Dalglish arrives at Newcastle and Horlick moves to ABN Amro. The common theme: do teams, or individuals, capture the returns to outstanding performance?
Rarely do sporting stories make the front page of the FT, but the departure of Kevin Keegan and the arrival of Kenny Dalglish at Newcastle United is big enough news to be an exception. It had to vie for space, however, with Nicola Horlick’s suspension from Deutsche Morgan Grenfell, amid allegations that she planned to lead a group of her fund management colleagues to ABN Amro. Two different events, but with a common financial theme. What determines the distribution of the returns to outstanding performance?
Five years ago, I and some colleagues confirmed most people’s worst fears about economists by bringing econometrics to bear on the football league*. We analysed the performance over twenty years of all the clubs and related their performance to what they had spent on players. Our measure of expenditure included both wages and transfer fees.
We found that you got what you paid for. Manchester United had done consistently well – but no better than you would have expected given the quality (as measured by price) of their playing staff. There were, however, some exceptions.
There was a Brian Clough effect. Derby County and Nottingham Forest, which had been managed by Clough, won more matches than they should have while he was manager, only to revert to the mediocrity our equations predicted when he left. More recently, Kenny Dalglish seems to have achieved the same result. This phenomenon seems to apply to managers rather than to players. Of course, there are superstar players just as there are superstar managers, but superstar players mostly seemed to extract the full value of their contribution for themselves.
Perhaps it is important that there is an active (and accepted) transfer market in players, but not for managers. During the period when football league clubs operated a cartel to hold down players wages, the returns to stars became capitalised in transfer fees, so the profits from their skill went to the clubs which had groomed them to stardom. Today, these returns are split more evenly between the club and the players themselves.
So the palpable, if diminishing, reluctance of clubs to compete too actively for the services of the top managers may keep some of the returns to managerial talent in the hands of the clubs rather than the managers. And top managers like Clough and Dalglish may reduce their potential earnings by being difficult people for boards to get along with. In the City similarly the breakdown of traditional self-restraints in competition between firms has led to a steady transfer of returns from the profits of firms to the earnings of individuals. That is why Ms Horlick earned £500,000 a year – and why she could think it was not enough.
But some football clubs showed superior performance even without superstars. Most conspicuous amongst these was Liverpool, which over the twenty years of our analysis had not only been the most successful of all clubs but had achieved this with an expenditure on players far less than the experience of other clubs suggested was necessary. Liverpool was one of the few cases where there was evident value to the team – the whole was worth more than the aggregate of the points. And if you can’t buy a superstar on the cheap, this is the only way in which you can translate outstanding performance by the organisation into profits for the firm.
So how had Liverpool done this? To understand it better we built a simple economic model of the game of football, in which a player with the ball at his feet can either kick it towards the goal, or pass it to a player better placed than him. What he decides to do will depend on two things – the extent to which his rewards reflect individual or team performance, and what he thinks his colleague will do with the ball. If the player who receives the pass is himself inclined just to kick it towards to goal, the first player is less likely to make the pass.
This dependence of individual decisions on expectations about what other players will do creates the possibility of two different kinds of team. In one, the normal behaviour of the team is to pass: in the other, to shoot for goal. Moreover, these patterns of behaviour are stable. A player from a passing team who joins a shooting team will find that passing behaviour is unproductive there: a player who moves from a shooting team to a passing team will discover that unless he changes his behaviour, he rarely gets the ball.
Now it is easier to change from a passing game to a shooting game than it is to move in the opposite direction. That means that if the passing game gets more goals for the club – and it probably does – then those clubs that manage to achieve it will not only outperform but discover that their outperformance persists. But the returns from this outperformance will be earned by the group of people who have passing behaviour in common – by Ms Horlick’s team, rather than Deutsche Morgan Grenfell as a whole. Only when, as in Marks & Spencer, the team becomes so large that it is inseparable from the organisation with its achievement be securely established as assets of the company rather than assets of the individuals and groups who make it up.
Designing structures which achieve a balance between co-operation and competition, which combine team behaviour and individual motivation is one of the hardest parts of building organisations, or designing economic systems. And the more individualistic culture of the post-Thatcherite world does not always make it easy to achieve.