To find the APV of the operational phase, we need to determine the cost of debt as well as the standard terms for wind farm project finance in Denmark. Based on this information we can calculate the value of the interest tax shield and find the APV of the operational phase.
5.3.1. Project Finance in Denmark
In Denmark the two most active project finance banks are Ringkjøbing Landbobank and VestjyskBANK with headquarters in western Jutland, an area with a long tradition for wind farm construction. It is therefore likely that EE will finance their wind farms through one of these two banks. In addition, Nordea also has a project finance office, although it is not as active in the Danish wind farm market. The following information about project financing is provided through interviews with the three banks. The interviews are summarized in Appendix 16.
The three banks all structure the financing of Danish wind farms in a similar way. The loans are structured like an overdraft, where the entire cash flow is used to repay the debt. The debt has to be fully repaid before the equity holders receive anything. The length of the loan thus varies according to the principal, the production of the turbines and the cost of debt etc. Typically the banks’ target is that the loan should be repaid approximately 8-13 years into the operational phase, which determines the leverage of the individual project. This typically equals a leverage of approximately 65-90% of the initial investment. Based on previous experiences from their Italian and German wind farms, EE aims for a leverage ratio of 80%. We have therefore chosen 80% as our initial leverage ratio, which results in the loan being paid back in 7 years.
102 The cost of debt is in the three banks determined by the 6-month Copenhagen Interbank Offered Rate (CIBOR) and the debt risk premium as seen in the formula below.
Formula 5.3 Cost of Debt
𝐫𝐝𝐞𝐛𝐭=𝐂𝐈𝐁𝐎𝐑 𝟔+𝐃𝐞𝐛𝐭 𝐑𝐢𝐬𝐤 𝐏𝐫𝐞𝐦𝐢𝐮𝐦
Source: Ringkøbing Landbobank, Vestjysk BANK and Nordea
The CIBOR is not equal to the risk-free rate since it includes a credit risk premium of lending to commercial banks. The existence of such a credit risk premium is on international markets usually expressed by the TED spread. This spread measures the London Interbank Offered Rate (LIBOR) versus the 3-month T-bill interest rate (Cheung et al. 2010: 88), thereby expressing the difference between the short term risk-free rate and the interbank rate.
However, the debt risk premium is more individual and typically depends on the individual project specifications, the professionalism of the developer, the relationship to the bank etc.
5.3.1.1. The 6-month Copenhagen Interbank Offered Rate
The 6-month CIBOR is a daily reference rate based on the interest rates which banks lend and borrow unsecured fund for in Copenhagen’s wholesale money market. Since CIBOR is a daily rate, it is unlikely that it will remain constant through the life of the project and these fluctuations will change the actual interest payments and thereby the ITS of the wind farm. It is therefore necessary to identify a reasonable proxy for a longer period. This is technically the same problem as in chapter 4, when we discussed the risk-free rate used in the CAPM. There we pointed out that the theoretically correct approach is to discount cash flows with a discount rate of a matching maturity.
In spite of this we saw that in practice a 10-year government bond was used to set the risk-free rate for most corporate valuations, and that in general the discount rate is kept constant through the life of the project. Based on a similar approach we have chosen to apply the 10-year Danish swap reference rate, as a proxy for the long term CIBOR. This rate was 3.90 % (Nationalbanken) the 30th of December 2009. We use the swap reference rate instead of a government bond as this can be perceived as the long-term fixed-for-floating interbank rate (Finansraadet 2010).
5.3.1.2. The Debt Risk Premium
Besides the interbank rate, the cost of debt also contains a risk premium to adjust for the project’s risk of default. This is typically denoted in basis points (BP). When estimating the debt risk premium, we face many of the same issues as the ones from the beta estimation in section 4.2.3.2.
For example, the debt risk premium should reflect the individual project’s risk of default and not EE
103
as a company. One way to estimate the risk premium is to identify a frequently traded corporate bond or index within the same industry to calculate a debt beta (Koller et al. 2005: 327). In our case, finding such a bond or index is not possible. Furthermore, this technique suffers from the problem of debt betas declining as the debt matures making the correct estimation very difficult (Cooper and Davydenko 2007: 91).
