Data was collected from two independent data sources. Investments (additions in RE capacity) have been drawn from BNEF, which possesses one of the most comprehensive databases in the field of clean technology financing (Cardenas- Rodriguez et al., 2013). It contains information on installed electricity generating capacity, date, transaction type, financing type, amount of equity, and amount of debt. We used data from 2003 to 2011. The complete database includes 5840 Solar
Investments, 9643 Wind Investments and 2889 Biomass & Waste Investments. These three RE subsectors account for 72% of the RE funds invested and with 74 GW installed for 75% of all capacity additions in the period, and have therefore been selected for further analysis. The records of the BNEF database start with the year 2003 and we further decided to use the data up to the year 2011, as data quality of the following year did not meet the standards of previous years at the time of research in early 2013.
Policy indicators were drawn from the IEA/IRENA Policy and Measures (PM) database which includes policy measures in OECD countries from 1974 onwards. These indicators have been used by scholars to analyse the impacts of aggregated policy instruments in Europe (Cardenas-Rodriguez et al., 2013; Marques & Fuinhas, 2012a, 2012b) and globally (Aguirre & Ibikunle, 2014). We used the IEA/IRENA policy measures data from 2000-2011 which corresponds to the availability of investment data. The database includes a total of 957 distinct policy measures active in the respective years resulting in 7835 policy data points (see appendix Table 18).
We structured the data according to sectors (Multiple RE sources, Wind, Solar, and Biomass) and applied additional data processing: First, we removed cases with missing values and included only completed deals. Second, our interest in the influence of policy measures lead to the exclusion of countries with less than three consecutive years of investor activities (DV). The case-selection is carried out for each sector analysed (see section 4.3.3. for the procedure and Table 17 (in the appendix) for a list of countries in the subsamples). The selection of policy instruments included in the model is described in section 220.127.116.11.
As control variables we further included macroeconomic and energy market data from the US Energy Administration - EIA (carbon intensity of the economy), OECD (Longterm interest rates, share prices) and from the World Bank (Total energy consumption, gross domestic product (GDP)) in our analysis.