The Role of Export Credit Agencies in Project Financed
Offshore Wind Farms
University of Amsterdam, Amsterdam Business School Master in International Finance
Master Thesis Student: Maurits Bos Student number 10884041 Thesis supervisor: Dr. S.R. Arping October 2017
Abstract
This thesis is intended to provide an exploratory overview of the role of Export Credit Agencies in project financed offshore wind farms in Northern Europe, a market said to be in an incipient stage. The hypothesis that is tested and confirmed in this thesis, is that in more recent years and especially after the financial crisis, financial commitments of Export Credit Agencies are a significant important factor in the debt tranches provide to project financed offshore wind farms. Project finance as a mechanism to finance projects also significantly increased in the more recent years. It is also found that
government support by means of multilaterals institutions is key in the overall debt structure of these investment projects. Furthermore it is shown that the characteristics of project finance (significant leverage ratios and debt on a pure non-recourse basis) as can be found in the literature, are applicable to project financed offshore wind farms. I find that the leverage ratios are conform earlier research on project finance in general but the full non-recourse basis of debt is limited due to guaranteed debt tranches, not confirming findings in earlier research.
This theoretical framework is tested by constructing a database of 86 investments in offshore wind farms worth USD 107.7 billion in 8 countries over 17 years. A Probit regression is performed to test for significant difference in probabilities between corporate financed investment projects and project financed investment projects. A sub sample of 27 projects is used to test for the significant
Table of Contents
Abstract 2
1 Introduction 4
Research question 5
Relevance and aim 5
Methodology and research design 5
Structure of the report 6
2 Literature review 7
Renewable Energy investments in offshore wind farms 7
Project Finance 9
Export Credit Agencies 10
Summary 12
3 Hypothesis and research methods 14
Hypothesis 14
Research methods 15
4 Data and descriptive statistics 17
The dataset 17 Descriptive statistics 20 5 Results 22 6 Discussion 26 References 29 Appendices 32
Appendix 1 – Example of data imperfections 33
1 Introduction
The idea of this research topic originates from examples I encounter during my daily work. I see participation of government-backed Export Credit Agencies (ECAs) in the field of offshore wind financed through project finance (PF) initiatives. Arguments for their presence are the immature and non-proven technology on the long run. Participation of ECAs in mature technologies and western world markets might indicate a form competition with commercial parties and could distort normal market functioning. This thesis provides an overview of the participation of ECAs in the PF of offshore wind farms. Furthermore a distinction is made whether the situation has changed after the financial crisis.
This thesis combines two long standing practices in finance. There are examples of project finance structures that date back until the 17th century. Investors finance trading expeditions on a voyage-by-voyage basis. Upon return, the cargo and ships would be liquidated and the proceeds of the voyage-by-voyage split amongst investors (Bodnar & Comer, 1996). In 2015 the total bank debt provided to PF initiatives was $278 billion (Project Finance International, 2016). Export credit agencies stimulate the export of companies since almost 100 years. In 1934 the Berne Union was established. Their mission is to support and grow trade and investment globally (Berne Union, 2017). To date the Berne Union has 84 members and as of 2016, ECAs underwrite or finance more than $400 billion of international exporting and business transactions.
Global developments and incentives such as the Paris Agreement striving to lower carbon dioxide emissions, are the driving force for developments in renewable energy initiatives. One of such initiatives is wind energy. Specifically for this thesis we explore energy from offshore wind farms. An asset class that initiates from 1991 when the world’s first offshore wind farm (OWF) “Vindeby” was built in Denmark and is becoming more and more mature market since 2003 when the first OWF exceeding 100MW, “Nysted” became in operation. These capital intensive projects with limited long term proven technology are financed through balance sheet, corporate financing, or by project
financing. The lifespan of OWF can be broadly separated into two phases, the construction phase and the operational lifetime. The balance sheet financing is usually done by large utilities. They absorb the construction risk and when the operational stage has been achieved the utility sells part of the de-risked asset. There is very little information about the financials of such projects before the public asset sale. It is tested if there is a significant difference in using PF over CF. Project financed offshore wind farms provide more insight in the financial stakeholders involved. Besides commercial banks other financial institutions such export credit agencies, multilateral organisations and development banks are involved and provide their expertise in order to make the projects “bankable”.
Research question
The main purpose of this thesis is answering the question: What is the role of government-backed Export Credit Agencies in the financing of offshore wind farms? This is done by looking at the various possibilities of how the projects are financed, what the main considerations are and if there is a difference observable before and after the financial crisis. The first research question focusses on the difference between project financed and corporate financed offshore wind farms. What are the main determinants for project financed offshore wind farms? The purpose is to look for an explanation on why developers choose project finance as form of financing. The second research question zooms in on project financed offshore wind projects. What are key variables in explaining the involvement of Export Credit Agencies? Answering this question results in a clear statement of when and why Export Credit Agencies are involved in the financing of offshore wind farms.
Relevance and aim
The aim of this research is to provide an exploratory overview on the financing of offshore wind farms in northern Europe and the role governments play by providing support via export credit agencies. Since the asset class is in a fairly junior stage of development, but is maturing rapidly, providing a first overview of the financial developments in this area gives some of the first insight in this area that can be used for further study. This combined with a relatively uninvestigated topic of the role export credit agencies have in the financing of major construction projects sheds the light on a very mature but somewhat invisible approach in the field of structured finance.
Methodology and research design
This thesis is executed using an exploratory and empirical approach to the research questions. By first providing insight in the relevant theoretical framework, subsequently the constructed dataset is used in order to perform several tests in order to observe significant effects.
The first research questions is answered by constructing a linear probability model (Probit regression). The Probit regression is required when the dependent variable is binary and it explains the probability of the dependent variable with an increase in the independent variables.
The second research questions is answered by executing an ordinary least squares regression. The OLS regression coefficients measure a partial correlation between the independent variables and the dependent variable The dependent variable here is a continuous variable. It will be tested if the expected explanatory variables provide sufficient and significant impact on the dependent variable.
All data that is used is manually compiled by using various data sources. There is no single data source that combines information on both project-, debt tranche and export credit agency specific information. The number of total investment projects in this specific field is limited due to the fact that this is a relatively new, but rapidly growing assets class. The sample used for this research is equal to the total population within that period of time.
Structure of the report
In chapter 2 of this thesis a literature review is presented to provide a framework for the reader sufficiently to understand the relevance and positioning of the hypothesis. The literature will cover the reasoning behind the creation and development of the asset class offshore wind. This is followed by a description of the main literature related to project finance in paragraph 2.2. Paragraph 2.3 provides insight in the role of export credit agencies and the last paragraph of chapter 2 summarizes the findings in the literature. Chapter 3 provides the hypothesis and research methods in more detail. In chapter 4 a detailed description of the creation of the dataset and the general descriptive statistics are discussed. Chapter 5 discusses the results on the hypothesis tests and the and in chapter 6 the conclusions are discussed that derive from the outcome of the data that are confronted with the findings in the literature by means of a discussion.
