Exchange with Olivier Leruste,
Head of Financing at Akuo Energy

In early March 2019, Akuo announced the start of work on a solar power plant with 187,000 solar panels (50 Mw) in Kita in Mali. We thought it would be interesting to look into the financing of this sort of project. Kita will be the biggest solar energy plant in Africa (excluding South Africa). The project was launched through a Build Operate Own and Transfer (BOOT) concession which means that Akuo will operate the power plant for a period of 30 years.

Project financing is based first and foremost on an analysis of the risks involved and on the allocation of these risks to the various stakeholders:

Construction risk. Olivier Leruste informs us that the construction risk in the case of Kita is low. The technology involved has been tried and tested (these are not floating wind turbines), it is relatively simple to put in place and does not require any specific engineering structures (such as a dam). The construction risk has mostly been transferred to the companies to which the construction has been subcontracted (companies specialising in Engineering Procurement and Construction or EPC).

Generation volume risk. The operator does take on this risk but it is relatively limited for a solar power plant (+/-5%) and it’s a risk that tends to fade over the years of generation. Wind turbines, for example, are more erratic and so the generation risk is higher.

Price risk. Firstly, it should be borne in mind that the price of electricity generated by a solar power plant is competitive compared with a thermal power plant. In the case of Kita, the price is guaranteed by a fixed-price Power Purchase Agreement (PPA) with the national operator in Mali, Électricité du Mali. The PPA has been signed for 28 years and guarantees Kita a purchase price for all volumes generated. So, it won’t be very hard to fill in the figure for Sales in the Business Plan!

Policy risk. This risk may appear to be substantial in Sub-Saharan Africa, but it can be covered by a country risk insurance policy offered by specialised insurers (Coface, Euler-Hermès, Lloyds, etc.). Akuo notes that recourse to a multinational agency for financing is a solid guarantee of the durability of the agreements made with the Malian government and its statutory bodies.

Currency risk could be a real issue. PPAs are generally based on a strong currency (the US dollar). This is a major political challenge for countries that are then forced to make long-term spending commitments in a foreign currency. In the case of Kita, the PPA is in CFA francs which has a fixed parity with the euro. This reduces the risk and additionally, the contract includes a clause providing for the revision of the price of purchasing electricity from the plant in the event of a devaluation of the CFA franc. 

For each of its projects, Akuo produces a risk matrix showing the risk assumed by each party and the remaining risk assumed by whoever is providing the shareholders’ equity (i.e., in this case Akuo). Olivier Leruste believes that the risk is at its highest for Akuo the day before closing. This is because at this stage, major design costs have been incurred, and these costs will be written off if the project fails to materialise. The risk declines sharply after two or three years of operation when the power plant has demonstrated its generation capacity and maintenance costs are under control.

So, negotiating a watertight PPA is crucial as this is the agreement that, for the most part, defines the allocation of risk between the operator and the purchaser of electricity (i.e. the Malian government). The more risk allocated to the government in the agreement (price, currency, volume, governing law, possible termination, indemnity clauses, etc), the more bankable the project will be, but obviously it will be less advantageous for the government. It can be quite difficult to find the right balance.

The total cost of the Kita project is around €85m. The financing is more or less standard with a tranche of senior debt, a tranche of junior debt and shareholders’ equity. In France, the rest of Europe and in the US, the debt would have been provided by a commercial bank. In emerging countries, local commercial banks are often too small and do not have sufficiently specialised know-how that is needed to analyse project finance. Most western banks are generally a bit nervous about taking on country risk so it’s usually local or international development banks that step in.

The senior financing was determined based on a business plan with robust banking assumptions and a 90% probability rate, which is standard for this type of project. The debt was then calibrated so that debt servicing would account for 2/3 of free cash flows (Debt Service Cover Ratio (DSCR) of 1.5). This is a relatively cautious figure as in western countries, the DSCR would be around 1.15. The senior financial leverage on the Kita project is 70%.

The senior debt of €54m over 15 years is being provided by Emerging Africa Infrastructure Fund Limited (a fund within the orbit of the World Bank), the West African Development Bank (BOAD), the National Agricultural Development Bank of Mali (BNDA) and the Netherlands Development Finance Company (FMO).

Junior financing accounts for 10% of the total investment in the form of a mezzanine debt of €8m over 20 years, provided by Green Africa Power (GAP, fund for developing renewable energy in Africa funded by the British government).

In addition to these debts, there is a guarantee provided by GuarantCo which secured payments for a year (which means that over-financing in order to create a cash reserve in the project can be avoided).

All of the equity capital for the project is being provided by Akuo. The amount of equity capital has been calibrated based on a scenario that is significantly less prudent than that relied on by the lenders (with a 50% probability of completion). The high cost of equity remunerates this risk!

With the exception of Akuo, the financing is being provided public or multinational entities. As shown by the ICA’s figures it is still rare to see private financing in Africa, especially private debt financing. National or multinational development banks are still banks. Not only are they staffed by teams that are extremely competent in the field of project finance, but their aim is also to make a profit. We even note that as they do not have a commercial goal, they have become highly demanding when negotiating contracts (and in particular, PPAs).

*             *             *

Solar energy remains underdeveloped in Africa (with the exception of South Africa). Solar power plants can be constructed quickly (more quickly than thermal power plants), maintenance is simple and not very expensive, electricity is generated at a competitive cost, the size of plants can easily be adapted to requirements, etc. Akuo has proved that properly planned projects can be financed and completed.

