LORV | website

Ashby's Law of Requisite Variety

This excerpt from 'The Little Book of Beyond Budgeting' explains, in simple terms, how Ashby's Law works and how it can be used to analyse the design and operation of traditional budgeting systems.

Ashby’s Law tells us that, to reliably achieve its goals, the flexibility of a control system (which is what budgeting is) needs to at least match the volatility of its environment.

Because budgets are inflexible and the real-world environment in which it is applied is unpredictable, this means that budgeting is incapable of fully achieving its intended purpose: to manage the performance of an organisation back to a large set of fixed targets.

Budgeting is designed to fail.

The law of requisite variety – A tool for managing complexity

What Ashby’s Law tells us may seem like simple common sense. But if it is, it is common sense that is widely misunderstood or ignored.

Where Newton’s Law deals with mass, Ashby’s Law uses a measure of complexity called variety.

Variety describes how many states a system can adopt.

For example, a heating system that can only be turned on or off has a variety of 2, whereas one that has 4 heat settings has a variety of 5 (off, plus the 4 ‘on’ settings).

Ashby’s Law describes the relationship between the varieties of three interacting systems:

First there is the environment. A stable environment has low variety; a volatile one has high variety.

Second there is the variety of the regulator or decision-making system. A system that provides a lot of flexibility has high variety (such as the heater with 4 settings) whereas one that has few control options (e.g. a heater with just an on or off switch) has lower variety.

Finally, there is the variety of the objective or goal. Taking our heating system example, if we can tolerate a wide range of temperatures in our room, we would have a high variety or ‘loose’ goal, but if we need it to be held within a narrow range, we will have a low variety or ‘tight’ goal.

Ashby’s Law is important because it allows us to describe what qualities a regulatory or decision-making system (like budgeting) needs to be able to consistently achieve the goals of the organisation it serves.

Expressed in simple terms Ashby’s Law states:

The variety of the regulator

must be equal to or greater than:

The variety of the environment

divided by:

The variety of the goal

This tells us how much flexibility we need to build into a decision-making system (the requisite variety) if we are to meet our defined objectives in any given environment.

For example, the range of control you need over your heating system is determined by how much the outside temperature varies and your target internal temperature range.

The implications of the LORV for a heating system are obvious but the same principles also apply in much more complex situations.

For example, the regulatory variety of a sailing ship depends on the design of its hull, the number of masts and different sails it has, whether it has radar to enable it to sail at night or an auxiliary motor to enable it to move when there is no wind.

It also depends on the crew – how many and how skilled they are.

Whether any particular boat is adequate firstly depends on the variety of the environment: the range of weather and sea conditions that the boat encounters.

But it also depends on the job it is required to do, which is expressed as a goal.

If it must call at specific ports, at specified times to pick up passengers, the goal set has low variety. This is because the range of acceptable ‘system states’ is small, defined by a precise combination of time and place.

If, on the other hand, it is on a lazy pleasure cruise, its goal variety is high.

The mathematics underpinning Ashby’s Law can become convoluted and in practice it is almost impossible to measure the variety of extremely complex real-world systems.

But while you cannot use the LORV to precisely engineer solutions in the way you can with the Law of Gravity, it is important to understand and respect it when you are designing organisational processes.

A good design is one that consciously aims to balance the variety equation, as shown in the graphic below.

On one side of the scale there is the variety of the environment. This must be balanced by the variety of the regulator plus the variety of the objective on the other side.

The importance of getting the design of the variety balance right is illustrated by what happens if the common-sense principles of Ashby’s Law are ignored – as they usually are with budgeting.

Budgets regulate the organisational system through the allocation of resources. This controls what can and can’t be done, just as a heater switch controls the energy input into a room.

Because budgets are prescriptive and fixed for a period of a year the regulatory system has low variety.

The budget also defines the goal set – the system of targets. This also has very low variety, because every element of the budget is broken down into quarterly or monthly targets that are used to measure and manage performance.

What Ashby’s Law tells us about this situation is that it is only possible to meet all budget goals if the environment is highly predictable – if it too has a very low variety.

If the environment is not predictable, as in normally the case in business, the variety equation does not balance, and something must give way to restore equilibrium. And this usually has negative consequences.

One thing that could ‘give’ is the level of environmental variety on the right-hand side of the variety balance.

Unlike our example of a boat and the weather, it is possible for an organisation to influence its environment.

For example, choosing not to serve customers with complex needs can reduce environmental variety. You can also constrain their demands by narrowing the product range or by forcing them to use a low variety channel such as an automated telephone answering system.

If the variety of the environment isn’t reduced, Ashby’s Law tells us that the variety on the left-hand side of the balance must increase. There is no other alternative.

One way to restore the variety balance is to ignore (or fail to meet) some targets, since this increases the output variety of the system. In these circumstances, there is a high risk of prioritising those goals that are explicit, visible and tied to incentives, over those that are not. But this may not be in the best interest of the business.

The only other way that the variety equation can be balanced is by adding variety (flexibility) to the regulatory system.

Ideally flexibility will be built into the system, but if it isn’t, people can create it for themselves by ignoring, bending, or breaking the rules.

Often the motives for not following the system are not malign and no harm is done.

But if the stakes are high, deliberate ‘law breaking’ can prejudice the reputation or prosperity of the business, as illustrated by the mis-selling perpetrated by banks when employees are given very aggressive (low variety) sales targets.

To be clear, it is not necessarily bad for environmental variety to be constrained, or some targets missed – indeed, it may be unavoidable. The real problems start when the variety equation is balanced by accident, rather than design, and in a way that can negatively impact the whole organisation.

The real problems start when the variety equation is balanced by accident, rather than design, and in a way that can negatively impact the whole organisation.

If these failings were recognised to be the result of a poorly designed system, the damage can be contained. Usually, however, the finger of blame is pointed at individuals or ‘the culture in the business’.

But no amount of punishment, exhortations to behave better, or even litigation can improve matters if the problem has been misdiagnosed.

None of this – the performance failure, poor customer service or bad behaviour – is necessarily the fault of any individual or groups of individuals working within the system. It is usually the fault of the system.

If blame needs to be attributed anywhere, it lies with those who failed to design the system in line with Ashby’s Law.

The Systemic Failings of Traditional Budgeting:

It ignores the nature of the environment since an annual cycle is used irrespective of the nature of the business environment.
It has low regulatory variety because the level of detail and the difficulty of making changes means it is difficult to respond to unanticipated threats or opportunities emanating from the environment or any other source.
It has low goal variety in the form of a large number, specificity, and fixed nature of targets

Like Newton’s Law, the Law of Requisite Variety predicts what cannot work.

Budgeting inevitably leads to suboptimal performance for the same reasons that having a centrally planned economy helped the Soviets lose the Cold War.

But Ashby’s Law does not guarantee success.

Designing a better system for running a business needs an understanding of how to apply the science, just as a successful space mission involves more than strapping thousands of tonnes of rocket fuel to a metal tube.

To apply Ashby’s Law successfully the whole system needs to be designed and run in the right way.