2 Literature review
This chapter highlights the main theory, practice and developments related to the main parts of the study, each part is divided into a separate paragraph. Paragraph 1 starts with an overview on the developments in the global energy transition and in particular mentions the activities related to offshore wind investments. Paragraph 2 discusses the main literature on project finance and paragraph 3 gives insight in the role and responsibilities of export credit agencies. Paragraph 4 concludes with a summary.
Renewable Energy investments in offshore wind farms
Worldwide actions on climate change initiatives have gotten attention from the mass after the introduction of the Kyoto Protocol, adopted in 1997 and entered into force in 2005, for reducing greenhouse gases by setting internationally binding emission reduction targets. The details for later additional initiatives such as the Paris Agreement followed. Subsequent to the Paris Agreement, the European Union has formalized climate change by the State of the Energy Union. It highlights Europe's contribution to the Paris negotiations (European Commission, 2015). Amongst various renewable energy sources, such as solar and biomass energy, wind energy is one of the main contributors to achieve the goals set. Especially offshore wind, because in many countries there is either no space for onshore energy, by also the Not In My BackYard (NIMBY) effect plays an important role. Esteban et al. (2011) find that offshore wind power can be considered an incipient market. However, just at this moment, the growth of this technology finally seems to be happening, being several countries at the top of its development namely the United Kingdom, Denmark, Holland, Sweden and Germany. The Economist (2014) gives two explanations for Britain’s enthusiasm for offshore wind turbines. It is committed by European law to generate about 30% of its electricity from renewable sources by 2020. Secondly Britain sees a change of settings up a lucrative futuretechnology market for export. With a lot of shallow sea, Britain has a great opportunity to experiment. Esteban et al. (2011) explain in their paper what the main reasons are for using offshore wind and its future. They conclude that the technology used for offshore wind is still in its learning curve and is yet not economically feasible and thus needs government support, but is promising due to the enormous resources available and its cost efficiency compared to other renewable technologies. Vindeby Offshore Wind Farm in Denmark was the first offshore wind farm in the world, made in 1991. The technology was unproven and was mainly amended from onshore wind farms. Major adjustments had to me made in order to make it sustainable for offshore conditions.
The construction of offshore wind farms is so expensive to build that it requires a significant size for sustainable economics of the projects. Offshore wind is still dependent on government support and receives this in many different ways such as subsidies via feed in tariff and contracts for difference
(CfD), green certificates (Green bonds) and direct loans/insurance. Some forms are more direct and obvious and noticeable than others.
Parties involved
Main parties involved in the development and construction of offshore wind farms are utilities, and developers such as construction companies and wind turbine manufacturers. This last group is referred to as Independ Power Producers (IPP). For utilities offshore wind farms are just another power plant. Utilities have the in-house management capacity to deal with the complexity of such projects and to manage the cheaper multi-contracting route. It also means that they want to keep control of the project, and avoid unnecessary interference from outsiders, especially bankers and their multiple advisors (Guillet, 2010). The recent developments on the debt markets have also not let to a forced push for other forms of financing than regular corporate debt finance.
For developers and IPPs, project finance is necessary, and its requirements cannot be avoided. The project structure and the contracts , should be ‘bankable’ on a non-recourse basis (Guillet, 2010). He further finds that banks do not care about a sharp pricing and upside potential, they are very reluctant for any downside risk and therefore study these investment projects upfront very carefully by all kinds of scenario analyses. However it can be observed in the market that banks are willing to finance construction risk on project finance. They however do not allow this for projects for utilities, because they do not allow them to solely depend on their own due diligence processes. Utilities will need to reserve significant amounts of contingency budgets in order to have the construction risk financed by banks. A very inefficient and costly process for these parties. Pricing for pre-construction and post-completion are not very different from another. The main costs for debt arise from other factors such like the parties involved, quality of the team working on the project and the sponsors.
Utilities tend to take a different strategy. They develop the project first, consequently they divest part of the investment to mainly long term investors such as pension funds. This process that is called de-risking. Part of the project is refinanced through for example project finance. Main goal for utilities is to remain the major shareholder with an ownership of at least 51%. Project finance for a (partial)
refinancing is not considered in this research.
In general construction risk is the most critical risk on the offshore wind farm. Farms are built at places where wind conditions are perfect for generating energy. For construction works, this means the most challenging locations to work. Offshore wind operates at the intersection of industries, marine
construction and turbine manufacturing. Both industries are more or less equally present, so none of these can take the lead (Guillet, 2010).
Project Finance
Within the field of structured finance, project finance (PF) is an odd case. When using PF, both the capital investment decision as well as the financing structure are depended on one another. This embodies an exceptional condition highlighted as part of the “irrelevance” proposition of Modigliani and Miller in where they state that the financing structure does not affect firm value, Modigliani and Miller (1958) and Esty (2003a). PF does require the creation of a separate legal project company, a Special Purpose Vehicle (SPV) or Special Purpose Entity (SPE). The structure of the SPV allocates the risk to those participants who are best able to handle it via legally binding contracts Girardone & Snaith, 2011). PF is sometimes referred to as contract finance (Esty, 2004) this is a major advantage of PF because it create the possibility of risk allocation towards the most equipped participant. This related to most of the encountered risks, but political risk remains undiversifiable between the sponsors, lenders and contractors. The coverage of political risk is where Export Credit Agencies (ECAs) find their role in the project structure.
The leverage ratios for project finance initiatives is high, 70% on average (Esty 2004) Vaaler et al. (2008) find that debt levels average 75.02%, and this debt is usually financed by the mechanism of limited or non-recourse bank loans. PF debt levels are thus much higher than corporate debt levels. The nature of PF debt is syndicated lending. Finnerty (2007) notes that lending syndicates, which include commercial banks, export credit agencies, insurance companies, pension funds, equipment vendors and multilateral development institutions, provide about 90% of PF debt.
Banks can only securitize their debt service (repayment and interest) by the cash flow generated from the SPV-asset. Equity providers (sponsors) use project finance as a mechanism to finance large capital intensive and possible high risk investment projects because of various key characteristics limiting risk, exposure to a certain asset class or project and creating an option to walk away against an upfront known costs (reputation risks not taken into account). According to a report from the World Bank (2016) one of the primary advantages of project financing is that it provides for off-balance-sheet financing of the project, which will not affect the credit of the shareholders or the government
contracting authority, and shifts some of the project risk to the lenders in exchange for which the lenders obtain a higher margin than for normal corporate lending (World Bank, 2016). Project finance creates value by reducing the agency costs associated with large, transaction-specific assets, and by reducing the opportunity cost of underinvestment due to leverage and incremental distress costs (Esty 2003a). Some of the downsides of project finance are the time it takes before financial close can be reached. Due to the number of bilateral contracts and extensive due diligence this may take up to at least one year. Another disadvantage is the margin which is, in normal cases, higher compared to standard corporate finance.
