In March 2007, the European Commission presented the “Renewable Energy Roadmap”, a document establishing a binding target for all European Union’s Member States to increase their renewable energy consumption rates from sources such as wind, solar, hydro or biomass. The overall value (for all EU Members) was set at 20 percent and should be reached by the year 2020. The renewable energy production was introduced in the energy production sector just after the energy crises of the 1970s as private attempts of few countries to deal with prospective energy security issues. In addition, renewable energy technologies contributed in pollution abatement, which has been a challenging issue for the European Commission ever since, due to their non-depletable nature and the decreased amount of greenhouse gas emissions compared to the ones produced by fossil fuels. In December 2008 the European Union adopted the proposal of the Commission (2008), which stated each Member country’s target share of renewable energy consumption, calculated on the basis of per capita gross domestic product. The Directive included in-between targets (such as 25 percent of the overall target between 2011 and 2012) but no binding constraints on their implementation. Besides the increase of the Renewables share, another solution which has been proposed by the European Commission has been the concept of “energy efficiency”. This idea is defined as decreasing the energy consumption by 20% by the year 2020. In contrast to the renewable concept, energy efficiency has been implemented following regulations suggested by the European Commission. The general policy instruments which have been used remain the same with those of Renewables but they have been more clearly defined for all sectors (households, industries, transports and services). Despite the fact that the Renewables have proceeded as a concept, energy efficiency has been adopted widely and faster. These two trends have been the most representative concerning the energy changes which the European Union started promoting the last decades. Their effect is multi-oriented: prevention from energy depletion and dependence on foreign countries and pollution abatement, according to the directions of the Kyoto protocol. The success of the implementation of environmental friendly technologies has been related to the technological change in the energy field. Many researchers tried to connect environmental innovation with various characteristics, such as R&D expenditures or policy instruments, in order to suggest effective means for their promotion. Characteristic researches have been these of Popp (2005) with “lessons” over environmental innovation and connection of innovation to energy prices, Brunnermeier and Cohen (2003) who modeled, for the US industrial sector, the determinants of environmental innovation, Vries and Withagen (2005) who connected environmental innovation to stringency policies and Johnstone et al. (2008) who modelled the environmental innovation determinants and specified their research in policy instruments. Johnstone et al. (2008) tried to define a relation between environmental innovation and energy policies for the Renewables using an empirical model. The model included some general variables, such as electricity prices or energy consumption, and then defined binary variables to operationalise the policy measures. The environmental innovation was defined as the number of patent counts for specific classes of the International Patent Classification. The present study’s model is based on Johnstone’s et al (2008) model. It follows the main lines of their research by using the same innovation output, i.e. the patent counts, and by having two sets of variables: the general ones and the policy-related ones. The patent counts were considered to be the most appropriate output indicator for this empirical study, as well, as they provided the necessary information to modelling (such as priority date and country of application) and were widely available (from the European Patent Office - EPO). Johnstone et al. (2008) defined a simple “input-output” model for six renewable energy sources. For each one of the sources, they used a specific classification in order to gather all patent counts relevant to each source. They also included seven policy types which were implemented for the renewable energies for a specific time period. Finally, they included a set of “explanatory” variables, which were considered as input innovation indicators. This study based on their model also used patent counts as innovation output. The main “modelling” differences are the introduction of more explanatory variables, such as R&D personnel, international trade indicator or R&D intensity, the different policy instrument operationalisation (from binary to semi-quantitative) and the empirical analysis methods. Also, Johnstone’s model is implemented for a different set of countries and a different time period than the one we examine. Our focus is on the effects of policies on environmental innovation for the EU15. The reason for that is that it is interesting to measure the effects of policies within the borders of the European Union (for policy recommendations) and that the data collected are from the EPO. Johnstone et al (2008) mention that, when coming from non-European country it is less likely to patent at the EPO. Also, our information is set for a shorter period of time for practical reasons (there was no available information before 1990 for the renewable sector and, by definition the European Commission has not started yet promoting the Renewables). The new framework has been yet once modified in order to include the effects on energy efficiency energy trend. Using the same variables, we created a new output set in order to incorporate the policy measures implemented for this second energy trend. Using a logarithmic scale, we performed two types of regression analysis to model our data, the linear regression using the ordinal least squares, which provides a linear relationship between the dependent variable (our output) and the independent variables (our input) and gives the best-fitting relation by minimising the least squares error and the general linear regression, which takes into account possible attributes of the dataset, such as heteroscedasticity or correlation, and, is also more efficient in analysing datasets with cross sectional data (such as time series) included in our model. The model used for both energy trends several explanatory variables, determinants of patenting activities. From these we extracted valuable information concerning the R&D input, the energy consumption and the electricity prices. R&D was measured in two separate ways: the R&D intensity, which reflects the R&D expenditures ratio, spent on each energy trend and R&D personnel. Both of these variables showed a positive linkage to the output. Positive impact also came from energy consumption. As for the electricity ratio, a subsidy for the electricity prices for households and industries, the results diverged, depending on the energy sector. For the renewable energy, the electricity ratio had a negative impact denoting the importance of the industry electricity price as innovation determinant and for energy efficiency; the electricity ratio had a positive impact denoting the importance of household prices as innovation determinants. At the same time, we should mention the fact that policy instruments follow this observation: for energy efficiency, most measures are implemented in the household sector, while on renewable energy most measures affect the industry sector. The regression analyses showed interesting results concerning the policy instruments for both energy trends. While the renewables favour price based instruments, such as taxes and tariffs, the energy efficiency favours legislative (like quotas) and financial (like grants and subsidies). This observation is very important as it shows the tendencies of the last 17 years to promote different energy trends. For renewables, which were boosted right after the energy crises, the measures have a more obligatory character in order to succeed. This information combined with the fact that renewables were mostly promoted in the industry sector (which by definition consumes the largest amount of energy) links renewable innovation to stringency policies. On the contrary, energy efficiency has a more liberate character. This is also connected to the fact that it is mostly implemented in households. After the conclusions, the policy measures provide a clearer picture of how environmental innovation is linked to them. In conclusion, the environmental innovation is strongly related to existing policy measures, and, in fact, different policy instruments have greater effect on different environmental friendly energy trends than others. Price based measures are most effective in inducing innovation for renewable energy technologies, while legislative and financial measures are most effective on energy saving innovation. During the study, there were found no significant results on voluntary measures. In general, public policies are linked directly to these two energy trends. This conclusion could urge the promotion of the two trends in parallel so that the targets (pollution abatement and security of supply) of the European Commission are achieved as scheduled.