Accsys Technologies — Arnhem expansion completed but Hull delayed

On 12 September, Accsys published a trading update, with two positive and two negative pieces of news. On the positive side, the fourth reactor in Arnhem is now operational and construction of the Accoya plant in America is on schedule to be operational by April/May 2024. The negatives are the larger than expected volume decline in 5M23 and continued delays with the construction of the first Tricoya plant in Hull, with Accsys in disagreement with its partners about financing.

Accsys states that Accoya wood sales remained flat year-on-year at €39m in 5M23. Due to the shutdown of all three reactors in Arnhem in April/May as a result of problems with the start-up of the fourth reactor, overall volumes dropped by 24% to 18,803m3 (considerably more than we expected). Multiple sales price increases over the past 12 months have fully compensated for this volume drop. On 1 May, Accsys also introduced an energy price premium due to volatile and higher gas prices (acetyl raw material costs are linked to the gas price), which continues to provide the company with a mechanism to pass on the effects of higher acetyl costs. The company also has a partial natural hedge through the resale of its by-product.

The fourth reactor in Arnhem increases total capacity by 33%, or by 20,000m3 to 80,000m3. Accsys has started the first commercial batches and will start ramping up production from now on. The goal remains for the fourth reactor to be at full capacity within two years (for comparison, the third reactor was at full capacity in nine months). If the demand for Accoya is as good as Accsys says it is, full capacity could be possible much sooner than the intended two years. Our model takes these two years into account, and at the current price level this provides a turnover of €160m. So, after around a four-month delay for this two-year project, the bigger Arnhem plant is now well underway.

Construction of the Accoya plant in America (with joint venture partner Eastman Chemical Corporation) is on schedule and will take two years from the start in April 2022. This timeframe includes the period of commissioning, thus commercial production could start in April/May 2024. On the positive side, Eastman is overseeing construction of the plant and has a lot of experience in building plants (Accsys has less experience). The construction contract has a fixed price structure, so cost overruns should not be as much of an issue as they have been in Arnhem and Hull. Initial capacity will be 43,000m3 and we estimate a potential revenue level of €86m at full capacity (at current price levels). At Kingsport, Tennessee, there is potential for total capacity of more than 160,000m3, offering ample room for further growth in forthcoming years in America, which is a big potential market for Accoya.

The first Tricoya plant in Hull is problematic. There is still no agreement with the partners on the financing of €7–10m (on 25 May, Accsys stated that it had sought an extension of the Natwest facility for €3m and, on 30 June, another €4–7m for cost overruns), while in its trading update the company said it is likely that project costs will be higher than the previous estimate of €94–103m on 30 June. According to Accsys, the consortium partners do not want to share the additional costs and apparently prefer not to increase Accsys's stake in the Tricoya UK construction. Accsys provided a bridge loan of €17m last year and has added an uncommitted €8m for these extra expenses (that were announced on 30 June) to allow commissioning of the plant to continue. It has reduced activity levels to cut costs and, according to Accsys, monthly run rate costs were reduced in recent months from around €4m to €1m in September (including the costs of subcontractors). There have also been problems with the software during commissioning, which has added to the delays. An external expert will investigate the current status of commissioning and the remaining work required to complete and the related costs which are still needed. Accsys will then appeal to the partners again, but it seems unlikely that they will want to share the costs.

Because of Accsys's current liquidity position, there is not much financial room for Accsys to fund new additional costs at Hull on its own. On 25 May, Accsys commented that a total investment of €57m will fall in FY23, being largely the finalisation of the fourth reactor in Arnhem (€8–9m), finalisation of the Hull plant (€15m) and the investment commitment of €28m for the Accoya US joint venture (which will be equity accounted, and therefore these investments are not part of the capex for property, plant and equipment), for which capital was raised in May 2021. More than 90% of the total investments for the company and for the US joint venture are expected to fall in H123. Taking these investments into account, together with the additional bridging loan facility of €8m, our calculations show narrow cash liquidity at the end of the first half. For the company’s cash liquidity, it is important to note that the bridging loan facility of €8m to Tricoya UK is uncommitted and Accsys is not under any obligation to lend the full amount and can withdraw it at any stage. According to the company, it currently does meet the covenants, with the most important one for the ABN AMRO facility being net debt/EBITDA, which should reduce over time from 4.0x initially to 2.5x.

Hence there is uncertainty about the amount of additional costs, the financing of these costs and therefore the plant’s further progress and pace of construction. Accsys is also looking at what impact the higher gas price will have on Hull's profitability, which it has always said could be EBITDA break-even on 40% capacity. Slower construction may spread the higher costs more evenly, but start-up will take longer. Accsys now says it does not expect Hull to start before year end, but does not doubt the start-up itself. It is looking at all possible options, including changing the structure of the consortium partners or self-funding Hull from Arnhem's incremental cash flows over a longer period. There are currently no plans for another share issue as Accsys raised €19m (net) in May 2022. Hull is planned to have capacity of 40,000m3 with potential revenue of €50m to be realised within a three-year ramp-up period.

We have reduced our estimates after the lower than expected 5M23 revenue level and further delays in Hull. There is also uncertainty about when the plant in Hull will start up. We have reduced our FY23 revenue estimate by 10% and our EBITDA estimate by 20% (larger decline due to operating leverage and the impact of higher input prices) but, as we expect Hull to come on stream in the next financial year and Arnhem being operational for the full year, the adjustments are less pronounced for FY24 (-3% for revenues and -14% for EBITDA). These lower estimates reflect revenue growth of 15% in FY23 and 32% in FY24, driven by additional capacity and pricing, while EBITDA margins are expected to improve, driven by scale benefits and pricing.

Despite the short-term uncertainty, there is long-term potential, with the fourth reactor in Arnhem adding 20,000m3 of capacity from now, the Hull plant adding 40,000m3 potentially from April 2023 and the Accoya plant in America adding 43,000m3 from April/May 2024, bringing total capacity to 160,000m3. With four full reactors in Arnhem, one in Hull and two in America, potential revenues on a fully consolidated basis are €296m versus the €121m reported in FY22.

We use a discounted cash flow (DCF) model as there are no other listed companies with a business profile close to Accsys’s. The company is trading at 1.9x EV/sales and 16.4x EV/EBITDA in FY23e.

Our DCF model is based on four reactors in the Arnhem plant and one in Hull. We use a DCF to value separate Accoya and Tricoya plants and add a value for the Accoya plant in the US, which is expected to be operational in mid-2024. We also add a value for the Tricoya plant in Malaysia (or an alternative location) with a probability rate of only 50% as the final investment decision on that project has not yet been taken. An investment decision will be taken after the Tricoya plant in Hull has been operational for at least six months, so timing is currently uncertain but it will be delayed until at least Q3/Q423.

Most of our assumptions for the DCF model are unchanged except for the equity risk premium, which we increase from 5% to 6% to reflect the short-term uncertainties. This lifts our WACC from 8.9% to 10.0%. On our lower estimates and the increased risk profile, our DCF indicates a value per share of €1.21 per share (previously €1.83).