Large international banks fund the majority of the loan, often with a smaller contribution from
multilateral development banks such as the International Finance Corporation (IFC). The participation of such institutions is one method by which political risk is addressed in project finance (Gianturco, 2001; Girardone & Snaith, 2011).
Project financed companies are interesting to study because as standalone greenfield asset, project companies do not have access to internally-generated cash flow. They must raise all of the capital from external sources. Second, they are created to finance illiquid assets with long, but usually limited, lives. Third, for agency reasons related to the use of free cash flows project companies have very high leverage ratios and capital structures comprised almost entirely of amortizing bank debt (Esty, 2003b).
Hainz and Kleimeier (2007) find that the use of PF increases with both political risk of the country in which the project is located and the influence of the lender over this political risk exposure. Girardone and Snaith (2011) complement this finding with the result that the relationship between disaggregated measures of political risk and project finance loan spreads differs as a function of country
development. The cost of funds is negatively related to the effectiveness, quality and strength of a country’s legal and institutional systems, lower levels of government stability and democratic accountability are more likely to be associated with lower loan spreads, with little evidence found to support a role for political risk in developed countries.
The global financial crisis (GFC) started in August 2007 and I will be using the end of 2008 as the end of the GFC. I use the same time horizon as Ivashina and Scharfstein (2009) use in their paper. They find that during the timeframe of the financial crisis new loans to large borrowers fall significantly. Especially in the fourth quarter of 2008 where the loans fell by 79% relative to the peak of the credit boom in Q2 2007. Guillet (2010) describes that during the financial crisis of 2007 and 2008 there was a time where there was no syndication market. Banks were not committed to take large stakes per transaction and only on a take and hold basis. For the large scale offshore wind projects this would result in a group deal of a significant amount of banks or bringing in multilaterals with their specific requirements and constraints. Post-crisis banks are more conservative and risk-averse. If you would incorporate this debt structure in an investment project it would result in an uncompetitive financing structure. According to the European Wind Energy Association (2016) given the scale of the offshore wind sector in 2015, project finance remained an important tool in the market.
Export Credit Agencies
The “unsung giants” of international finance Gianturco (2001) are the world’s export credit agencies (ECAs)-highly specialized financial institutions that cover about $600 billion of exports each year. Export credit plays a role of central importance in international trade (World Bank, 2006). The basic role of an Export Credit Agency (ECA) is to support and encourage exports and outward investment
by providing financial support, through insuring or guaranteeing international trade and investment transactions and, in some cases, providing loans or finance directly (Morrison, 2012). I limit my research with governmental supported ECAs, commercial credit export agencies are left out of the thesis. Examples of ECAs are Coface (France), Export-Import Bank of the United Stated (EX-IM Bank), Atradius (The Netherlands), EDC (Canada), EKF (Denmark), GIEK (Norway) ONDD (Belgium) and Euler Hermes (Germany). In Appendix 2 some examples of responses on active projects by the ECAs are provided.
An ECA is usually set up as a statutory institution that aims to facilitate foreign trade primarily between private parties. In reality, a government, through the ECA’s operations, directly or indirectly becomes party to a private trade transaction. A range of countries have established one or more such
specialized financial institutions, each of which may have a distinctive organizational structure and business model (World Bank, 2009).
A definition of ECAs provided by Gianturco (2001) is (1) a highly specialized bank, insurance
company, finance corporation or dependency of the government, (2) offering loans and/or guarantees, insurance, technical assistance etc. to support exporters, (3) covering both commercial and political risk related to export sales (4) with the backing or approval of the national government, and (5) dedicated to support the nation’s export. ECAs can be governmental, mixed or privately owned. Governments provide officially supported export credits through Export Credit Agencies (ECAs) in support of national exporters competing for overseas sales. Such support can take the form either of “official financing support”, such as direct credits to foreign buyers, refinancing or interest-rate support, or of “pure cover support”, such as export credits insurance or guarantee cover for credits provided by private financial institutions. ECAs can be government institutions or private companies operating on behalf of governments.
ECAs are considered the facilitator of last resort in financing and are complimentary to the existing commercial markets. If a commercial party is willing to insure or invest, ECAs are not involved. This implies that from a purely commercial angle there is no appetite to cover the project, liquidity is low or technologies are not mature, these agencies are expected to act as a kick-starter. ECA financing is vital for mobilizing capital and for keeping the cost of capital at a level that produced manageable tariffs for the emerging economies (Gianturco, 2001). There is no such thing as one ECA. Every ECA has its own characteristics based on history and culture.
Hainz and Kleimeier (2006) show that lending syndicates with multilateral development banks and export credit agencies often provide PF debt, and these multilateral development banks act as "political umbrellas" that mitigate the risk of expropriation. Woodhouse (2005) shows that the involvement of multilateral institutions significantly affects the outcome of a renegotiation in favor of the investors.
The Organisation for Economic Co-operation and Development (OECD) has a long tradition of rulemaking in the area of officially supported export credits, dating back to 1963. Export Credit Agencies in OECD countries are constrained to work according to the Export Credit Sector Understandings which prescribes the Arrangement on Guidelines for Officially Supported Export Credits, which stipulated maximum repayment periods, minimum cash payments, and minimum interest rates. Given domestic interest rates, the arrangement introduced ceilings on the subsidies conferred by officially supported export credits. Over time, the governments have reduced the prevailing ceilings by increasing the minimum interest rates (OECD, 2017) Specifically for the
renewables sector, the on Renewable Energy, Climate Change Mitigation and Adaptation, and Water Projects (CCSU) directive is in place.
Both project finance and ECA are very specific and case-by-case situations. Therefore it may be hard to investigate and generalize the outcome of this research to a large extend.
Another important government body that is an enabler, and is required to provide some detail about, is the European Investment Bank. This multilateral institution contributes chiefly to economic and social cohesion, environmental protection, research and innovation, support for SMEs, the development of trans-European transport and energy networks, and sustainable, competitive and safe energy (EIB, 2016). It is intended to stimulate EUR policy and in this respect it has become a priority objective to promote renewable energy initiatives. According to its own website it allocated EUR 2.2 billion on renewable energy projects in 2008 (EIB, 2009).
