LBS logo London experience. World impact.

In the middle of a change reaction

True and long-lasting change only happens if the structure of the underlying system is changed.

By Costas Markides and Anita McGahan . 26 May 2015


In 30 May 2008 a hit-and run incident was caught on CCTV in Hartford, Connecticut. It showed 78-year old Angel Arce Torres crossing a two-way street when he was hit by a speeding Honda. The amazing thing (captured on video) was that nobody bothered help him, or even stop the traffic.  The incident provoked an outcry in the press. The Hartford Chief of Police Daryl Roberts said that the lack of empathy shown from passers-by exposed the nation as having lost its moral compass. “We’ve got to look at ourselves and understand that our moral values have now changed,” he said. “We have no regard for each other.”

The police blamed our “changed” moral values for our behavioural attitudes. But instead of blaming individual behaviour, let us consider two things: we must first understand the underlying structure that created the problem and then change the structure to encourage favourable behaviour and thus address the issue.

What could be the possible explanation as to why rational human beings do not fulfil their basic human need, empathy? Are moral values in modern society so far south of optimal on our moral compass? Is culture to blame for encouraging individualistic behaviour?  Or, does the problem lie in the time pressures we face and our shift in priorities?

The structure of the system

The idea that the structure of a system creates the behaviours within it is one of the most fundamental principles in management literature. What it implies is that individuals should not be blamed for behaving in a suboptimal way. It is only by changing the underlying structure of the system that we will get them to change their behaviours.

You can apply this principle to real-life situations – business or social – immediately, but we need to go beyond just manipulating the underlying incentives. Jay Forrester commented: “All [social systems] seem to have a few sensitive influence points through which the behaviour of the system can be changed.” Find these influence points in a system and you are halfway to achieving deep and sustainable change in that system.

What exactly is the underlying structure that needs changing? In the case of Angel Arce Torres, we argued that values, culture and incentives are three important elements of this structure. Donella Meadows’ book Thinking in Systems eloquently defines a system: “A system isn’t just any old collection of things. A system is an interconnected set of elements that is coherently organised in a way that achieves something.”

Thus, to understand the underlying structure of any system, we must look for three things:

First: The system’s purpose 

1. The accepted philosophy on which the whole system is based. This is the set of unwritten and deeply-held assumptions and beliefs about how the world – and not just the system – works.

2. The goal of the system – what are we trying to achieve? The goal is important because it determines our strategy.

Second: The main elements that make up the system

1. The rules in the system. These may be formal or informal, physical or social. They define what is acceptable and provide people with the parameters to operate within.

2. The physical structure of the system which includes the underlying processes as well as the structure of the hierarchy.

3. Information flows in the system. What information is delivered in the system, by whom and how?

4. Time delays in the system. These may be information delays or delays between action and effect.

5. Positive and negative feedback loops in the system.

Third: The interconnections in the system

1. The incentives in the system. These may be financial or nonfinancial, positive or negative.

2. The cultural norms and values in the system.

3. The way the system absorbs change (called buffers or stabilising stocks).

Changing systems

There are several factors that produced the suboptimal behaviour – incentives, time pressures, underlying moral values, group-think and so on. Failure to ‘correct’ the underlying factors will lead us to the same suboptimal behaviours again and again. Up to 70 per cent of our behaviour can be explained by our situation – our personality accounting for only 30 per cent. Academics in the field of system dynamics describe the collection of factors that influence how we behave in a different way. To them, factors make up the ‘underlying structure’ of the system in which we operate. The ‘system’ may be mechanical, electrical, economic or social. Whatever it is, system dynamics argues that the structure of a system creates the behaviours within it.

Systematic (beer) failure

System failure is an obvious principle in the ‘beer game’, an educational exercise – traditionally the ‘production-distribution game’ – developed by Professors Jay Forrester and John Sterman at MIT. In the game each board consists of a team of four players: Retailer, Wholesaler, Distributor, Factory. Customers purchase from the retailer who then places an order with the wholesaler. At each of these stages, there are time lags, representing ordering and shipping delays.

The players’ objective is to minimise total team cost. The players have to balance two costs: enough beer to supply, but not too much to create costly unused inventory. Crucially, no communication is allowed between players.

