In 1949, the economist Bill Phillips constructed a machine that modelled the British economy—using water.
By controlling taps and valves and observing the flows of liquid, trainee economists learned how to model the economy using mathematical equations. The problem was that the models did not always match reality.
The extraordinary Bill Phillips
Bill Phillips was born in New Zealand in 1914.
At the age of 15 he took an apprenticeship as an electrical engineer.
Six years later, in 1935, he moved to Australia where, among other casual jobs, he hunted crocodiles and worked as a goldmine electrician. After a year in the outback he moved to the UK, aged 22, and qualified as an engineer.
After the Second World War ended, Phillips studied sociology at the London School of Economics (LSE).
However, he was quickly drawn into 1940s experiments that used mathematical models to explain how the economy worked.
There has been an increasing use in economic theory of mathematical models … However, those students of economics who, like the present writer, are not expert mathematicians, often find some difficulty in handling these models effectively.
Bill Phillips (1950)
Phillips encountered a problem: how could non-mathematicians learn these new mathematical approaches?
To solve it he drew on his engineering background, building a two-metre tall machine that simulated the equations using pipes, valves, tanks and pumps. Phillips completed it in 1949.
Money flows like water
Bill Phillips used water to represent money as it flowed around the economic system. Valves could be opened or closed to represent variable effects, such as the rate of interest on savings or investment.
Graphical curves, describing things such as the way interest rates varied over time, could be cut into plastic sheets and physically ‘read’ by the machine as it operated.
Watch how Phillips' economics computer modelled the economy using the flow of water:
What made the machine particularly realistic was the time it took to settle down after making a change. As the water ebbed and flowed, gradually reaching a stable position, so too did the economy. Phillips' machine was so well received by senior LSE academics that he was offered a job as a lecturer before he had even finished his degree.
How did Phillips' model help teach economics?
The importance of teaching economics in society had been apparent for many decades before Phillips completed his machines.
In 1885, Henry Cunynghame, an economist who influenced John Maynard Keynes, made an ‘equation-solving machine’ as part of his work ‘to educate the rising generation’ and ‘support justice’, as he put it.
Science Museum Group Collection
Today, there is a multitude of resources to help us become more informed about how the economy works.
So how was Phillips' model influential?
All in all, Phillips made 14 of his hydraulic economics machines.
The 1949 prototype found its way to Leeds University. Improved versions went to universities across the UK as well as overseas institutions and corporations in Australia, New Zealand, South America and the USA, where the device was dubbed the ‘Moniac’.
Two more were used at the LSE to train a generation of economists in the mathematical approach to economics.
The educational role of the machines was immediately spotted by commentators at the time.
The American business magazine Fortune noted in 1952 that ‘interest in popular economic education’ and ‘the need for demonstrating fiscal problems before congressional committees’ meant that Moniacs would be in demand for teaching economics in ‘thirty fascinating minutes’.
Mathematical models in the real world
Countless people use mathematical models to describe economic or financial activity, but they do not always know when to stop following them. As the Nobel-Prize-winning economist Peter Diamond has said, ‘taking a model literally is not taking a model seriously’.
Economics is simply the combined actions of real people making difficult decisions about money, decisions that are often based on emotions as well as rational analysis.
The simple mathematics modelled by Henry Cunynghame on his equation machine, and by Bill Phillips in his hydraulic computer, have been overtaken by mathematics that incorporates ideas of probability: the likelihood (rather than the certainty) that people will behave in a certain way.
This starts to make allowances for irrational decisions and the psychology of human behaviour.
It is hard to see how any mathematical model will ever truly capture that human behaviour under all conditions so it is vital we understand the limitations of the models.
The more we arm ourselves with knowledge about the way the economy works, the more we can hold those who use the models to account.