Summary
Since the Kyoto Protocol in 2005, awareness on climate change significantly increased and has been formalized into concrete actions plans. One of such concrete examples is the State of the Energy Union drafted by the European Union. The European Investment bank contributes to this plan by providing direct loans to renewable energy projects. Esteban et al. (2011) paraphrase Offshore wind as an incipient market, confirming the scarceness of the financial theories and research papers found. This gap brings an interesting opportunity for this thesis to provide an exploratory overview of the market from a financial angle. The construction of offshore wind farms is expensive and to date requires direct subsidised electricity prices from host governments. A less directly observable form of government support, is de participation export credit agencies in offshore wind.
Independent power producers are forced to use project finance. Utilities usually choose corporate finance because of relaxed conditions and long lead times. Project finance requires the creation of a separate legal entity and is usually financed with considerable leverage between 70% and 75%. The debt is provided on a non-recourse bases, limiting banks to only cash flows from the project, which increases their risk and the margin they require from customers (World Bank, 2016). Political risk in
general project finance is allocated to ECAs. It has also been concluded that the global financial crisis was a driver of very limited liquidity during the period of the GFC and has changed to a risk averse attitude after the crisis.
The role of government supported ECAs is to support and encourage exports. Literature focusses on exports to developing countries and covering political risk, developed country involvement is hardly researched. The role of ECAs role is to be complementary to the commercial market, so involvement is only required if the risk will not be taken by other parties. Effectively a government becomes a party to a private trade transaction.
3 Hypothesis and research methods
This chapter describes the hypothesis that are derived from the available literature and the
expectations of the researcher in order to provide an exploratory view on the market. The models that will be implemented and tested is described in paragraph 3.2 research methods.
Hypothesis
Based on the literature review, we can hypothesise and statistically test the total investment market and expectations on participation of ECAs in particular.
The first hypothesis test that is conducted is the expectation that the total project value, the capacity of the wind farm, the country and the year of financial close are main determinants for a project being financed through project finance. This is important because the form of financing is the first
explanation why a certain financing form is chosen. Accordingly if this would result in PF, the second hypothesis can be tested to further drill down into the considerations of using project finance combined with ECAs. The total project value is expected to have impact because the larger the project value, the higher the risk for developers to finance the project on balance. I expect that a large project in terms of value is more often financed through project finance making use of the non-recourse characteristics of PF. The capacity of the wind farms is expected to have an equal impact. If only looking at the value of a project, we might miss valuable information since more expensive does not always imply higher capacity. On difficult locations (deep waters far offshore) a relatively lower capacity wind farm might be suitable, but more expensive. Equal for the opposite, a large wind farm on a relatively simple to build location, might be relatively less costly. Country is also an important determinant of using PF or CF. As mentioned in the theoretical framework, the United Kingdom was one of the early adopters of offshore wind farms because of both the renewable energy targets it has to comply to and the opportunities it sees to capitalize the knowledge as a possible export product. Another key
determinant is the subsidy regime the countries are exposed to. Each country has its own regime for providing subsidy and calculating subsidy prices. According to a study of PWC (2017) the United Kingdom mechanism differs from other countries on the European mainland. The so called Renewables Obligation Certificates (ROCs) leave interpretation of the subsidy price partly to the market development granting investors not fully covered investment. Therefore I expect that the CF/PF ratio will be different in the UK. I expect that participants in project finance require, as discussed in the theoretical framework, require certainty over margin and care less about upside potential (Guillet, 2010). The last variable considered in the regression is the year of financial close. I expect that in earlier years more projects were financed through CF due to the pioneering phase of utilities. Furthermore, financial institutions were not eager to invest or guarantee investments in unproved technological projects. This forced utilities to finance on balance, and kept project finance
out of the market. Especially after the financial crisis, when restraints on financial institutions were reduced, an increase in new investment opportunities such as offshore wind is expected.
The second hypothesis tests for possibility of the explanation of the value of ECA involvement by the expected major distinguishers value, capacity, leverage ratio post global financial crisis and a difference per country for which is controlled. I expect that for the very large and thus risky projects ECA involvement is inevitable because banks do not prefer to take the full risk. ECA involvement means shifting non-recourse risk towards participating governments. Post crisis, banks are risk averse, so it is expected that ECA involvement is higher for these projects. I further control for the leverage ratio. I expect the involvement of ECAs to be higher if the leverage ratio is higher. Higher leverage ratio means more risk for lenders. They prefer guarantees in return such certainty can be provided by ECAs. In the different countries, ECAs might play a different role. For Denmark ECA involvement might be limited due to the fact that most wind turbine manufacturers are located in Denmark.
Since the dataset is of limited size some key observations related to these parameters will be provided to give direction for further study.
Research methods
For the first research question a Probit equation will be used to test for significant difference in
probabilities between corporate financed investment projects and project financed investment projects. Due to the binary nature (limited dependent variable) of the dependent variable the Probit equation is the suitable solution. This model is not linear and is therefore not estimable using OLS but instead maximum likelihood is used (Brooks 2014). Brooks also indicates that both the Probit and Logit approaches are much preferred to the linear probability model. The only instance where the model may give different results is when the split in the dataset is very unbalanced e.g. 90/10 ratio. In this case logit is preferred. Since the dataset is not unbalanced, the Pobit mode will be used.
The following model will be tested:
( ) = 1 + ( ∗ 1 ∗ ∗ ∗
In words, the dummy project finance is 1 when the investment project is financed with project finance, 0 when corporate finance. Total value is the total USD value of the investment project.
Capacity is the capacity of the windfarm in Mega Watt. A dummy variable country has been added in
order to control for the effects per country and lastly, a variable for the year in which financial close has been obtained is part of the equation.
The second research question is answered by performing an Ordinary Least Squires (OLS) regression. The robustness of this model is confirmed by testing the model conform 5 assumptions that have to be confirmed in order to use the OLS regression. The output of the model including the robustness checks can be found in chapter 5, results.
= ∁ + + + + +
+ + + ℎ
Export Credit Agency involvement in USD can be explained by the total contract value, the capacity of the wind park, the leverage ratio of the investment, a dummy variable indicating the post financial crisis years, investments in the United Kingdom, Germany and The Netherlands.
4 Data and descriptive statistics
This chapter contains a detailed description on how the data was retrieved from the various data sources and which variables have been created in order to be used for the statistical tests.
The dataset
The basis of the dataset is retrieved from the website 4COffshore.com, the full dataset is accessible behind a paywall. The dataset used for this research is the version of July 3rd 2017. It contains an Excel spreadsheet with a total of 1539 wind farms located throughout the world. Total reported maximum capacity is 534,518MW. This figure is somewhat misleading because the dataset reports wind farms in various stages, from Development Zone to Decommissioned and everything in between. The projects that I require for this analysis are offshore wind farms located in North-West Europe with a capacity of 25MW or more that have reached financial close or final investment decision between 2010 and 2017. When limiting the dataset to these characteristics, the total amount of relevant projects is 329 with a reported maximum capacity of 136,799MW. The dataset of 4COffshore does also register the stakeholders registration for the projects. This includes, amongst others, the Export Credit Agencies, multilaterals and commercial banks. After detailed analysis of this data it was concluded that this information is not registered in deal tranches, the data was not always up to date and the detailed level of registration was not sufficient.
