“It is excuses after excuses” thundered Roy Keane this weekend, in a damning assessment of Liverpool’s collapse over recent weeks. Having won the Premier League last year, they are all but out of the running barely halfway through the season this time around. The ferocity of Keane’s criticism is unsurprising: he has little natural affinity for Liverpool and his job description involves winding up their supporters. What is notable is the defensiveness of the response. People have raised a host of mitigating factors (mostly the loss of numerous key players with injury, but also the absence of crowds, tough scheduling, refereeing decisions and just the unfortunate ‘bounce of the ball’). However, they have been at pains to insist that these are ‘explanations’ and not – as Keane has claimed – ‘excuses’. I think that’s protesting too much: excuses have a valuable role to play in helping us accurately assess and improve performance, not just in football, but in business and policy too.
What is the difference between an explanation and an excuse? As far as I can make out, both involve giving accounts of what went wrong, but the key difference is that excuses additionally involve trying to escape blame or responsibility. It is no surprise that rubs Roy Keane up the wrong way – a man with a leadership style apparently cribbed from Stephen Fry’s Duke of Wellington in Blackadder – will always want to know who to shout at. But why should the rest of us be so allergic to excuse-making?
I think the issue links back to one of the fundamental challenges of performance management – the principal-agent problem. In many places across our economy and society one person or group of people, ‘principals’ (eg football fans, business owners, taxpayers) need to commission others to act on their behalf ‘agents’ (eg managers and players, workers, public servants). But principals may be suspicious that the agents they have hired are pursuing their own interests and objectives rather than doing what they have been hired to do. A football manager may be more interested in building their reputation and maximising their salary than winning games. A worker might prefer to shirk rather than being productive. A bureaucrat may be more concerned about covering their back than producing optimal policy.
The solution to the principal-agent problem is to try to align incentives, so that the objectives of both sides overlap: it is hard for a football manager to have a good reputation and earn a lot without winning games. More generally, holding people accountable for outcomes is a way of ensuring that our agents care about the things we want them to care about – that is the underlying objective of things like performance-related pay or school and university league tables. The problem with excuses is that they break this connection. The person making the excuse is saying that we shouldn’t judge them on whether they could secure the things we wanted (winning the league) but on some other metric (whether they signed the right players or got their tactics right). Again, this is not just football managers. The Prime Minister is pulling the same move when he insists his government “did everything we could” to fight the pandemic: asking voters to shift their attention from the death count towards the actions he took.
The problem is that excuses often are valid. Agents can’t control everything, and it is rather perverse to hold them responsible for things beyond their control. Novel viruses and damaged knee ligaments are close to acts of God (though we might argue over how predictable they were), so why should we pretend that anybody is to blame for them? In fact, it is not just perverse, but possibly self-defeating. One of the major lessons of the burgeoning field of football analytics is that we should attend more to process and less to outcomes. In any given game or run of games, a team can do most things right and still lose, but if they stick with it over a long enough period, things will turn around. Refusing to hear excuses carries the risk of knee-jerk overreaction – sacking managers, transferring players, changing tactics – when staying the course would lead to better outcomes.
That won’t always be the case, but the valuable role that excuses play is in helping us to determine when a disappointing outcome is the result of circumstance and when it is the consequence of a bad process. Excuses help us sift effective but potentially unlucky approaches from the ones that will never work.
An under-appreciated point is that this works both ways: bad processes can lead to outcomes, just as good processes can lead to bad outcomes. It is striking that the concept of an ‘excuse’ only really applies to failure: we don’t have a word for a mitigating factor that explains why a person doesn’t deserve credit for their success. Yet by failing to recognise such factors, football fans risk reading too much into lucky runs and society and the economy risks rewarding the wrong people and institutions. If we are to get better, those of use responsible for managing performance should listen a bit less to Roy Keane and a bit more to people’s excuses.