On paper, nothing looks easy (complex geography, no major commercial banks in the region, etc.). This was in fact Akuo’s initial reaction when it committed to this project initiated by R20-Regions of Climate Action, an NGO founded by Arnold Schwarzenegger. So, Akuo started out by merely advising R20 and the Malian government on the feasibility of the project with a view to focussing more on development aid than commercial financing for this project. When the Malian government offered the concession to Akuo (through a direct auction as part of the emergency plan following signature of the peace agreements), this young French renewable energy company (which also happens to be the leading producer of renewable energy in France) said: “Why not?”. It is true that Akuo rarely shies away from projects in geographical areas that are, in theory, complex (Montenegro, Indonesia, etc.).

Since chart, compiled by Deutsche Bank and published by the Financial Times, is very instructive in many ways.

Firstly, it shows those who are not happy when dividends and share buybacks exceed capital expenditure, that this rule has never been observed since 1998, the start date of this chart, thus casting doubt on the validity of the rule. In our view, there is no reason to seek to limit to the amount of the former to that of the latter. Dividends and share buybacks are reflections of the immediate past, i.e., the profitability of investments made in the past. Capital expenditure, for their part, are proof of faith confidence in or a gamble on the future. There is therefore no reason to link them in this way.

Then, we can clearly see the effects of the Trump tax reform[1], which cleverly put an end to an aberrant tax system, eliminating the double tax regime for a foreign subsidiary (on profits repatriated to the United States) and its parent company (on dividend upstreams from its foreign subsidiaries). This system had led US groups to put off repatriating the earnings of US subsidiaries to their parent companies, which they then parked in tax havens (Bahamas, Ireland) pending this reform, which had been announced on numerous occasions but was never implemented, until one day in 2017 ... Also, a large portion of these funds was paid to shareholders in 2018, in the form of record increases in dividends and share buybacks, but which we won’t be seeing these levels again any time soon.

Then, we measure how much dividends could have fallen during the 2008 crisis (three times lower in quarterly amounts compared to the 2007 high), but share buybacks fell by even more (seven times lower in quarterly amounts), which is consistent with their even more discretionary nature compared with that of dividends.^[2] The former took eight years to get back to their historical high while the latter only reached it in the very particular context of the Trump reform.

Lastly, others will note the stagnation of investments since 2012, at $175bn per quarter, and will seek to explain this by the simultaneous rise in dividends and share buybacks. We don’t go along with this for four reasons.

The first is the reduction in the number of listed companies in the US and the continued rise in the number of unlisted companies financed by private equity funds, which makes it difficult to generalise across the whole economy.

The second is the growing share of the service economy, of which GAFAM is a striking illustration, to the detriment of the industrial sector, part of whose investments is recorded on the income statement under expenses (R & D, advertising, start-up losses) and no longer on the cash flow statement under capital expenditures.

The third is that investments related to manufacturing (see Apple and Foxconn) are often made outside of the US, by subcontractors in places such as China, Southeast Asia or Mexico, and do not appear in these statistics.

And the fourth is that we’re having a bit of trouble understanding why US business leaders, spurred on by investors who are vigilant, even activist, and whose compensation is performance-linked, wouldn’t realise all the value-creating investments that they can identify at a time when credit is cheap and easy to get and when most of them have low net debt to EBITDA ratios (1.2 for the median net debt/EBITDA ratio[3] of the S&P 100 in 2018).

With Simon Gueguen, Senior Lecturer at the University of Cergy-Pontoise

IPOs are a good deal for investors because of the discount on the offering price. This discount varies depending on the period and the market. In the 2010s it was around 5% in Europe and a little over 10% in the US[1]. Most of the time, demand for shares far exceeds the offer, and the issue of how shares are allocated among the various investors comes up. The article we look at this month[2] supports the idea of quid pro quo in the allocation process, with banks favouring investors from whom they receive the most in fees on other transactions.

Jenkinson et al rely on an original database for this study, created on the initiative of the Financial Conduct Authority (FCA) in 2014. It covers IPOs carried out by banks located in the UK between 2010 and 2014. In total, they studied 220 IPOs worth around $160bn (three-quarters of the European IPO market for this period).

The first result concerns offers made by investors. All other things being equal, investors who make offers at a limited price get more shares than those who make offers for large amounts (without a price limit). Such offers provide the bank with more information on the offering and the quid pro quo here is the success of the operation. This effect is, however, only visible for some of the banks in the sample. Other characteristics of offers assumed to provide information (early offers, offers revised during the course of the procedure, etc.) do not have a significant impact on the allocation of shares to investors.

We see a more spectacular result when we look at links between banks and investors. Investors in the top quartile of those who contribute the most to banks’ earnings, get (relative to their offer) 60% more than those in the bottom quartile. In the sample studied, total revenues of banks linked to investors participating in IPOs amount to nearly $40bn. Of this $40bn, most of the revenue is from brokerage fees. Fees on IPOs themselves account for less than $400m. This is why banks favour investors that bring them more in fees generally. It’s called quid pro quo!

This strategy is not necessarily implemented to the detriment of the issuer. It’s possible that this quid pro quo could be the partial reason for lower fees (although this remains to be proved). Whatever the case, the results of this study suggest that there is a key role to be played by the market authorities in controlling IPOs. MiDIF 2, which came into force in January 2018 (so after the period selected for this study) requires investment banks to justify their allocation procedure. It would be interesting to look into whether this greater transparency has any impact on the practice of quid pro quo.

If the valuation you get using the discounted cash flows (DCF) method is a pre-money valuation (i.e. before factoring in the capital increase), is the valuation you get using a discounted dividend model (DDM) automatically post-money?

Yes, the value calculated by discounting dividends is a post-money value of equity because the dividends that are discounted are those in the company’s business plan which assumes that investments will be made (as is the case with DCF) AND that financing will be granted. Accordingly, we get a post-money value.

^[1]  Like it here