Do Your Own Research

Insightful analyst, fund manager and accountant lan Robertson shares his tips to help investors understand what questions to ask when making investment decisions.

<< Back to Do Your Own Research archive

April 9, 2024

Try to stay awake – Seeing Machines and questions on accounting

Our earlier blog focused on how IAS38 can be bit daft from an accounting and auditing standpoint.

In this blog, I look at the issue of capitalisation of development costs from the viewpoint of an investor or analyst, using the example of Seeing Machines.

Capitalising development costs can provide reported figures that are prima facie more attractive than the simple full expensing path, and more accurate in the minds of management.

However, as we shall see, in search of insight and consistency, informed and diligent investors may well unwind all the IAS38 capitalisation and amortisation on their way to making an investment decision.

And as we will see, on their way investors may find themselves asking ‘why’ quite a lot. That is rarely a good thing.

This is certainly not meant as a full investment analysis of the company; there is plenty for investors to consider other than the accounting.

seeing machines ebitda paths

Seeing Machines – should be simple, but it’s not

There are plenty of UK-listed companies that could be used to illustrate the analysis of capitalised development costs. But I chose Seeing Machines because the board has its fair share of accountants and the end markets work around long-term product programs, so there should be reasonable visibility on eventual volumes – albeit it through the fog of the vehicle manufacturer customer’s innate optimism and Seeing Machine’s Tier 2 supplier status.

Seeing Machines has been developing technologies for vehicle driver monitoring and awareness for several decades and has been on AIM since 2005. It has not really got involved with the profitability thing.   It has actively capitalised development costs over this period in accordance with the relevant accounting standard, IAS 38. And as we have considered elsewhere, there is some ‘margin’ regarding how it gets applied, and one management can be hostage to its predecessors’ actions.

The considerable regulation-driven growth expected at Seeing Machines in coming years, notably from EU General Safety Regulation, may the grab the investor or analyst’s initial attention. But what matters to investors is the path to profitability and cash generation.

Development costs need not be such an issue

In most cases where there is capitalisation of development costs, investors, even professional ones, will look at the capitalisation and amortisation figures. If these are not too far apart and not significant in the scale of revenues or profits, the investor will usually ‘take a view’ and move on from the issue.

In FY23, Seeing Machines capitalised US$23.7m of development costs and expensed US$11.3m, of which US$2.4m related to amortisation of development costs, set against US$57.8m of revenue.  No sensible investor is going to ‘take a view’ here, when the difference the accounting policy makes is to reduce EBITDA losses by 70% and the amount capitalised is over a third of revenue.  They will add the capitalised development spend to the operating costs and then, if relevant, adjust out any impact of the amortisation in the profits figure.

Need to look at the relevant cycle – economic, product, management, or hype

Quite how far an investor should go back in time to analyse a company naturally varies from company to company.  Following the pandemic, there is a need to revisit the ‘normal’ in the years prior to 2020. And, even with tech companies, we must look at what happens over a cycle – be that economic, product, management, or hype.

Seeing Machines – starts off nice enough

For much of its first decade or so as a public company, Seeing Machines worked away building its technology base and credibility, and it gained traction.  Contract and development announcements began to flow.

seeing machines share price chart

2018 would seem to be a good place to start looking at the numbers in detail. The share price chart would certainly suggest that things were getting interesting.

At that time, Seeing Machines had secured production awards with two premium German automotive OEMs and one with a global automotive OEM headquartered in North America. Shortly after the June year-end, there was an announcement regarding a Chinese OEM.

In FY18, Seeing Machines was already capitalising some, but not much, of its development spend, with a policy of amortising this over three to five years. It expensed US$15.7m, significantly more that the US$0.1m it capitalised, and with that expense there was US$1.2m amortisation relating to development spend capitalised on products being sold. None of this could be said to be unreasonable.

Unfortunately, not everything came to pass.  It is automotive and technology after all. But thankfully most wheels remained on the still-unprofitable wagon.

In FY19, Seeing Machines posted R&D expenses of US$25.7m and capitalised US$0.1m.  At this stage, it had seven OEM production programs across six automotive OEMs, expected to deliver more than AUD$150m (US$108m), mainly over the 2021 to 2024 period. This was a waiting game year, but clearly one where management felt justified in ramping the R&D spend.