Due to these problems, it is not possible to use market data to estimate the debt risk premium of our loan. Instead, we have chosen to base this on the interviews conducted with the three banks and historical risk premiums provided by EE for wind farms in other markets. The three Danish project finance banks that we have interviewed have all mentioned risk premiums for professional investors at around 300 BP, depending on the individual project and the developer. This is slightly higher than the risk premiums of around 150-280 BP, which EE has experienced in other markets. This could reflect a higher risk in Denmark, because a large part of the tariff is the market price of electricity.72 According to project manager Andreas Von Rosen from EE, the financial crisis has made the banks increase the debt risk premium compared to earlier – therefore a risk premium somewhere between 200-300 BP seems more realistic according to him. Based on these considerations, and the information provided by the bank, we have chosen to set the debt risk premium at 250 BP.
Having estimated both the debt risk premium and a proxy for the long term CIBOR rate for our loan, we can now estimate the cost of debt for out project in the formula below.
Formula 5.4 Cost of Debt (Calculation)
𝐫𝐃𝐞𝐛𝐭=𝐂𝐈𝐁𝐎𝐑+𝐃𝐞𝐛𝐭 𝐑𝐢𝐬𝐤 𝐏𝐫𝐞𝐦𝐢𝐮𝐦 𝟔.𝟒𝟎% =𝟑.𝟗𝟎 % +𝟐.𝟓%
Source: Interviews with banks and EE73
5.3.2. The Value of the Interest Tax Shield
With a cost of debt and a leverage ratio, we can now estimate the value of the ITS. The ITS was defined as the amount of the interest rate payments that can be deducted from the income tax of the project and is calculated using Formula 5.5 below.
72 This risk premium estimate was provided by EE as estimates from project financings of both solar and wind farms undertaken during the last three years in Germany, Italy and Spain.
73 It is interesting to note how close the cost of debt is to cost of equity for wind farms in Denmark. The proximity of the numbers can be explained by different factors such as the low market risk premium for equity, the low beta for IPPs, the high leverage and the CIBOR being higher than the risk-free rate.
104
Formula 5.5 Value of Interest Tax Shield
𝐕𝐈𝐓𝐒=�𝐫𝐝∙ 𝐃𝐭∙ 𝐓𝐂
𝟏+𝐫
𝐭=𝟏
𝑽𝑰𝑻𝑺: Value of interest tax shield 𝒓𝒅: Cost of debt
𝑻𝒄: Marginal corporate tax rate 𝑫𝒕: Outstanding debt
r: Discount rate Source: Own construction with inspiration from Brealey et al. 2006: 521. 74
In the above formula, the value of the ITS increases with all three variables in the numerator. While academics agree on the terms in the numerator of Formula 5.5, the discount rate in the denominator is a debated issue. Luehrman (1997: 151) argues that the cost of debt can be used, based on the notion that the interest tax shield will only be used in the situations where debt can be repaid, and therefore it is approximately as risky as the interest payments and installments. Others argue that since it is not sure that the future ITSs can be used, it should be discounted at a rate according to the prevailing business conditions. This seems particularly appropriate if the company is constantly rebalancing its debt-equity ratio. Based on these two arguments, Luehrman (1997: 151) ends up discounting the ITS with an interest rate subjectively set “somewhat” higher than the average cost of debt.
On one hand, the idea that the risk of the ITS is not equal to market risk makes intuitive sense, as its value is determined from the Cost of Debt (rd), The Initial Outstanding Debt (D0) and the Marginal Corporate Tax Rate (Tc), which are all relatively fixed variables. On the other hand, the project must generate an income to shield for the ITS to have a value. In Denmark, the issue of not having enough income to shield is not as big a problem as elsewhere, due to fact that the ITS of a given year can be “carried forward” to future years, where they can be used.75 Hence we find it appropriate to use the cost of debt as a discount rate to find the value of the ITS.
Following these consideration we can now calculate the value of the ITS in Figure 5.1 below. The calculations should be seen as an additional calculation to the free cash flow of the operational phase from Figure 4.4 p. 71, and can be seen in full scale in Appendix 19.
74Formula 5.5assumes a full tax deductibility of interest rates, this is the case in Denmark but not in many other countries e.g.
Germany, where only 50% of the interest rates can be deducted.
75 Information on Danish accounting principle provided by Chief Accountant Peter Christiansen from EE.
105
Figure 5.1 ITS Value in the Operational Phase
Source: Own construction
As can be seen from the table above, the project does not pay any tax in the first years, due to the large asset depreciation. Hence the ITS will be carried forward until the project starts to generate a tax, which is then shielded. In our case the last cash flow from ITS will occur in year 2020. The Tax Saving from ITS represent the DCF value of the tax benefits when project financing is introduced and represent the value of the ITS. To adjust for the timing of these cash flows, we add a mid-year factor and get a value of the ITS of 2,155,734 DKK.