Tony Jackson on why improving shareholder values means looking at factors beyond the obvious interests of the company.
Creating shareholder value, some might think, is in theory pretty simple. All you need do is slash your workforce, shove up your prices and squeeze your suppliers. Up go your earnings and share price, and the shareholders are happy.
Real life, of course, is not like that. Things bite back, and so do people. Maltreat your customers and suppliers, and they will damage you in return. Everyone, in the jargon, is a stakeholder: and in the long run, no stakeholder will stay in the game unless they get more out than they put in.
Every worthwhile chief executive knows that, and instinctively adjusts the business accordingly: instinctively, because the interaction of the variables seems too complex to calculate. But must this be so? In the age of the computer, is it not possible to build models which will predict how the players will react?
Yes, some would say. The answer lies in a technique known as dynamic simulation.
In manufacturing, this has been employed at least since the 1960s. If you are running a cement plant, you can build a model of how it functions. You can then feed in variations in ingredients, temperature and so forth.
Each variation creates feedbacks elsewhere in the process. If you run the model repeatedly, allowing those feedbacks to take effect, you can quickly establish how the plant's efficiency will be affected over days or weeks.
One European utility, which prefers not to be named, claims to have applied this principle successfully to management. For example, says the executive in charge of the programme, take the question of whether or not to impose a wage freeze.
The cash savings of this can be quickly calculated, but not the costs. At issue is the level of service the utility can achieve, which depends heavily on staff morale.
In response to a wage freeze, experienced people will leave and have to be replaced by the inexperienced. Those who remain will perform less well. To an extent, this will damage revenues. But how much?
Granted, the executive says, you cannot measure the impact with any precision. But by asking the experts in your organisation - in this case, the human resources director - you can form a rough idea.
Begin with the assumption that a wage freeze will increase staff turnover from 5% to 6%. Run that through the model, and see whether the cost outweighs the savings. If the answer is no, increase the assumption to the highest level you think is feasible - say, 10 per cent. If the answer is still no, you can be pretty sure you have a robust decision.
The important point, the executive says, is that this attempts to measure soft data as well as hard. 'If you only model the things you're sure about,' he says, 'you end up describing only a very small part of your business. That tends to dominate people's thinking, and the softer things get a very low weighting in your decisions.'
The point is reinforced by Professor John Sterman, head of the systems dynamics group at the Sloan School of Management. 'There's nothing wrong with making assumptions,' he says. 'The question is whether they are reasonable, relative to your purpose.
'If morale goes down, and the best people leave first, that's a reality. But you don't have numerical data for it. If, as a result, you don't put it into your model, you're assuming zero impact. And that's the one effect we know is wrong.'
At this point, the sceptic might object that an example drawn from a utility is of limited relevance. The reason why dynamic simulation works in a cement factory or chemical plant is that there are physical constants involved. If a reaction happens once, the odds are it always will.
Similarly, there are more constants to a utility than the average business. The phones do or do not work: gas leaks get fixed promptly, or not. The essential service remains the same.
But at the other end of the spectrum, take an internet company. There, everything is variable, and the future is wholly unclear. Modelling seems out of the question.
But according to one specialist in the field, the global management consultancy PA Consulting, applications are not as restricted as all that. For instance, in the early 1990s PA performed dynamic simulations for Mastercard, which ended up demonstrating that a large amount of shareholder value would be created by co-branding: that is, the creation of an Exxon Mastercard, and so forth.
In other words, shareholders stood to gain by consulting the interests of Exxon as a stakeholder. PA suggests another example, which broadens the concept of stakeholders to include competitors.
In 1991, it claims, one of the big UK brewers adopted an aggressive strategy designed to take its market share from 23 per cent to 30 per cent by 2000. The result, its advisers told it, would be an increase in shareholder value of £600m.
Unfortunately, the other brewers reacted in kind. The result was a five-year price war which reduced the value of all of them.
This illustrates a general point: that to increase shareholder value on a consistent basis, it is not a good idea to grab value from other stakeholders. Not only does this create negative feedback: it is also a finite strategy, since the whole cake is not growing, and may even shrink.
For a variety of reasons, say the enthusiasts, the use of dynamic simulation will increase. 'This whole field,' says Prof Sterman, 'is exploding.'
Granted, he says, some companies have tried to use it and failed. But however complex the variables, you still have to try to measure their effect.
The real question, he says, is whether you regard the reaction of other stakeholders as exogenous: that is, as external to your business. 'If you assume that competitive behaviour or price movements are exogenous,' he says, 'you're in deep trouble. By assuming things are exogenous, you exclude them from your model.'
Prof Sterman takes the argument further, into the basic question of how companies are structured. There is a parallel, he argues, with aircraft design. 'You cannot imagine a company developing a new aircraft today without simulation. What's at stake when people manage a business is at least as big as when they go into an aircraft, but you don't see anything like the same effort going into it.'
In future, he says, people will design organisational structures as they do aircraft. At present, managers are rather like pilots, struggling to keep the aircraft in the air in rough conditions.
'It will become routine for managers to have an organisation which will perform well in extreme weather and high winds,' he says. 'More and more, managers will become organisational designers rather than pilots. And the whole nature of management education will change.'
Copyright (C) Financial Times Ltd, 1982-1998