To be able to test the detailed effects of financial involvement of the various parties, the 4COffshore dataset is complemented with detailed tranche specific debt and equity financing information. It turned out that these data are not easily accessible. For this reason I have combined 3 financial databases in order to build the dataset so that all information sources could complement and/or verify each other. The databases used are TagMyDeals.com, Infra-Deals.com and IJGlobal.com. These websites are commercial companies that track financing activities in the market. They acquire data both from market research, news items on a topic, and from registration by involved parties, that in return receive access to the data of other projects. All databases are accessible behind the paywall on the individual websites. All 3 websites provide detailed information on individual projects and disclose the way the loan tranches are set up, parties that are involved such as banks, multilaterals and export credit agencies and sometimes they provide margins per individual debt tranche. Both IJGlobal and Infra-Deals also disclose information on the equity side. This allows me to get a view on the total financings of the individual projects SPVs and calculate the leverage ratio.
The combination of sources allowed me to build two datasets. The first dataset provides a per project detailed registration on individual debt tranche level. If a project was financed through Corporate Finance, the total value of the project was registered as an equity tranche. In case the project was financed through Project Finance, the equity part was registered as one tranche. Subsequently the
different debt tranches are registered. Per tranche the notional in USD, measured in millions, debt structure, tenor, reference rate, margin and guarantor are identified. For the borrowers I made a distinction between Commercial Banks (including pension funds), Multilateral Institutions, and in case of direct lending, Export Credit Agencies. Subsequently I registered in case of ECA backing, which ECA was covering which debt tranche. Based on the before mentioned data I am able to create a set of dummy variables that are required for the regressions. I created dummy variables for the
participating debt provider Dummy_commercial_banks. I create a dummy on non-recourse funding and registered loan tranches where the data source specifically mentioned that it concerned non-recourse debt, dummy_non-non-recourse. Since in many of the tranches the data source did not specifically mention the non-recourse characteristic of the debt, I gave my own interpretation on the information and created a dummy called Dummy_Non-recourse_Assumption that informes if a debt tranche is, what I assume, on a non-recourse basis. The other dummies created are
Dummy_Multilateral_institution_involved, Dummy_ECA_guarantee, Dummy_ECA_direct_lending, Dummy_ECA_involved, Dummy_any_government_support. All resulted in a set of 247 tranches with a total value of USD 107,7 billion, 116 equity tranches with a total value of USD 75,9 billion and 130 debt tranches of in total USD 31,9 billion.
Graph 1: Total value of offshore wind farm investments divided by financing type
The second dataset that is created is a dataset on the aggregates per project. The total number of projects that meet the criteria stated above is 86. The total value of all projects is (of course) equal to the total of all debt and equity tranches, USD 107,7 billion. This dataset is built in order to show the aggregates per project. The basis of the information comes from the 4COffshore dataset and includes information such as project name, unique project number, country of execution, date of financial close/final investment decision. I added to this set a couple of extra fields. The first is the total value of the project, the way of financing (project finance/corporate finance), the leverage ratio of the project, ECA_involvement_count which counts the number ECAs per project. ECA_involvement_y/n shows if an ECA is involved, ECA_involvement_value provides the total value in USD that one or more ECAs provide to the specific project. These same variables have been incorporated for Multilateral Institutes. I also created the variable for government support GovSup with all the above sub registrations. GovSup indicates the total involvement of governmental institutions such as ECAs and multilateral banks. The next steps are adding a variable for the year in which the Financial Close (FC) or Final
Investment Decision (FID) was obtained (FC_Year). Accordingly the start and end date of the Global Financial Crisis (GFC) are determined. The start and end date chosen are in line with the paper of Ivashina and Scharfstein (2009). By allocating individual projects to one of the groups, this resulted in 3 variables: Pre-GFC, 22 projects, 5 project in_GFC and 59 projects Post_GFC. Furthermore the dataset of 4COffshore contains the variables Capacity MW (Max), Turbine MW (Max) and
DistanceFromShoreAuto that have been included in the set. All other, mostly technical registrations, are removed from the database.
Descriptive statistics
Descriptive statistics dataset 1
The first dataset contains investment projects that obtained financial close from 2000 to 2017 in the countries Belgium, Denmark, Finland, Germany, Ireland, Netherlands, Sweden and the United
Kingdom. The total number of observations is 86. The average capacity for the projects was 259 Mega Watt (M = 258.7, SD 202) and the total average project value per investment is USD 1.25 billion (M = 1.25 billion, SD 1.05 billion). 27 of the projects are project financed, the other 59 financed using corporate finance. Project Finance Dummy Total project value in USD Capacity in MW Mean 0.3139 1252.455 258.7128 Median 0 948.485 208.15 Maximum 1 4176 1218 Minimum 0 40.5 25.2 Std. Dev. 0.4668 1057.568 202.0441 Observations 86 86 86
Table 1: Descriptive statistics on dataset 1
The correlation between the project value and the capacity is positive and high (0.88) which can be reasonably explained because in many cases you would expect that an increase in capacity causes an increase in the USD value. Other correlations are low and are presented in the below table.
Project Finance Dummy Total project value in USD Capacity in MW Project Finance Dummy 1 0.17 0.07 Total project value in USD 1 0.88
Capacity in MW 1
Table 2: Cross-correlation diagram dataset 1
Descriptive statistics dataset 2
In total 27 projects are project financed. ECA involvement per tranche is on average USD 297 million (M = 296.9, SD = 542.1). The maximum involvement of ECAs in various tranches for one investment project is USD 2.6 billion. The value of the projects is USD 1.5 billion on average (M = 1523, SD = 975) with a maximum project value of USD 4.1 billion. Leverage ratio for the project financed projects is high on average, 74% (M = 0.74, SD = 0.08) remarkable is the extremely high leverage ratio of one of the projects namely 93.6%. Only 2 projects were not closed after the financial crisis. 7 projects have
been closed in Belgium, which is the reference country in this dataset. In Germany 11 projects were executed, United Kingdom 5 and in The Netherlands 4.
ECA involvement Value
Total project value in USD Capacity in MW Leverage ratio Post GFC Mean 296.8942 1523.83 280.5148 0.7418 0.9259 Median 87.31 1235.82 288 0.7235 1 Maximum 2604 4123.99 600 0.936 1 Minimum 0 268.85 48.3 0.5603 0 Std. Dev. 542.1155 976.1584 145.0755 0.0832 0.2668 Observations 27 27 27 27 27 Table 3: Descriptive statistics on dataset 2
The only high correlation observable in the dataset is between capacity and USD value (0.91). Other correlations are low and can be found in table 4 below.
