Taiwan’s 8MW pilot offshore wind farm Formosa 1 Phase 1 was commissioned in 2017. Two years later, 5.5GW worth of offshore wind farms have been awarded to the world’s most experienced offshore wind developers, and the island’s first large-scale offshore wind farm reached financial close on June 8, 2018.

This project – Formosa One Phase Two – has been described as a pathfinder for two reasons. Firstly, it is a rare example of conventional project finance in the region. What’s more, the development of offshore wind is still at a nascent stage in Asia. With the exception of China, it is yet to take off is lacking an established regional supply chain.

The path ahead for the remaining capacity may not be a straight one. Months after financial close, the government slashed proposed feed-in tariffs (FiT) for 2019, putting much of the remaining 5.5GW pipeline at risk and forcing one developer, Orsted, to halt all project development.

Can the debt financing structure used for Formosa 1 Phase 2 still provide a template for investors active in the sector?

In the Taiwanese market lenders will need ECA cover. Future transactions are likely to need 50% cover or more if they are to go through

The answer lies in the lender mix and Taiwan’s approach to construction and operational tariff payments. 

In November 2017, local project sponsor Swancor approached Taiwanese banks for a 16-year syndicated loan to finance Formosa 1’s second phase. The following month, the project secured a 20-year Power Purchasing Agreement (PPA) with state utility Taipower at at FiT rate of NTD 5.849/kWh. By March 2018 it became clear that there would be heavy participation from foreign banks.

With a project cost totaling USD 795m, a consortium of seven international and four local banks eventually gathered to provide a New Taiwanese Dollar-denominated USD 596m loan, based on a debt-to-equity ratio of 75:25. In order to placate concerns from apprehensive lenders, Danish Export Credit Agency EKF is covering 60% of the debt.

Having a diverse pool of lenders is proving to be complementary. Without a doubt, international lenders to offshore wind projects in Europe bring with them rich industry experience. But they often lack a deep understanding of the local regulatory and business environment, according to a Taiwanese banking source that spoke to Inframation for this article. That is a gap their local counterparts are able to fill, while sharing their experience with Taiwanese IPP and solar projects that were developed under PPAs.

With Formosa 1’s financing finalised in June 2018 and project construction beginning this year, all eyes are keenly monitoring the structure of subsequent projects in Taiwan’s 5.5GW awarded capacity allocation and how the economics of the reduced feed-in tariff structure plays out.

ECA cover will be integral to the financing of offshore wind projects in the next two to four years, says Jørgen Kragh, head of team offshore and project finance at EKF: “There is a general understanding in the Taiwanese market that lenders will need ECA cover. Future transactions are likely to need 50% cover or more if they are to go through”.

Given that the capacity of a single ECA will not be enough for cover on the wind farms lined up, Kragh expects to see two or even three ECAs involved in the same transaction – and indeed, as reported by Inframation’s sister publication Sparkspread, the 360MW first phase of WPD’s Yunlin wind farm is set to be 50% covered by EKF, Hermes, and potentially Atradius. This assertion is corroborated by the source at a Taiwanese bank, who adds that ECA cover increases the chances of gaining internal approval to lend into an offshore wind farm, particularly for local banks.

One element of Formosa 1 that is unlikely to differ in subsequent projects is the 75:25 debt-to-equity ratio, according to the same source, who adds that the involvement of state-owned banks, such as the Bank of Taiwan, will be pivotal in the financing of future projects by providing much-needed liquidity.

This point is echoed by Quentin Slight, head of energy and infrastructure for Asia at Crédit Agricole. Slight adds that in Taiwan, the involvement of state banks that remain apprehensive about offshore wind would positively “affect the confidence of both local and international banks”. Regarding debt margin pricing, Slight does not expect subsequent projects to see much of a premium over the one found in F1P2, which was reported to go as low as 125bps.

Regarding tenor length, most international banks would prefer shorter term financing than the 16-year tenor found in Formosa 1 Phase 2 – mainly due to the short-term hedging products available and banking regulations that constrain the lending period. It is however important to keep in mind the two-tiered system option offered to developers, which essentially provides for a high feed-in-tariff during the PPA’s first 10 years and a lower one during the latter 10. This option, which is the one picked by for both Formosa 2 and Yunlin, can go some way in mitigating the long tenor issue as loans will likely be paid off earlier due to higher returns in the first 10 years.

