically I put the hypothesis that we studied and the proxy variables that we're going to measure. So for example, for environmental policy, we look into the three components of EPS, the EPS market, the EPS non-market, and of course the EPS total and so on. Not sure if this table, so this table basically shows the mean and the standard deviation along with the mean and the minimum and the maximum of the indicators that we used. It's probably not of high interest to you right now, but the paper is available online. If you have specific questions, please come and find me. Okay, excellent. So we're down to results. Back to the first hypothesis, which is the role of environmental policy stringency. In the case of, remember, we're going to go back a little bit. So we studied the impact or the determinants of, sorry, the determinants of both R&D expenditure and R&D related patents. So for R&D expenditure, we find that environmental policy stringency has a positive and significant effect only on electricity R&D. So not on the power R&D. So the impact is significant or the impact of environmental policy stringency is only significant and positive on a sub-component of R&D expenditure, not on the total R&D expenditure on the power sector. For patents, however, the inducement effect of market-based instruments, so tax and feeding tariffs is larger for environmental patents. So in this, for the case of patents, the impact is larger for the wider classification or the definition of patents. Now, putting this into perspective, one unit increase in environmental policy stringency, which is equivalent to one interquartile range change. So moving from the 25th quartile to the 75th quartile results in, for the market-based instruments, so for tax and feeding tariffs, it increases power patents intensity by around 1.5 percent. But for environmental patents intensity, it increases by more than 3 percentage points. For the non-markets, so that is the subsidies, the impact of environmental policy stringency is that it increases power patents intensity by between 1.2 and 1.5 percentage points, but environmental patents intensity by over 2 percentage points. Now, putting this into perspective, it might, the changes might seem small, but in case of the market-based score, moving from 25th to 75th percentile is equivalent to moving from the stringency, the environmental policy stringency of Belgium to that of Finland in 2010. So Belgium in 2010 had a rather lax or weak stringency in environmental policy, whereas Finland was already red, so it had very high policy stringency back then. In the case of non-market scores, the same movement, so moving from the first quartile to the third quartile, would be equivalent to moving from the policy stringency of Portugal in 1995 to that of Portugal in 2010. So Portugal has made a high jump. In 1995, it had a very weak level of policy stringency. By 2010, it had already almost caught up with Finland. Moving on, good governance. So this slide looks into the results of both good governance and political orientation. The takeover message is that stronger economic institutions promote innovation. Coming into the details, we know that this is a critical driver, good governance. One unit increase in governance indicator has a positive impact on both power R&D and patent intensity. So for power R&D, the impact is between 62 and 96.4 percent. So these are huge increases. For patent intensity, it is between 6.5 and 30 percent. But again, putting this into perspective, a one unit change in good governance is equivalent to change from Portugal, which was 1.02, to that of Sweden in 2010. So it's a big special move as well. For political orientation, we only find that it only positively affects power and energy R&D intensities, not the patents insensities so much. Changing from right cleaning to left cleaning governance increases power intensity, R&D intensity by 11 percent and energy intensity by 22 percent. So again, I like the example of Portugal. Portugal moved from change to a left cleaning government in 1995, whereas Canada and Sweden went the opposite way. The final set of results, this is for lobbying. So we also find that research distribution, market side effect and lobbying has positive impact on R&D intensity. This is mostly because a larger energy sector can lobby for more R&D allocation. A 1 percent increase in value added share of energy intensive industries increases power R&D by more than 0.5 percent. Lobbying seems to have a greater impact on the inward oriented sector, such as power. We looked into the energy sector as well, but energy sector also has R&D innovation or R&D expenditure from oil and gas industries, which tend to be more international in its nature, and its lobbying has less effect on that. Energy price has a negative effect on power and energy R&D. This is most likely because high energy prices increases energy expenditures in both the private and public sectors, reducing resources that are available for other users, including R&D. The final finding can be complexing for some of us. It shows that trade openness reduces incentives for R&D innovation. And now the explanation could be that once countries learn or once the trade development of countries improves, when their relationships improve, they're able to purchase R&D innovation instead of investing money in it. These are our progression tables if anyone is interested later on. So, in conclusion, what we find is that both market and non-market-based incentives result in dynamic efficiency gains. We show that better governance promotes energy innovation, and that left-wing governments are more likely to devote R&D resources to the energy sector. But this does not translate to the innovation output, that is the patents. Finally, a large distribution of resources toward energy-intensive sectors can induce market-sized effects. Some implications. So, political economic factors can act as barriers even in the presence of stringent environmental policy. So even if you have stringent policies to promote innovations and provide long-term incentives, if your quality of governance and the institutional quality in your country are not strong enough to enforce those policies, then there are no use. So, in order to move towards a greener economy, in order to provide long-term sustainable incentives, countries, and these are intuitive in a lot of times, they have to improve the institutional qualities. They also need to consider the influence of governance political orientation, because we see that sometimes left leaning governance are very good at providing incentives to the labor-intensive inward-looking sectors, but when it comes to sectors that are more international in stature, they're lacking. Lobbying also has significant impacts, and countries need to be careful of that as well. So, the final message here should be that focus on determinants of energy innovation, but we need to go beyond that of the environmental policy instruments. The focus really needs to be on the political economic factors as well. Otherwise, we'll end up with a crisis. Again, thank you very much.