ECA involvement Value Total project value in USD Capacity in MW Leverage ratio Post GFC ECA involvement Value 1 0.53 0.41 -0.01 0.16 Total project value in USD 1 0.91 0.05 0.23
Capacity in MW 1 0.24 0.14
Leverage ratio 1 -0.04
Post GFC 1
5 Results
The below figures gives a visual overview of how the two research questions relate to one another. The total dataset contains 86 projects (The # indicates the number of projects). All projects are used to answer research question 1. The sum of the total project values is USD 107.7b and the total capacity is 22,249MW. 59 projects are financed using corporate finance and 27 projects are financed using project finance. For research question 2 only the project that are financed using project finance are used. A further distinction within these projects can be made based on the involvement of Export Credit Agencies. Of the 27 project financed projects, 16 have involvement of ECAs with a total value of USD 26.1 billion.
Figure 1: Visual overview of how the datasets of the two research questions relate together
The first hypothesis is whether there is a significant difference between the financing of offshore wind farms via corporate finance or project finance results in the following statistical output. The following model is tested using a Probit regression:
( ) = 1
We interpret the significant probability of year in such a way that an increase in year makes project finance more likely. Financial Close Year is significant with p=0.0251. Another observation is the importance of the country code for determining the likelihood of project finance over corporate finance (p=0.0058). When zooming in to the dataset, it can be observed that especially the United Kingdom plays an important role as a country being an important factor for the negative coefficient. The McFadden R-squared is 0.19 which indicates the fit of the model. This is not a very large score, according to Brooks (2014) this is often the case for limited dependent variable models. Total value in USD is almost significantly impacting the choice for project finance, but with p=0.087 not statistically significant.
Dependent variable: Project finance dummy Number of observations: 86
McFadden R-squared: 0.1931 Observations with Dep=0: 59 Observations with Dep=1: 27
Variable Coefficients Std. Error z-Statistic Probability Constant -201.7728 90.1264 -2.2387 0.0252*
Total project value in USD 0.0006 0.0003 1.7111 0.087
Capacity in MW -0.0032 0.002 -1.5821 0.1136
Country code -0.17 0.0616 -2.7592 0.0058**
Financial Close Year 0.1004 0.0448 2.2397 0.0251* Note: * p<0,05 ** p<0,01
Table 5: Output Probit regression
Over the total period 69% of the wind farms, 59 in number, have been financed using corporate finance. Of the total value this is 62% or USD 66.6 billion and of the total capacity in MW this is 66% or 14,675 Mega Watt.
Explanation for this is expected to be found in the location of the project. In the United Kingdom, the projects utilities are the main owners of the wind farms. Utilities such as DONG Energy and Vattenfall have large balance sheets and tend to use corporate finance for financing the investment projects. As the hypothesis describes, this is a logical finding, since utilities started exploring and investing in this market being capital intensive companies. Project finance was also not expected given the subsidy regime in the UK. Financial Close Year is also a significant determent factor for project finance. As observed in the data, in the more recent years, we see more project finance. This can be explained by increasing willingness from the financial sector due to available investment funds but also an
increased proof of work by successful operating wind farms.
In order to be able to test if the Probit model fits, the Huber/White method has been used in order to work with robust covariance. This was directly applied to the model when executing the regression.
Furthermore the Hosmer-Lemeshow test for the goodness of fit was performed. The result (p = 0.4346) indicates that the null hypothesis need not to be rejected and the model has a good fit and can be used for performing the regression.
Research question 2
In part two of the research question we only look at the investments financed through project finance. The total number of observations is 27. It is tested what the effect of the dependent variables total USD value, capacity, leverage ratio, post global financial crisis, United Kingdom, The Netherlands and Germany are on the dependent variable ECA involvement in USD, measured in millions. Belgium in this case is our reference country. The results are presented in table 6.
= ∁ + + + +
+ + + ℎ
The model provides insight that the overall fit is relatively high with an R-squared of 0.76 and is significant p<0.05. Meaning that the variables chosen explain the availability of ECAs in offshore wind farms. Both Netherlands and Post Global Financial Crisis are not significant variables for determining the value of ECA involvement, all other variables are. However, Post GFC is almost significant with p=0.06. The main take away from this model is that with an increase of 1% leverage, the ECA involvement increases by USD 29.9 million.
It can be observed that the total USD value has a positive impact USD with the proportion of total value, country. About 66.6% of the variation in the proportion of ECA involvement can be explained by the explanatory variables.
Dependent variable: ECA involvement Value Number of observations: 27 R-squared: 0.7563 Variable Coefficients Std. Error t-Statistic Probability Constant -1441.089 672.5726 -2.1426 0.0459*
Total project value in USD 1.1458 0.2283 5.0184 0.0001***
Capacity in MW -4.4738 1.384 -3.2324 0.0044**
Leverage Ratio 2993.928 1012.967 2.9556 0.0081**
Post Global Financial Crisis -619.5411 313.1809 -1.9782 0.0626 United Kingdom -1362.273 277.9306 -4.9014 0.0001***
Germany -331.8866 157.0288 -2.1135 0.048*
Netherlands -87.2357 233.5662 -0.3734 0.7129
Note: * p<0,05 ** p<0,01 *** p<0,001
Various assumption been tested in order to confirm that the hypothesis tests using the OLS-regression could validly be conducted. In order to be able to use the standard OLS regression, 5 assumptions have to be tested to confirm that the estimation technique can be used (Brooks, 2014). The 5 assumptions are discussed in high level. Assumption 1: Average value of the errors is 0. A constant term was added to the equation, so this assumption is not violated. Assumption 2: Assumption of homoscedasticity. This assumption tests if the errors have a constant variance. If not, they are hetroscedastic. This is done by performing White’s test. The output of the test is a value. The p-values for the test (p = 0.19) is in excess of p = 0.05 so there is no evidence for the presence of heteroscedasticity. Assumption 3: covariance between error terms over time is zero. This assumption indicates that the error terms are uncorrelated with one another, or the absence of autocorrelation. This is tested by the Breusch-Godfrey test. The test result (p = 0.65) is a value larger than p = 0.05 so we do not reject the 0 hypothesis of autocorrelation. Assumption 4: regressors are non-stochastic. Since E(u) = 0, this expression will be zero and therefore the estimator is still unbiased, even if the regressors are stochastic. Assumption 5: the disturbances are normally distributed. The p-value when performing the Jarque Bera test (p = 0.94) is larger than p = 0.05. I therefore do not reject the null hypothesis of normality at the 5% level. Based on the additional tests of the individual assumptions of the desired properties, the assumptions are not violated and the hypothesis tests regarding the coefficient estimates could validly be conducted using an OLS regression.