No one changes accounting policy without reason – more distractions

In FY20, Seeing Machines reported a figure of US$20.8m for R&D expenses, and with a revised amortisation policy of 5-7 years it ended the year having fully amortised its historical costs. It didn’t capitalise a cent.

With increasing references in the announcements to lifecycles and programs, the extension of the amortisation period doesn’t look out of place. But it still begs the question, why?

In the FY21 results, R&D expenses had fallen to US$7.4m after capitalisation of US$6.2m of development costs. There was also a reclassification of the costs, with the FY20 R&D expense now being shown as US$16.6m – a movement of US$4.2m or 20%. We have no insight into quite what the FY19 reclassified figure would have been. FY21 was a good year for ‘whys’.

seeing machines development capitalisation

In FY22, Seeing Machines capitalised US$18.6m of development costs, with US$11.3m expensed to the income statement, of which just US$0.8m was the amortisation of the asset.

This year, the policy had changed to 7-10 years. Ten years is a long time in technology. It’s also a long time in automotive when faced with the transition to electric at the end of the decade. Fewer ‘whys’ than FY21, but bigger ones.

In FY23, Seeing Machines moved to reporting in US dollars rather than AUS dollars; a reasonable move given the dominance of US customers, but a pain for the investor who has to recalculate everything in US dollars. The figures here (FY17-FY21) are recalculated on average rates for the respective years.

Underlying profitability and cash generation is not going to be achieved any sooner by changing the reporting currency.

Seeing Machines capitalised US$23.7m of development costs in FY23 and expensed US$11.3m, of which US$2.4m related to amortisation of development costs.  So, as it stands today, it doesn’t look like amortisation is going to equal capitalisation anytime soon.  In fact, even if Seeing Machines held its level of development cost capitalisation constant, investors would not be seeing amortisation broadly level with capitalisation until the end of the decade.  That’s a long time to be putting off investors.

Of course, Seeing Machines could see the two match sooner if the investment / capitalisation were to slow.  As a fund manager, and as an analyst investing in and covering the successful UK-listed IP companies, two salient factors were (1) that higher licence payments meant lower royalties and (2) that maintained or increasing R&D was required to defend market position and long-term royalty rates.

This capitalisation policy isn’t providing insight into either the cashflow or the income statement. Furthermore, management’s policy of including capitalised development costs within the R&D spend it shows within the operating costs breakdown in its presentation only adds to the muddle.  Where do they think it should go?

seeing machines opex and capex mix

The only way to get a clear and consistent picture is to restate without capitalisation of development costs, as shown in the table below.

Everything that Seeing Machines has done appears justifiable and consistent with IAS 38, but it is not necessarily helpful to the investor. Things would be so much clearer to the investor if, in line with US GAAP, the company had just expensed development spend.

Seeing machines development capitalisation impact

As can be seen above, the resulting numbers might not be pretty, with EBITDA loss increased by 231% for FY23, but the path to profitability has to be easier to explain and understand than the mystery tour described above.

Even though taking the US GAAP path might, using the house broker’s forecasts (source: Refinitiv) and FY23’s capitalisation figure, delay EBITDA breakeven to FY26 and PBT breakeven possibly to FY27, everyone is well aware that capitalising development costs (and repeatedly adjusting policies) does not help generate cash or underlying profits.

What it does do is generate questions. And the more questions investors have, the less likely they are to invest.

Ian Robertson

irobertson@progressive-research.com

07768 276595

This communication is provided for information purposes only, and is not a solicitation or inducement to buy, sell, subscribe, or underwrite securities or units. Investors should seek advice from an Independent Financial Adviser or regulated stockbroker before making any investment decisions. Progressive Equity Research Ltd (“PERL”) does not make investment recommendations.

Opinions contained in this communication represent those of PERL and/or our affiliates at the time of publication and PERL does not undertake to provide updates to any opinions or views expressed. PERL does not hold any positions in the securities mentioned in this communication, however, PERL’s directors, officers, employees, contractors and affiliates may hold a position,  and/or may perform services or solicit business from, any of the companies or related securities mentioned.

Any prices quoted in our research are as at the previous day’s close.