Two lessons from the failure of Carillion

Originally demerged from Tarmac 20 years ago, the company had grown through many acquisitions into a large and diverse network of construction and service businesses. In its final form, it lacked a very clear purpose, save for its role as a major player in government outsourcing and public-private partnerships.

Whether you prefer economic activity to be undertaken by the public sector or by the private sector is, to some extent, a matter of faith. The immediate response of Labour Party leader Jeremy Corbyn to the news about Carillion was to call time on the ‘outsourcing racket’. There have certainly been some bad private finance initiative (PFI) deals, but using Carillion to make this argument is a bit of a stretch – rather than making out like a bandit the company in question went bankrupt.

Outsourcing is just another word for buying a service rather than doing it yourself. Whatever the frustrations of dealing with builders, you wouldn’t build your own house extension. You wouldn’t suggest that a car manufacturer with an immensely complex supply chain, that is, with a reliance on outsourcing, should make its own steel, tyres and radios, or provide its own banking, catering and insurance. When we outsource, we identify the things that are best done ourselves and the things that are best left to others. We access other people’s resources. We reduce complex economic activity into manageable and accountable parts.

So like everyone else, the UK government will continue to outsource, but it has a responsibility to do it properly and with great care. This is the first lesson from Carillion. The company was probably an accident waiting to happen – its margins were too thin and it had a balance sheet that was too weak for the economic risks it was taking. But government kept giving contracts to Carillion. That vote of confidence from government increased the eventual damage to the shareholders, the banks, and any number of suppliers and employees that Carillion took down with it. And it transferred substantial wealth to the hedge funds that had been astutely short selling Carillion for several years.

The second lesson is the limitations of today’s financial statements, and the limitations of audit. We need the financial statements to give a true and fair view of the financial condition of a company, year by year. That way, all of the many stakeholders of the company who rely on the financial statements – customers, suppliers, employees, investors – can make a timely assessment of the risks of dealing with it. But, rather, accounting has become a game of financial hide and seek. Like many companies, Carillion appeared to believe that the purpose of an annual report was to paint the business in the most positive possible light and, over the years, they carefully curated their financial disclosures to that end.

The government’s duty in outsourcing

When government outsources, it has a unique responsibility to demonstrate the highest standards of procurement practice and responsible behaviour. Government cannot hide behind ‘caveat emptor’ like just another player in the market. Some SMEs (small and medium-sized enterprises) that may be bankrupted by this say their confidence in subcontracting to Carillion came from the government being the ultimate customer. They have a justifiable complaint. These small contractors do not have the resources to do full financial due diligence. When government gives a significant amount of work to a supplier, it effectively hangs a ‘By Appointment’ badge over that company’s front door. Government is negligent if it does not understand the significance of the signal it is sending.

So what is best procurement practice? It means, among other things, closely monitoring key suppliers for signals of financial weakness. It means taking an entity-wide view of the exposure to individual suppliers and, where possible, using multiple suppliers to limit the systemic risk. It means using one’s economic power in the relationship to enforce good behaviour.

There was a time when you could have been sure that the public sector was not doing any of this. But around five years ago the UK created exactly this capability, in the Crown Commercial Service within the Cabinet Office, for a select group of suppliers that were large enough to be ‘strategic’. Carillion was one of those.

These suppliers were subjected to close financial monitoring. Procurement was ‘joined-up’ across government departments so that, when necessary, contracting with certain suppliers could be limited or ended altogether. A Crown Representative, typically an independent senior industrialist, would be installed in the company as a minder or enforcer. We need to find out why this didn’t work out in the case of Carillion.

From my own experience, having observed the government’s relations with industry at close hand, there are several potential dangers. These schemes are very vulnerable to political interference. And they are prone to rapid entropy. They depend crucially on the, usually short-term, hire of people with commercial skills from outside the government. Some career civil servants will understand the issues, but the constant rotation of officials to other jobs means this knowledge is easily lost. In parentheses, I noted several years ago that the government had stopped sending officials to London Business School, where we do teach these things.

 

What did Carillion’s financial statements show?