The data consists of 27 projects of which 7 in Belgium, 11 in Germany, 4 in The Netherlands and 5 in the United Kingdom. Two of the projects achieved financial close before the global financial crisis. None of the investments during the financial crisis and the majority of 22 after the GFC. Of the total project value of USD 41.1 billion, USD 30.6 billion is debt. This accumulated to a leverage ratio of 0.74. Participation of ECAs in the various debt tranches accumulates to USD 8 billion which is 26% of the total debt over 16 investment projects. The total level of government support in the debt tranches is USD 22.7 billion, this is 74% of the debt.
An interesting observation from the individual debt tranches, is the level of pure non-recourse financing. According to the assumptions made in the dataset, the majority of 92 of the 128 debt tranches have a pure non-recourse basis. Zoomed in on the notional of the individual tranche, this relates to USD 21.1 billion of the debt value which is 67%.
6 Discussion
The aim of this thesis was to discover what the role is of government-backed Export Credit Agencies. The study shows that in order to observe ECA involvement, a project has to be financed using project finance. A determining factor for project finance is the country in which the investment takes place, mainland Europe has more project financed investments. The year of financial close is also determinative. In more recent years, more project finance is observable. Additional research on a subset of investments leans us that with an increase of 1% leverage, the ECA involvement increases by USD 29.9 million.
Based on the theoretical framework, the market for offshore wind farms is still in an early phase of development. Theory also focuses on the participation of export credit agencies in the developing world and not so much in the developed world. Some characteristics of standard project finance models have been described.
The data learns us that during the years, the more recent years predict more offshore wind farms financed through project finance. This can be explained by the fact that in the starting years of offshore wind, mainly utilities were the key players in the field, because of their expertise but also the undesired hassle and lead times of banks they choose for the on balance financing. It has not been confirmed that this also relates to the cheap cost of funds during the post global financial crisis, this could be an interesting field for further research. Detailed information on corporate financed offshore wind farms is difficult due to the limited disclose of utilities. When utilities have de-risked their assets for a sell-off, part of the project value can be observed via public market information.
ECAs are of growing importance to the offshore wind market. In 60% of the cases, an ECA is involved in the financing. Detailed observation of the data learns that it are mostly ECAs from Japan, Denmark and Germany which can be explained by the fact that turbine manufactures are located in these countries. In order to stimulate their export, ECAs are in the game. In one of the cases, a home country ECA was involved in the financing, raising the question of a level playing field amongs ECAs and the interpretation of the sector understandings for renewable energy. Since it relates to an individual case, this exemption might not be interesting for further research.
Leverage ratios for project financed offshore wind farms is in line with the 70-75% range mentioned in literature. The average leverage ratio of the sample is 74%. The non-recourse characteristic of PF is also interesting. Based on a created dummy on the interpretation of non-recourse, the dataset learns that of the total debt, only 69% is expected to be on a pure non-recourse basis, the other 31% is covered by ECAs or multilateral institutions. This lower risk profile is interesting for both the bank, theory learns that this is considered to be government risk instead of project risk. Also the project
entity benefits from this due to lower margins give the risk profile. The regression shows that both the leverage ratio and total project value have a positive influence on the amount of project finance.
The theory says that a growing political risk in project finance is allocated to ECAs. However, in northern Europe this political risk is almost non-existing, so there must be another reason if they participate. Given the fact that ECAs are additional to the market, it can be assumed that the involvement is necessary in order to reach financial close, which would not be achieved by involving commercial parties.
Given the exploratory nature and the creation of these models, one should be very careful in
interpreting the results. Both models provides significant results on independent variables, but it is not possible to determine whether this related to a correlating effect, or that the model really lays out an causal effect of the independent variables.
The most important limitation that I experienced during this research, is the amount of available existing literature on ECAs. There has not been much written on the practices of these institutions in the developed world. Useful literature, if any, focusses on the activities in de developing world. Despite the fact that I could find financial information related to all projects, there are some limitations which I will discuss to provide an objective opinion on the data. First, there is no single database that both provides all the required information and can also be used to answer the detailed and specific questions of this thesis. By combining the data sources, I found that even though all sources have information on some projects, the reliability is not always granted. I have included an example comparison in Appendix 1. I concluded that however that the various sources have different registrations on the same deal. It cannot be taken completely as a fact, but it provides at least a direction of the truth and can be used to compare as a sample without looking too much to individual projects. I have always used the data source with the most extensive and in debt information and assumed this data could be provided because the most reliable information was disclosed to these sources.
Margins usually change over time. During the construction period margins are usually higher and during the first years of operations lower. Over time margins may gradually increase. Because information on margins is available in a very limited number of tranches, I have taken the average margin that was published (e.g. (Construction margin + short term margin + long term margin)/3). For the regressions, I unfortunately had to leave the margins out because they were provided only incidentally.
The value of tranches was at least available in the currency of the debt (usually EUR or GBP). Some sources also provided the USD counter value. If stakes were not provided in USD, this has been calculated accordingly using the exchange rate at the day of financial close.
The different data sources labelled the portions of debt differently. Sometimes the qualification of the tranche such as ‘senior debt’ was clearly stated, in other cases the debt tranche was just referred to as ‘debt’. Due to this difference I was not able to make a clear distinction between subordinated debt and senior debt. Sometimes the subordinated debt was clearly stated, sometimes not. Therefore I assumed that all debt tranches indicated on the websites are considered as senior debt in order to calculate the leverage ratio per project.
For future research on role of ECAs in the developed world, especially in offshore wind, it would be interesting to zoom in on the credit spread of the various ECAs and if these are in line with the level-playing field as indicated by the OECD although it might be very hard to obtain such data. This study could be accompanied by a further comparison between the willingness of ECAs to participate in deals for the developed world.
