Financing Offshore Wind – Part 4

This is the fourth part of what will be a 10–part series of blog posts, which will ultimately be published in full as a single report. Two parts will be published each week for the coming weeks.

Earlier instalments:

Part 1   Part 2   Part 3

BARD1

BARD1 is a project that is partly forgotten by the offshore wind community, as it is an orphan project in many ways, with a bankrupt developer, a defunct turbine and a secretive bilateral funding, even though it was the first 400 MW project to actually get built in German waters.

Its developer, BARD, was one of the many smaller players that participated to the first generation of projects in Germany, but it was unusual in that it was a fully integrated operation, including design and manufacturing of its own turbine, its own foundation concept (a tri–pile structure which was meant to be serial–produced), its own installation vessel, as well as project development. Founded by an immigrant Russian engineer who had played a prominent role in the Russian gas industry (heading Gazpromstroy, the construction arm of the gas giant), Arnold Bekker, it tried to do everything in–house and almost succeeded.

The distinctive BARD tripile, in the Cuxhaven harbor, before installation (photo by author)

The project got lucky in that it received a EUR 200 million bridge financing from Unicredit’s capital markets team in the summer of 2008, just before the financial crisis, which closed financial markets and any hope of the “real” financing to be raised on the public markets (which had quite different requirements than project finance lenders) and the bank was then left with the unpalatable decision to either write off the loan or continue to fund the project on its own to the end, requiring at least another billion euros. It did decide to continue with the project (at the time it still had credible buyers for it, once it was built) but unfortunately the project’s installation vessel suffered from delays in construction and then design defects that prevented it from fulfilling its job. Construction was delayed by over 2 years, and cost overruns, while never fully made public, probably reached above EUR 1 billion, largely borne by Unicredit.

Active construction of the BARD1 project, June 2013 (photo by author)

The saga saw Unicredit take over the projects as well as the company, completing the project, selling the other BARD projects (including what became Gemini in the Netherlands and Veja Mate in Germany) after multiple complications.

After the death of the founder and the delayed completion of the project (at the time, the first of the multiple 400 MW utility–scale offshore wind projects in Germany to actually get built, beating to the punch the largest utilities), it still suffered many operational difficulties, including serial defects with the turbines, harmonics issues with the grid, while having to manage orphaned technology (the turbine was not used on any other project), suppliers no longer committed to the (defunct) manufacturer, and the perception in the market that it was a troubled project. A dedicated project team managed to solve most technical issues after several years of effort, and finally bring the project to almost 100% technical availability, allowing the project to be sold as a viable concern to and finally selling it in 2019 to Green Investment Group (the Macquarie–owned successor to the UK Green Investment Bank), an experienced offshore wind player.

Global Tech 1

Global Tech was the first 400 MW project in Germany to receive standard non recourse construction finance (it was the second project in Germany to be so financed, after the 200 MW Borkum West II a few months earlier), and the first, alongside Meerwind, (288 MW, project–financed almost at the same time) to benefit from the public KfW offshore wind financing programme. Promoted by another colourful pioneer of the offshore wind sector, Willi Batz– run Windreich, it managed to bring together the largest bank consortium ever put together for what was then the largest financing of an offshore wind project. Together with Borkum West and Meerwind, it heralded the arrival of Germany in the offshore wind sector and the industrialization of the sector.

Unfortunately, these projects, like the others built at that time by utilities (Riffgat, Dan Tysk, Nordsee Ost), had to face a major obstacle: delays in the construction of the grid connections to shore. In Germany, the choice was made to allocate all grid connection construction and ownership to the grid operator, and to coordinate that via offshore wind hubs – massive sub–stations that would collect power from several offshore wind farms and then transport the power to shore via high voltage direct current lines (HVDC). That made sense given that projects in Germany need to be quite far from shore and this would avoid building too many parallel cables to shores through environmentally‑sensitive coastal areas. The problem was that not a lot of HVDC lines and transformer stations had been built in the past, and TenneT, the grid operator, was suddenly obligated to build several almost at the same time. The technical challenges of that task lead to severe delays, and forced projects to delay construction (or, if they had started, to find costly ways to maintain their installed turbines with no power available on site). Global Tech I, like Meerwind and Borkum West, was a victim of that situation, and the contingency budgets were stretched beyond breaking point. A combination of complex financial engineering and limited additional support from the sponsors was required to get the projects through. There was an implicit commitment by the grid operator to pay penalties in case of delays, but this situation forced the government to make that commitment explicit and provide a detailed formula to calculate the exact amount – this was put into law in early 2013, after an accelerated process.

Tasks that had been scheduled at certain dates could not be performed, and vessels that had been reserved for such tasks could not be used at the time –  substitutes needed to be found at the later dates when the tasks could actually be performed. Managing these delays, and the full reorganization of the construction schedule, was a full-time job for the project and its lenders. All projects became fully operational at least a year beyond the scheduled date, and definitely beyond the expected date of maximum tolerable delay.

Banks could have decided that this was the sign that offshore wind was too risky, and stopped lending to the sector, or at least made terms a lot more conservative, but actually, the lessons the banks drew from this episode was that problems in offshore wind could be resolved, and what mattered was to have good project management teams, who were proactive and transparent, willing to listen to past experience and learn the lessons from previous episodes. The philosophy of lenders could be summarized as “don’t tell us you don’t have problems (we won’t believe you or trust you), show us how you solved the problems you had”, and Global Tech was one of the core projects where that experience was built.

The project became fully operational in 2015 but was only declared complete by the banks (which allowed dividends to be paid to owners) in 2018. It was successfully refinanced in 2020, after having had to deal with the fact that its turbines had also become de facto ”orphaned” (Owner Areva ceased activities and was dismantled, with operations and maintenance for the small existing fleet transferred to a Siemens entity).

Lincs

In parallel to the negotiations for the first financing of Global Tech I, Centrica and DONG (later renamed Ørsted) were also trying to raise financing for their 270 MW project in the UK. That financing effort, started in 2009, only reached a conclusion in 2012, when construction was already well under way, after painful negotiations between the investors (two utilities) and their project finance lenders. At the heart of the conflict between the two was how to manage construction risk – utilities wanted to do it their way, with minimum interference, but asking banks to bear some of the risk, while lenders wanted more scrutiny and more say on how the contracts were structured. The very different approaches to construction risk were never really reconciled and left a bitter taste to all, with London‑based bankers and investors saying publicly and wrongly, for several years, that construction risk could not be financed even as their continental colleagues were doing multiple transactions where it was. The “curse” was finally lifted when the Galloper project was financed on a non–recourse basis, with construction risk, in late 2015

The two different approaches to allocating construction risk both work, but they are largely incompatible and the decision to go for one or another must be taken quite early on in the development of the project, as it influences the construction strategy and the negotiation of all corresponding contracts.

Which takes us back to how projects are financed, which will be Part 5 of this series.