Carillion is a textbook example of the limitations of today’s financial statements. In the 2016 report (strapline ‘Making tomorrow a better place’) Carillion’s tone had been typically upbeat and complacent, but the questions people really need to understand about the company’s financial condition went unanswered. Those questions are, fundamentally, two. Is the company making money, in terms of profit and cash flow? Does the company have financial resilience, in the form of a strong balance sheet?

Throughout the years leading up to the failure, Carillion carefully curated certain key financial indicators, in particular, ‘net debt’ and ‘cash conversion’. Below, I examine these in detail, amongst some other negative signals a well-informed reader should have found in Carillion’s financial statements.

One of Carillion’s central financial signals was its dividend, which was £79 million in 2016, or 64% of reported earnings of £124 million. In the 10 years to 2016, Carillion paid out £626 million of cash as dividend. Group finance director Zafar Khan boasted that “the board has increased the dividend in each of the 16 years since the formation of the company in 1999.” Until recent events took over, the growing dividend had supported the share price that, in turn, underpinned the remuneration of senior executives.

But at the same time, Carillion’s pension fund was in growing deficit – £811 million by 2016, which was larger than its £701 million balance sheet equity at that time. So Carillion‘s senior management had chosen to pay dividends to shareholders rather than using the cash to fund its pension commitment to current and past employees. In doing this, management must have known that there was some possibility that the company would ultimately default. That, of course, is what happened. There should be a fundamental principle in government procurement that precludes major suppliers from channelling cash that is needed to fund pensions, out of the business to the owners.

In the aftermath of Carillion, some people argued that companies issuing profit warnings, as Carillion did in 2017, should not get government contracts. In fact a profit warning is simply a statement by a company that earnings will be below what the market expects, so profit warnings are reasonably common events. What was striking at Carillion were not the profit warnings, per se, but the catastrophic performance they were communicating.

At the 2017 interim Carillion reported a loss after tax of £-1,165m for the half-year. This wiped out Carillion’s equity, turning it negative, at £-435 million. The cause was a series of sweeping impairments for cost overruns and other delivery issues, and for payment problems across several of Carillion’s major contracts.

But in the 2016 report published just six months earlier, there was no hint of this contracts bonfire, with no cloud on the horizon. The auditors issued a ‘going concern’ judgement that the company would survive until its next balance-sheet date and, strikingly, there was no goodwill impairment.

By far the largest asset on Carillion’s balance sheet was the £1.57 billion of ‘goodwill’, which is the accumulated difference between the price paid for its acquisitions and the identifiable assets acquired. When an acquirer overpays for an acquisition, that shows up as goodwill. When, perhaps because economic circumstances change, the value of an acquired business subsequently falls, then goodwill is overstated. For both reasons, the GAAP accounting rules – IFRS, in the case of Carillion – require goodwill to be subject to a rigorous annual ‘impairment review’. This involves revaluing the goodwill by making a conservative forecast of future cash flows based on current information about the prospects of the subsidiaries to which the goodwill relates. Evidently, at the end of 2016 there was no reason to write down Carillion’s goodwill.

When a business is only earning the thinnest of profit margins on quite risky economic activity, we need to assess its resilience to negative shocks. We look for what is colloquially called a strong balance sheet, which essentially describes the relationship between equity capital and net debt (financial debt less holdings of cash).

We look for a thick cushion of equity capital to absorb losses, backed up by readily saleable assets that can be sold in times of need. We look for reserves of cash and relatively low borrowing, signalling unused borrowing capacity that can be drawn on in times of need. If the company is borrowing, we like it to be long rather than short-term, since short-term borrowing may not be renewed in a crisis.

Carillion reported a very healthy balance between equity and net debt. At the end of each of the 10 years to 2016, equity averaged £800 million, while net debt averaged £110 million. But neither side of this equation bore close inspection.

This brings me back to the fragility of Carillion’s equity. The £1.57 billion of goodwill – Carillion’s principal asset in 2016 – was more than double the balance sheet equity of £701 million, so the company’s equity would have been extremely vulnerable to any goodwill impairment. Goodwill is not a readily saleable asset and, typically, has no realisable value when a company is liquidated.

Net debt was a financial measure that Carillion frequently used in its financial report. In 2016, the company’s net debt remained at a fairly healthy-looking £219 million – Carillion was borrowing £689 million but held cash of £470 million.

