Last week the Financial Times reported that two executive directors at Interserve earned a combined £1.99 million by the end of the 2018 financial year, roughly one third of which was annual variable performance related bonuses. But by April 2019 Interserve had been taken over by its lenders in a pre-pack administration.
It would be unfair to pin all of the blame for Interserve’s collapse on these two directors – they were hired in the Autumn of 2017 when much of the damage to the company had already been done. The bonuses also still have to be approved, so may not actually be awarded. But it does beg the question of why the recommended payments were so high when performance was so awful, with the firm teetering on the brink of bankruptcy. More substantially, it raises questions about whether the reward structures at board level over the long term may have actually contributed to firm failure.
The latter point merits consideration. Much of the problem at Interserve was the arcane way performance was measured for the purposes of calculating the board’s bonus payment. At the centre of this problem is the murky accounting concepts of ‘underlying profits’ and ‘normalised earnings’. Both strip out one-off, non-recurring items that notionally interfere with an investors ability to judge the underlying performance of the firm over time. However, when this concept is linked to bonus payments, it can introduce perverse incentives because there is a great deal of discretion when choosing what items to include and exclude.
In Interserve’s case, the 'underlying' operating profit for 2018 was reported as £92.7 million. But this figure excluded net costs and charges of £98.6 million, including a £55.2 million charge for goodwill and other asset impairments. Those costs turn a healthy ‘underlying’ profit into a formally recognised operating loss of £5.9 million. Worse still, the excluded net cost figure of £98.6 million included a realised gain of £70.6 million from a pension re-indexation. Without that re-indexation gain, the costs excluded from the ‘underlying profit’ calculation would have been £169.2 million and actual operating losses would have been £76.5 million.
Figure 1: Underlying vs Operating Profit at Interserve, 2018
Source: Interserve 2018 Annual Report
The majority of the 2018 bonus pay for the two executive directors came from the Annual Variable Pay component. The targets for that bonus focused on three indicators: ‘operating performance’ (one third), ‘operating cash flow’ (one third) and ‘strategic targets’ (one third). If each of those targets were 100% met, the two would receive an amount equivalent to 125% of annual salary. As the table below shows, the majority of the annual variable pay was awarded on the basis of Interserve’s ‘operating profit’ performance - they were recommended to receive 33.22% of annual salary because they had met 79.7% of the maximum target. This implies that the measure of ‘operating profit’ used in the calculation of this bonus was somewhere between £91 million and £101 million (figure 2).
Figure 2: Structure of Annual Variable Pay, Interserve 2018
Source: Interserve 2018 Annual Report
But we know from Figure 1 that actual operating profit was -£5.9 million, which would imply a ‘nil’ bonus award. The operating profit figure used in the bonus calculation therefore must have been in fact much closer to the £92.7 million figure used in the calculation of ‘underlying profit’. A rather clumsy and elliptical footnote reveals that:
"In determining the vesting percentage the Committee had the discretion to adjust the calculation and it had exercised that discretion which is reflected in the stated vesting percentage for the Operating Profit element. For example, the original targets may be adjusted for any discontinued, divested or acquired businesses to ensure the targets remained similarly challenging and it had made such an adjustment to the targets in respect of the disposal of Access Hard Services which is reflected in the vesting percentage disclosed."
It is not clear what this footnote means precisely, but the implication is that many costs were stripped out of the calculation of actual operating profit and it was on that basis that bonuses were awarded. This practice goes back some way. The bonus structure at Interserve has changed significantly over time, but the principle has not. Earlier bonuses were awarded primarily on the basis of ‘Normalised Earnings Per Share’, which are defined as:
"…headline EPS adjusted to reflect growth in underlying value created by (a) removing the impact of IAS 36 Impairment of assets and IAS 39 Financial instruments; and (b) recognising or removing “one-off” events at the discretion of the Committee"
In other words, goodwill impairments and other costs were stripped out of the calculation of earnings.
In hindsight, it looks reckless to strip costs of this nature out of any assessment of management performance because the upside earnings and downside impairments from an acquisition result from the same decision. When the board decides to acquire a firm at a particular price, there is always the risk of overpayment. Overpayments mean more goodwill booked, which will be impairment tested at a future date. But under these bonus metrics management are rewarded for the upside on the new earnings acquired, but take no punishment on the downside impairments because they are considered ‘one-off costs’ and excluded from the bonus calculation. They are fundamentally part of the same decision.
Remuneration structures like this encourage profligacy. They also encourage gaming around how certain income sources and assets can be recorded and gaming about which income and costs should and shouldn’t legitimately be included in the ‘underlying’ or ‘normalised’ bonus metrics.
In the case of the former, the present valuation of assets like goodwill or even contract profits involves subjective and speculative judgements about future cashflows and discount rates. Interserve themselves make this plain in their 2017 annual report:
"The recognition of revenue and profits on long-term construction and service contracts requires management to exercise significant levels of judgement involving a high degree of discretion and control"
The auditors even provide this handy visual in the 2017 annual report which shows that those items which may have the highest value impact on the firm are also (coincidentally) those items where management exercise most discretion: revenue recognition on contracts, going concern, the disclosure of non-underlying items and goodwill impairment. It should be noted that this does not imply anything illicit – discretion is part and parcel of the fair value regime.
Figure 3: Key audit issues, Interserve 2017
Source: Interserve 2017 Annual Report
In the case of the latter (costs), fair value rules are often interpreted in such a way, that management errors of judgement (overpayment for firms, over-booking of contract profits etc) are considered one-off, non-recurring, exceptional costs and are then stripped out of bonus calculations. This is ‘heads I win, tails you lose’ remuneration practice.
This all illustrates some of the paradoxes of financialization – that the beneficiaries of shareholder value driven strategies are not necessarily shareholders, but boardroom elites and their intermediary helpers. A recent book by Dumenil and Levy notes that managerial capitalism is still alive and well. Bill Lazonick’s 2015 article links the rise of inequality back to the firm, and the egregious management salaries that arise as part of the shareholder value revolution.
Within that context, Interserve is an allegory for late capitalism’s dysfunctions, where capturing gatekeeper positions inside a firm rather than generating real value drives economic outcomes and inequality.