References
Berne Union, (2017). About the Berne Union. Retrieved May 4, 2017 from: http://www.berneunion.org/about-the-berne-union/berne-union-vision/
Bodnar M., Comer B. (1996), Project finance teaching note, The Wharton School
Brooks C., (2014), Introductory Econometrics for Finance, 3rd Edition
Buscaino V., Caselli S., Corielli F., and Gatti S. (2012), project finance collateralised debt obligations: an empirical analysis of spread determinants, European Financial Management
Byoun S., Kim J., and Yoo S. (2013), Risk management with leverage: evidence from project finance, Journal of Finance and Quantitative Analysis
Corielli F., Gatti S., Steffanoni A. (2010), Journal of money credit and banking
EIB, (2009). EIB Financing of EUR 300 million for Belwind offshore windfarm. Retrieved July 12, 2017 from: http://www.eib.org/infocentre/press/releases/all/2009/2009-156-eib-financing-of-eur-300mio-for-belwind-offshore-windfarm.htm
Esteban, M. Dolores ; Diez, J. Javier ; López, Jose S. ; Negro, Vicente Why offshore wind energy? Renewable Energy, 2011, Vol.36(2), pp.444-450
Esty B. (2003a), The economic motivations for using project finance, Harvard Business School
Esty B. (2003b), When do foreign banks finance domestic investment? New evidence on the importance of legal and financial systems, Harvard Business School
Esty B. (2004), Why study large projects? An introduction to research on project finance, Harvard Business School
European Commission, (2015). The Energy Union on track to deliver. Retrieved August 4, 2017 from: http://europa.eu/rapid/press-release_IP-15-6105_en.htm
Gianturco, D.E., (2001) The Unsung Giants of International Trade and Finance, 1st Edition
Girardone C., Snaith S. (2011), Project finance loan spreads and disaggregated political risk, Applied Financial Economics
Goldzimer A. (2002), Globalization’s most perverse secret: the role of export credit and investment insurance agencies, Alternatives to Neoliberalism Conference; May 23-24; 2002
Grushkin J., Bartfeld D. (2013), Securitizing project finance loans: are PF CLOs poised for a comeback?, The Journal of Structured Finance
Guilliet J., (2010), The Principles of Project Finance – Financing Offshore Wind, pp. 183-191
Hainz C., Kleimeier S. (2003), Political risk in syndicated lending: theory and empirical evidence regarding the use of project finance, LIFE Working paper 03-014; June
Ivashina, V., Scharfstein, D., 2010. Bank lending during the financial crisis of 2008. Journal of Financial Economics 97, 319–338
Modigliani, F.,Miller, M. Theory of Investment, The Cost of Capital, Corporation Finance and the Theory of Investment. The American Economic Review, Jan 1, 1958, Vol.48, p.261
Mukherjee A., Chatterjee P. (2015), A framework for understanding and modeling risk in mega projects, The Journal of Structured Finance; Vol. 42, No. 7; 2010
PricewaterhouseCoopers (2017), Unlocking Europe’s offshore wind potential, Retrieved May 4, 2017 from: https://www.pwc.nl/en/publicaties/unlocking-europes-offshore-wind-potential.html
Sorge M., Gadanecz B. (2007), International Journal of Finance and Economics; 13: 68-81; 2008
Srivastava V., Restructuring project finance bank debt in India: information asymmetry and agency costs
UN, (2017). Paris Agreement: essential elements. Retrieved August 2, 2017 from: http://unfccc.int/paris_agreement/items/9485.php
UN, (2017). Kyoto Protocol. Retrieved August 2, 2017 from: http://unfccc.int/kyoto_protocol/items/2830.php
Vaaler et al. (2008), Risk and capital structure in Asian project finance, Asia Pacific Journal of Management, March 2008, Volume 25, Issue 1, pp 25–50
World Bank, (2017). Project Finance – Key Concepts. Retrieved May 12, 2017
from: http://www.eib.org/infocentre/press/releases/all/2009/2009-156-eib-financing-of-eur-300mio-for-belwind-offshore-windfarm.htm
Appendix 1 – Example of data imperfections
Example of imperfections in available data, confronted with validation dataset
Project:
NL18
Gemini
IJGlobal
InfraDeals TagMyDeals
Subordinated Loan
274
274
Commercial term loan + EDC
1344
1156
1160
Commercial + EKF
265
210
Commercial banks + Euler Hermes (95%)
480
481
Euler Hermes
480
EIB
809
274
275
EIB B + ONDD
247
247
EIB C + EKF
284
284
Standby facility
173
173
Total debt
2907
3152
2830
Total bank debt financing (sum on site)
2907
3176
2880
Equity mentioned on website
972
606
Implied Debt ratio
0.75
0.84
1
Implied Equity ratio
0.25
0.16
0
Tenor mentioned
17 1/12
17
18
Info on spreads
Yes
Yes
No
EUR USD rate used
1.37
1.38
Margins mentioned
300bps construction, 275bps completionc. 300bps
Appendix 2 – Export Credit Agencies at work
JBIC in Eneco Luchterduinen (Q10) offshore wind farm
“With offshore wind power generation being one of the new frontiers of renewable energy, Japanese companies have not yet managed to build a track record in such projects overseas. JBIC’s support for Japanese companies to participate in such projects through long-term financing will contribute to maintaining and strengthening the international competitiveness of Japanese industries.” JBIC (March 11, 2016)1
EKF in Northwind
Due to the financial crisis the banks were reluctant to offer loans to Belgium's new large offshore wind farm, Northwind. A commitment from EKF kick-started the negotiations for the establishment of the project which is based on wind turbines from Vestas and financing from among others, the Danish pension fund PensionDanmark.
Offshore wind farms typically require loans with a credit period of 15-17 years to ensure a proper balance between repayment and earnings. But the financial crisis turned the loan finance into a major challenge for Northwind.
”We found that it had become more difficult as well as more expensive to involve the banks in long-term financing. However, the banks would be willing to finance the project if we were able to present a strong business case”, says Francois van Leeuw, CFO, Northwind Offshore Energy.2
”We were therefore forced to provide security for the financing before initiating negotiations with banks and financial partners. This was to be achieved by obtaining fool proof contracts with suppliers, guarantees from ECAs and com-mitments regarding insurance of the wind farm,” says Francois van Leeuw.
EKF in Offshore Wind
EKF can help you obtain financing for your wind project whether you choose a financing model based
1 http://www.jbic.go.jp/en/information/press/press-2016/0311-47255 2 http://www.ekf.dk/en/WhatWeDo/Pages/Northwind.aspx
on corporate risk or in the form of project financing.3 4 reasons to involve EKF in connection with
financing: (1) EKF is the leading player in the market. We are the world´s leading export credit agency within wind energy. Over a period of 10 years, we have been involved in approximately 60% of the total wind financing covered by export credit agencies worldwide. Our long-standing experience and insight into the area have made us a central business partner. (2) EKF assumes the biggest risk. We are normally the ones among the syndicated lenders who carries the biggest risk, often assuming a greater risk than the banks. We enter into repayment agreements based on fixed terms which ensures financial predictability throughout the tenor of the loan. (3) EKF has an extensive network. We are in contact with two of the largest turbine producers in the global market and a large number of sub-suppliers from the wind industry. We have worked closely with banks, developers, sponsors and advisors from the industry worldwide, including the European Investment Bank (EIB), who has participated in several of our business transactions. (4) EKF assumes the risk even in challenging markets. We are willing to take a risk and participate in business transactions even in challenging markets where the banks are less accommodating.
