Carillion has entered the pantheon of cursed companies following its recent failure. Politicians and the media have worked themselves in another fit of righteous indignation about greedy management and incompetent boards. The search is out for people to blame, shame and even prosecute. Regulators, sensing the flow of the political wind, are climbing on the bandwagon and looking for blood.
Non-executive directors are reasonably enough feeling nervous. They are at the centre of corporate crises such as this. Already some Carillion directors have had to resign from other roles and the non-execs will be contemplating an enormous black mark on their CV’s. Being a non-exec in this form of collapse can be terminal for any career.
Do your due diligence
One lesson for non-executives is to do careful due diligence before you contemplate joining a board. Another lesson is that non-executives need to be sceptical of what they are told around a board room table, applying due diligence principles to what they see and read.
Most candidates take the financial health of a company, especially a well-known or large one, for granted. This is a dangerous assumption. Two companies that I joined as a non-exec proved to have potentially fatal toxic derivatives. Carillion recruited two new non-execs onto its Audit Committee in the couple of months before going bust.
Read the Annual Report
Although full of guff and regulatory noise, the Annual Report generally should tip you off to issues. In my experience with the toxic derivatives, they were indeed listed in the Annual Reports, but were so surrounded by reassuring jargon that it took several interrogations of the CFOs to confirm their real nature. In fact, the words in the Annual Report are very unlikely to warn you. Despite the regulators attempts to ensure that reports are balanced and fair, they are still largely promotional documents.
I am not attempting here a technical analysis of the Carillion Annual Report. Nor do I wish to criticise the directors (there are many others only too keen to do that). My objective here is to use 20:20 hindsight to look at whether there were clues even in the 2016 Annual Report that could at least have raised questions in non-execs minds, and to offer these as lessons for other non-execs doing due diligence.
Virtually all the text is confident and devoid of doubt, as is pretty standard for most annual reports. However, amongst all the good news, there was one clue: “In 2016, we made good progress in a number of our markets, while managing and mitigating the effects of more difficult trading conditions in others.” Given companies natural reluctance to air their problems, a reader should highlight any cautionary statement like this. The Board will have had a very good reason to include such a comment.
Overall however, you wouldn’t have got much balanced information from the words. Most of the clues lie in the numbers, particularly those that are not discussed in the text.
There had been two years of decline in ‘secured and profitable’ orders from £18.6bn to £16.0bn, which possibly was an early sign of slowing growth. Of this £12.2bn was support services. Given that support services annual revenue was £2.7bn, the ‘order book’ looks as if it is adding up all future contracted service revenues. This is therefore not as impressive as it looks. Furthermore, the reader should ask if this future revenue is actually all unconditional or subject to performance conditions.
Underlying operating margin had fallen for two successive years from 5.6% to 4.9%. This should have been of concern. However, given the revenue growth, the business recorded two years of underlying EPS growth from 33.7p to 35.3p. The question therefore is whether Carillion was buying less profitable business, just to keep its growth going.
Long term contracts can be inherently volatile. Costs are generally booked as incurred and revenue – and hence profit – are, broadly speaking, booked on a pro rata staged basis. This is calculated using % contract costs incurred so far against forecast total contract costs times the full contract revenue. If individual contract costs run to plan, there is little problem, but if costs start to escalate, then there can be a big swing from pro-rata profit to whole contract loss. There isn’t much discussion of this in the Annual Report, other than noting the accounting policy and that the Audit Committee had reviewed the accounting and found it ‘reasonable’. There is no way of knowing whether the Committee deliberately used such a mild word to suggest that they were not very enthusiastic about the accounting or whether this was a synonym for ‘true and fair’. Either way, a reader might have queried it. For all the length of the new external auditor’s report, it simply lists all the work they did, without drawing a specific conclusion on this issue.
There was a big increase in non-operating costs from £5m to £40m, excluded from ‘underlying profit’. These were largely redundancy and closure costs. Costs excluded from underlying are always of interest.
Net debt and cash flow
On the one hand, the company seemed happy with its cash flow and debt;
“Cash flow from operations represented 117 per cent of profit from operations.”
“The Group continues to have substantial liquidity with some £1.5 billion of available funding…The vast majority of the Group’s £1.5 billion of funding matures in November 2020 and beyond.”
On the other hand, there is a clue that they are not as happy as they appear to be;
“We will also begin reducing average net borrowing by stepping up our ongoing cost reduction programmes and our focus on managing working capital.”