We have seen some sponsors sell off part of their stake to outside investors, much to the chagrin of the Taiwanese government who had hoped they would maintain their stake in the long term. Macquarie and Swancor have already divested part of their stake in Formosa to Japanese utility JERA, while WPD is looking to sell a 49% stake in Yunlin. A developer source told this publication that Japanese and Korean developers and investors will “seek opportunities in owning partial equity in Taiwanese offshore wind projects”. As they would be regional players, this could be done with a view to create “synergy in the supply chain and learn lessons from the early projects [in Taiwan]”.

[Japanese and Korean developers and investors] will seek opportunities in owning partial equity in Taiwanese offshore wind projects

Potential new lenders are also waiting in the wings. There has been much talk about Taiwanese LifeCos, which are flush with liquidity, potentially lending into later projects. However, their regulations dictate that they are not eligible to take on construction risk, and they are thus essentially excluded from lending into the first stage of the projects’ financing. They should not however be restricted from getting involved in any subsequent refinancing, according to Slight. 

It is likely, however, that we will see more international banks come into the fold.

In Europe’s offshore wind sector alone, upwards of 50 banks have lent into various offshore wind projects, which indicates the significantly higher level of competition there. A barrier for international banks such as ABN Amro, HSH, Ipex, NORD/LB and Norinchukin that do not have local branches is that this disqualifies them from being able to lend in NTD. A solution would be to enter into a risk participation agreement with a local bank, by which the international bank can cover a loan provided by its local partner in NTD. As put by the local banking source: “We have the New Taiwanese Dollars and they have the credit appetite”.

Without a hedge, foreign exchange risk exists only during the construction period for ECAs and lenders, because that is when we see contracts in USD and EUR, according to EKF’s Kragh. Following construction, NTD are used. When it comes to interest rate risk, a hedge needs to be provided for in the longer term due to loan tenors falling between 15-20 years.

The severe losses sustained by the governing party in Taiwan’s district elections last November created doubts in some quarters regarding policy continuity in the island’s offshore wind industry, particularly following the government’s proposal to set the FiT rate for 2019 at NTD 5.106/kWh, down 12% from the 2018 rate of NTD 5.849/kWh – just days after the election. This was compounded by the frustration exhibited by developers, most notably Orsted, after they said they were unable to sign a PPA for their projects under the reviewed FiT. 

The government eventually halved the proposed cut to a 6% drop, setting the final FiT for 2019 at NTD 5.516/kWh. With the island’s 2020 general elections on the horizon, the Taiwanese banking source indicates that political uncertainty may still influence offshore wind projects. Should the governing DPP party lose, “there will be an adjustment period. The new government may re-examine Taiwan’s power supply and demand market and may not be as aggressive in pursuing renewable energy. We might well have a period of turmoil for offshore wind.”

Credit Agricole’s Slight is less concerned, given that even with a change in government, “there is still a real political incentive to sustain offshore wind. The government will still want to create the jobs resulting from these projects and position Taiwan as a supply chain hub for the rest of Asia”.

He instead points to a potentially greater risk: regulations on local content. While these may not apply to fast-track projects like Formosa 2 and Yunlin, subsequent projects will have to deal with higher local content requirements. Another potential risk is the development of a comprehensive grid system that will facilitate supply from the wind farms. This is not yet in place and will need to be constructed by Taipower.

Finally, perhaps one of the most obvious risks on an island that is routinely plagued by earthquakes and typhoons is the potential for a natural disaster. There is preparation taking place for this – the wind farms are designed in such a way that they can face dire natural phenomena, and they are of course insured. Nevertheless, if a natural event were to severely damage the wind farms, insurance costs for the projects would soar.

Aside from the existing 5.5GW pipeline, a third batch of offshore wind farms was expected to be announced this year. This has now been delayed while the government evaluates key milestones in the first two phases. Even so, the industry is set to continue attracting attention as projects line up to arrange financing, with Formosa 2 and Yunlin next in line for financial close.