Customer demand is kept constant, but despite this fact there are always fluctuations in the orders placed. And the fluctuations amplify: for example, if the retailers increase their order by two beers, by the time the shock is transmitted down the line, the factory is increasing their order by up to 20 beers (not ideal).

We have now played this game with hundreds of diverse executives and no matter who plays the game; the suboptimal behavioural patterns are the same each time. In a situation that is so simple, how can it be possible that people always behave so poorly? The first reaction of players is to blame other team members, while others blame the game itself. However, the important question is: why did we fail?

There are structural flaws in the game that create predictable behaviours. The multiple layers in the system amplify shock waves down the system. The lack of communication among players hinders. Information about pinch points is missing. And team costs are forgotten. Recent research demonstrates that if structural ‘faults’ are removed, behaviour will improve.

Changing the people without changing the system

The moment something goes wrong, we immediately search for scapegoats. For example, Charles Ferguson, the director of the movie Inside Job, said: “Three years after a horrific financial crisis caused by massive fraud, not a single financial executive has gone to jail.&rdquo

We hold individuals responsible for systemic problems because privileged leaders represent the system itself. We assume that the best mechanism for changing a system is to lead from the helm and that failure to change the system by a leader is sufficient grounds for keelhauling.

What will it take to prevent – let’s say – bankers from succumbing to the conflicts of interest in their professions? Changing the culture of banking requires thinking differently. Replace the people without changing the values, culture and incentives and the same mistakes will occur over and over.

Changing the incentives without changing the system

The same idea applies to incentives. To get people to behave in ways we want, all we have to do is to give them the right incentives, right? No – incentives are only one element that makes up the underlying structure of the system. In 1994, the World Trade Organization created an incentive for pharmaceutical companies to sell drugs in developing countries.

A new policy was adopted on the theory that the incentive – a patent in this case – would make it worthwhile for drug companies to distribute their drugs in poor countries, as imitators would not compete down profits. The goals: to help poor people as well as the companies themselves.

If the incentive worked, pharmaceutical companies could help battle malaria, tuberculosis and HIV and other diseases. But nearly 20 years later, the incentive has not worked.  Recipients of the drugs cannot afford the drugs; therefore profits for drug companies are limited. Even more fundamental than the ability to pay, is a system for delivering drugs. Too few clinics are available. Pharmacies, laboratories, hospitals and medical schools are in short supply. Distribution centres don’t exist.

As a result, patent incentives don’t work. No matter how strong the profit incentive, drug companies can’t effectively get their products distributed and administered. Incentives are only one part of a much bigger picture. What is needed is coordination and leadership to get the structure of the system in place so that elements of the system – such as incentives – can work.

Change in two ways

Remember, not all component parts of a system need to be changed for fundamental change to occur. Some component parts are more influential than others and this is where effort needs to be focused; and, second, small changes in the underlying structure of a system can produce big changes in behaviours in that system. The trick is to focus our efforts at achieving small changes in high-impact parts of the system. Knowing where to focus one’s effort is half the battle won. Succeeding in introducing small changes will (systematically) guarantee you success.


Ann van Ackere, Erik Reimer Larsen and John Morecroft, ‘Systems Thinking and Business Process redesign: An Application to the Beer Game’, European Management Journal, Vol. 11, No. 4, 1993; Chris Bartlett and Sumantra Ghoshal, ‘Rebuilding Behavioral Context: Turn Process Reengineering into People Reengineering’, Sloan Management Review, 37 (1), Fall 1995; Jay W. Forrester: Principles of Systems (2nd edition), Productivity Press, 1968; Jay W. Forrester, The Collected Papers of Jay W.

Forrester, Wright-Allen Press, 1975; Donella H. Meadows, Thinking in Systems: A Primer (edited by Diana Wright), Earthscan, 2009; Lee Ross and Richard E. Nisbett, The Person and the Situation: Perspectives of Social Psychology, McGraw Hill, 1991; Peter Senge, The Fifth Discipline, Century Business, 1990 

Comments (0)


Subscribe to LBSR

Get the latest ideas and opinion from London Business School’s experts, straight to your inbox