The year-end net debt ratio to EBITDAR rose from 0.6x to 0.8x. But average net debt was 2.7x bigger than the year-end debt, implying an average net debt to EBITDAR of 2.2x, a very different picture. There is no explanation in the annual report why year-end net debt is so much lower than the average. Maybe there was a consistent big seasonal swing, but this seems unlikely in this industry. In its absence, a sceptical reader might draw the conclusion that the year-end balance sheet was being managed aggressively.
There is evidence of working capital issues as the construction contracts receivables increased by 60% to £615m, perhaps indicative of cash flow problems with such contracts. Other receivables and prepayments also rose by 36% to £750m. Finally, Carillion was having a little trouble in getting paid, with trade receivables that were over 3 months old more than doubled from £25m to £55m. Carillion had managed its working capital by lengthening many payment terms to 120 days. As a result, its trade payables were rising and suppliers were being encouraged to use an Early Payment Facility, whereby they would borrow against their receivable from Carillion to keep their own business going despite such slow payments.
The group’s net debt was growing. Both year-end and average net debt rose by just under £50m, although £68m was due to foreign exchange offset by £34m from selling shares in PPP joint ventures. The report was a little dismissive of the foreign exchange loss, half of which was ascribed to a US dollar private placement debt. This latter was apparently fully hedged, but it is not clear where the corresponding hedging benefit appeared.
However, the group claimed to have ‘a strong funding position’. £1.4bn of funding was available at Dec 2016 (of which £0.7m was undrawn), plus additional funding then secured the following month, gave a total of £1.5bn. The only mention of any debt covenants in the report is in the going concern section, where the board confirms ‘comfort that funding covenants will continue to be met’. In retrospect, a reader might have wished for the details of those covenants.
‘The Group therefore continues to have substantial funding…over the medium term’, with only £85m of facilities maturing in 2017. When Carillion went bust, with a reported only £29m in cash, you would have assumed that it had at least £1.4m in debt facilities. It looks as if failure to meet the undisclosed covenant conditions caused facilities to be withdrawn, triggering the eventual liquidation.
Net promoter score (a measure of customer satisfaction) fell dramatically from +36 to +22, which the company ascribed to ‘challenges of mobilising new contracts’, although there was no evidence that new contracts were growing in number. Anecdotal reports suggested that there was a growing unhappiness with customers and a few high-profile disputes had surfaced.
“…the wider outlook for volumes and margins across the (Middle East) region is expected to remain challenging…”
Middle East construction services revenue grew by 19% to £428m, but at a tiny net margin of 1.9%. The commentary recognises the low margin, but point to a further £15bn of contract opportunities, without discussing whether these would be at a higher margin. The segmental analysis reveals year-end net assets employed of £206m, which with an operating profit of £8m, suggest a return on capital of only 3.9%. This assumes of course that the year-end net assets are representative, yet we know that the group’s average net debt is 2.7x higher, so the real return on capital was likely to be lower. It wouldn’t have taken much of a cost overrun for this return to be wiped out. There is no discussion of segmental capital returns in the Annual Report.
Other construction services also grew fast by 21% but at a net margin of only 2.1%, down from 2.9% the previous year. A margin target margin is given here as 2.5% to 3.0%. The Company is able to claim that it satisfies this target by adding in joint ventures where the £9.0m profit share exceeds its £8.9m revenue share. There is nothing like a 100% net margin joint venture to improve reported performance. Mention is made of ‘managing risks in order to deliver our target margins and cash flows’, although there is no further discussion of cash flows. The segmental reporting reveals year-end net assets of £190m. With a profit of £32m, this suggests a decent return on capital of 17%, although this is probably flattered by year-end working capital being below the average.
Recent acquisitions were underperforming. There isn’t any reference to this in the commentary, but buried in the financial review is £15.6m of non-recurring credit. This would seem to be good news, but is in fact a reduction in consideration to be paid for two acquisitions whose EBITDA is ‘lower than the stretching targets agreed’. ‘Nevertheless, these businesses have performed well…’. The report does not define what it means by ‘well’.
