It is hard to overstate the importance of segment reporting. The vast majority of listed companies that employ more than 1,000 people have more than one operating segment and much of the dialogue with investors revolves around these.
By definition, the profitability and prospects of the segments will vary over time, and understanding these trends is at the heart of much investment research. Unfortunately, the segment numbers are often much less robust than they seem and users are often unaware of these limitations.
The relevant standard, IFRS 8, Operating Segments, is relatively new and was, in my view, a huge improvement. The underlying aim is to align external reporting with the way the segments are managed, which is laudable, but this is also where the problems start.
Most companies now claim that there are synergies between their operating segments, not least to avoid the accusation that the head office serves no purpose. These ‘synergies’ are often actually shared costs and cross-subsidies, which are exceptionally hard to unpick, even for management.
Companies now claim that there are synergies between their operating segments, not least to avoid the accusation that the head office serves no purpose
Most larger companies have some form of matrix reporting to reflect operational complexity, whereas IFRS 8 in effect assumes that life is simple and one-dimensional.
For example, many truck companies also sell their engines for other purposes such as power generation and propelling small boats. The engines are all made in the same plant but the revenues are split by end market. Allocating R&D and production expenses across these segments is highly subjective. The profit reported by each segment will depend on private decisions about cost allocation.
Making and leasing
It gets more complex still when you look at companies with separate ‘making’ and leasing segments. The ‘making’ segment wants to book the profit immediately on sale, but the leasing segment has to spread the profits over the life of the lease, which typically requires a separate consolidation column. Who bears the residual value risk? Who is borrowing the money to fund the leasing?
Getting answers for questions like these is often impossible. In my experience, the ‘making’ profits are notional, the ‘leasing’ profits are provisional and the consolidation column will be largely opaque.
There is, though, an even bigger issue with segment reporting: corporate costs. It is extremely rare to find companies that report a separate number for corporate costs, even though this is permitted under the standard. I think that the real reason is embarrassment, especially for larger companies.
It is extremely rare to find companies that report a separate number for corporate costs. I think that the real reason is embarrassment
Corporate costs have a tendency to breed, like some form of progressive disease. Keeping a lid on them is really hard work, particularly for acquisitive companies. It’s much easier to sweep these things under the carpet by dumping them on the segments.
As an analyst, I occasionally got a true view of such costs when a segment was sold. The divested segment was often much more profitable than it seemed, and the legacy for the seller was one of my favourite bits of business-speak: stranded overheads, otherwise known as excessive costs. Large conglomerates are particularly susceptible to cost creep, which is one the many reasons that they are a dying breed.
The obvious temptation is to conclude that the standard was badly drafted, but for once I think this is unfair. Better disclosure for embedded leasing segments would be welcome but wouldn’t address the underlying issue of near-term subjectivity. And I cannot envisage any satisfactory way of separating corporate costs from genuine segment overheads.
I think it’s better to treat segment reporting as a starting point rather than a destination. Use the disclosure to have a better dialogue with management but don’t kid yourself that the numbers are especially representative.