The first question the forensic analyst would ask is, where is the cash? When a company has international operations, it frequently turns out that some of the cash is ‘trapped’ in overseas territories, making it inaccessible.

And what do we mean by debt exactly? As noted, Carillion’s pension deficit had spiralled to £811 million. At some point in the future that deficit will have to be made good so, effectively, when a company runs a pension deficit, the company is borrowing from the pension scheme.

Most intriguingly, Carillion used a financing device known as reverse factoring. ‘Factoring’ is a common arrangement where, rather than waiting for your customers to pay, you get your bank to lend you cash against the invoice, that is, against the receivable. Carillion’s ‘reverse factoring’ did something akin to this, but with suppliers.

In 2012, Carillion announced that, henceforth, it was not going to pay its suppliers for 120 days. At first glance, that would be crippling from many suppliers. But not to worry, because if suppliers wanted the cash earlier, say after 45 days, Carillion had arranged with a bank to pay them earlier. There would be a small bank charge for this, but Carillion would cover the fee.

So why did Carillion enter into this somewhat Byzantine arrangement? I believe part of the answer may lie in how it was able to treat it for accounting purposes. Essentially, Carillion was borrowing from the bank to bridge the gap in paying creditors. But this bank financing was recorded in Carillion’s balance sheet within ‘other creditors’ of £760.5 million, on p118 of the 2016 financial statement (this is apparently the case - there is no disclosure at all on this in the financial statements).

So, arguably, if we also include the pension deficit, Carillion’s debt-like financing liabilities in 2016 were the stated £689 million, plus £811 million and £760.5 million, which is approximately £2.26 billion in total.

Over recent years, Carillion’s customers were taking longer and longer to pay – by 2016, receivables were 38% of sales. Carillion was taking even longer to pay its suppliers. By 2016, if you include the reverse factoring, payables were around 50% of sales, or six months. I imagine it was this worsening working capital situation, and the reverse factoring arrangement, that first attracted hedge funds to Carillion as a company in trouble.

What was unavoidable in the published statements was that Carillion was a very low-margin business. Its operating margin (operating profit/sales) over the last 10 years was highly variable, but in the range 1% to 4%. Its earnings margin (earnings/sales) was between 2% and 3%.

But what did these numbers actually mean? Long-term contracting always rings alarm bells for readers of financial statements, because it offers companies discretion over when to report the income from a contract, with the risk that they will be tempted to report revenues earlier and costs later. Carillion’s financial statements offered no clue as to how it was accounting for long-term contracts.

A new accounting standard, IFRS 15, will shortly come into force, offering much more rigour and disclosure about the measurement of revenues on long-term contracts. This comes too late for Carillion. But in a disclosure in late 2017, Carillion hinted that the introduction of IFRS 15 would require the company to write down retained earnings by well over £100 million. This presumably suggests that the income on current contracts had thus far been overstated by this amount, relative to how the new IFRS 15 would require it to be measured.

Incidentally, users of financial statements have become increasingly wary of companies that habitually report their own ‘alternative’ version of key profit numbers, in preference to GAAP numbers. This became endemic in Carillion’s accounting. For example, in just one page of the company’s 2017 interim announcement, the key results page (p3), Carillion referred to ‘underlying’ numbers no less than 10 times.

Analysts on the lookout for companies that are inflating their revenues understand that overstated revenues simply get booked as increased receivables. These are deducted in calculating operating cash flow, so to catch the overstatement of revenues, analysts focus on ‘cash conversion’, which is the ratio of operating cash flow to operating profit. The idea is that the higher the cash conversion ratio, the truer the profit.

Uncommonly and, at first glance, impressively, Carillion therefore made ‘cash conversion’ a central part of its narrative. It was a key corporate KPI and a factor in top management remuneration. But what did cash conversion actually mean at Carillion?

This was where the reverse factoring arrangement really worked its magic. Apparently, Carillion included the change in the reverse factoring creditor – essentially a financing cash flow – as an operating flow in the calculation of operating cash flow and thus in the calculation of cash conversion. In 2016, Carillion reported operating profit of £145 million and operating cash flow of £115 million, which is a decent cash conversion ratio. But that operating cash flow was increased by the approximately £200 million increase in ‘other creditors’ during the year. Without that, operating cash flow would have been £-85 million.