The pension scheme
£47m cash pa was being put into the defined benefit pension scheme, whose IAS9 deficit had ballooned to £805m from £394m the previous year. The deficit payments look light, being based on a 15-year recovery period, as a rule of thumb is more like a 10-year period. This was a warning sign that cash payments would be likely to rise significantly at the next triennial valuation (due in 2017). The absurdity of pension accounting rules is that the P&L was actually being charged only £6m (the rest effectively goes through reserves). Unlike many companies that have closed their increasingly expensive defined benefit pension schemes, Carillion kept theirs open for employees when necessary ‘to meet the requirements of work winning’. A reader might ask whether the costs and risks of this were fully built into those contracts.
Goodwill & deferred tax
Carillion had a lot of goodwill on its balance sheet – £1.6bn compared to £0.1bn of fixed assets. The rules say that goodwill has to be justified on whether it can be backed by prospective profit flows. This is a potential double whammy. The reader should note that if performance falters, large potential write-offs in goodwill become necessary, exacerbating the financial pain.
A similar problem exists with deferred tax assets, which were £164m. These can only exist where you are forecasting sufficient forward profits to justify them. Poor performance can lead to these being run down quickly.
Carillion appears to have been exemplary in its corporate governance, as described in the annual report. The board fully complied with the governance code. Employee engagement rose from 68% to 73%; employee volunteering increased from 18% to 30%; and the gender balance improved, with 38% of employees being female. The board had an external review of its effectiveness and 29% of the board were female. The previous Remuneration Committee was approved by 80.6% of the votes.
Regulators might note this. If Carillion obeyed all the rules, then maybe the rules are not really the be all and end all to company performance after all. Perhaps Regulators could focus on understanding why companies fail, rather than grandstanding the latest governance flavour of the month.
The Annual Report boasts that; ‘Rigorous risk management processes that identify, manage and mitigate risk are fundamental to the success of our centralised operating model.’
However, this rigorous process did not identify liquidity as one of the top ten Group risks (although the pension liability was regarded as high impact/high risk). The top risk was winning new work, but it didn’t mention the risk that new work might be loss-making or cash consuming.
The new requirement for a Viability Statement was intended to highlight precisely such looming cash crises as Carillion, but stakeholders were reassured;
“On the basis of both reasonably probable and more extreme downside scenarios, the Directors believe that they have a reasonable expectation that the Company will be able to continue in operation and meet its liabilities as they fall due over the three-year period of their assessment.”
In fact, there was insufficient information in the Annual Report to give a reader comfort on liquidity and viability. Perhaps the regulators need to study this again.
There are many cautionary lessons here for a non-exec in using an annual report as part of their due diligence on any company;
- Pay limited attention to the words, except where there is any hint of caution or bad news. Accept that boards will naturally enough always accentuate the positive in writing about their business, but apply scepticism yourself.
- Always question whether numbers are unconditional. Future revenue and pipeline numbers are unlikely to be guaranteed. Banking facilities can often be withdrawn.
- Watch out for specialised accounting, such as long-term contracts. Some rules say that profit must be booked before it becomes unconditional. Do not assume that the accounting rules are designed to protect the investor.
- Take a close look at non-underlying costs. They may well be genuinely one-off or technical, but they may still be important.
- Cash is king. Businesses don’t go bust because they make a loss, but because they run out of cash, sometimes even when profitable.
- To understand the usefulness of borrowing facilities requires you to know not just the amount and maturity but also any conditions under which facilities might be withdrawn. This includes covenants, but there may also be other conditions too.
- Study working capital seriously. It is the most likely source of cash problems in a profitable business. If the company is making strenuous efforts to manage working capital, it may be under strain.
- Year-end quoted cash numbers are of limited value. Focus on the average balances and try to find out the peak numbers too.
- If there is evidence of general or growing customer discontent, treat this as potentially serious.
- Segmental profitability matters, but you will probably need to look at the numbers rather than rely on the commentary. Try to understand revenue trend, net margins, net capital employed and cash flows by segment. Check that the business is growing the high margin/return segments.
- Underperforming acquisitions are an amber light. Businesses under pressure may seek to alleviate this by acquiring other companies.
- Defined benefit pension schemes are now commonly in deficit, not because company contributions have fallen, but because monetary policy has led to very low interest/discount rates. However, higher cash contributions will follow and these are not reflected in profit.
- Good corporate governance is very important, but more important is high quality, experienced directors doing a diligent job in the board room. Box ticking governance codes in reality provides little reassurance on company performance.
- Risks and mitigations disclosed in the annual report are generally fairly meaningless, and provide no real reassurance. As in Carillion’s case, it’s often the risks that are not mentioned that prove fatal.
NB. All quotes are from 2016 Carillion Annual Report