1 November 2021
Six years ago, Daniel Mügge, Professor of Political Arithmetic at the University of Amsterdam, started researching economic formulas with his team. Why does this deserve our attention, and what message does Mügge have to share with us?
For Mügge, the interest in economic formulas began with the observation that while figures on national economic performance have an air of solidity and neutrality, there is plenty of leeway in their conceptualization and calculation. ‘If you look closely at the economic figures and the formulas that underpin them, you realize that something like unemployment, for example, cannot be measured objectively. Choices are made about what is and what is not heeded in economic measurements. For example: do you include the effect of rising property prices on inflation? Does the gross domestic product include unpaid work? Is it appropriate to set off government assets against its gross liabilities to get a better idea of the fiscal situation?’
According to Mügge, the list of choices to be made is long, but this ambiguity is invisible in the final figures. ‘They seem to speak for themselves, whereas the use of one formula instead of another can have a considerable impact on policy. Which region needs investment and which one does not? How meaningful are purchasing power projections if they ignore the housing problems of young families? In short, there is a political dimension to economic statistics, both in terms of the consequences they have and in terms of who actually chooses specific formulas and why.’
The statistical measurements we use today were quite appropriate for industrialized economies of some decades ago, says Mügge. ‘The formulas were developed at a time when the economy was dominated by the production, and when consumption revolved around tangible items, from agricultural products to television sets. Similarly, employment at that time concentrated on manufacturing in factories, but also, for example, around mining in Limburg. Full-time positions were primarily filled by men and easy to count. And by monitoring prices for everyday necessities such as butter or clothing, economists were able to compile inflation figures that gave a good idea of the cost of living for an average Dutch household. Economic figures fit the lived reality in the 1950s and 1960s pretty well,’ argues Mügge.
‘Fast forward to 2021, and we find ourselves in an economy and society in which boundaries have blurred,’ Mügge continues. He highlights the fading differences between our working and a non-working lives, between being in and out of work – think of self-employed people who sometimes work many hours and sometimes few, or flexible contracts – and between what is a product with a clear price and what is not, ‘like a free app on your mobile phone.’
Mügge postulates that the definition of ‘a good life’ has also changed and become more difficult to quantify. ‘Especially for people with a reasonably high income, naked purchasing power – how many cartons of eggs and sticks of butter can I buy with a monthly salary – is no longer a meaningful indicator of living standards. They also value peace of mind, stable house prices, access to the property market, affordable education for their children or, for example, the care with which a product is made. Goals such as climate change and sustainability, or the negative side of economic growth, are even more difficult to measure.’
Whereas the “old economy” could be captured in figures fairly well, Mügge argues, this is much less true for today’s economy and society. ‘Even so, policymakers still use those figures as a guide. The danger is that people no longer recognise themselves in the figures and feel that their problems are being overlooked, or that the figures do not reflect social problems as people experience them. This can lead to a loss of confidence in the economy and politics or foster a negative attitude towards politicians,’ warns Mügge.
Statisticians want to keep figures free from noise
Mügge emphasises that statisticians recognise these limitations of economic statistics, but often have no way to incorporate these new and vague aspects of the economy into figures without damaging their reliability. ‘They fear noise in the data, because there is so much more that requires interpretation – like the monetary value of unpaid work or a free app such as Google Maps. The desire for solid and reliable figures creates a bias that prevents us from including the new and blurry aspects of the economy in the figures that we use to interpret our society.’
The aim of Mügge’s research was not to develop better measurement methods. ‘That was not our job. Many others try to do that already – often with only moderate success.’ Neither does Mügge reject the use of economic figures altogether. His intention was to highlight the political connotations and limitations of economic statistics in order to improve the way they are used. ‘We argue in favour of a more nuanced approach to the strengths and weaknesses of these figures and the realisation that there are limits to what you can calculate well. We have to get number-smarter in that sense.’ For Mügge, this also means that governments should recognise that not everything can be substantiated with solid figures and should not be afraid to make decisive policies in pursuit of goals that are difficult to quantify. ‘You have to prevent good ideas from falling by the wayside because you cannot measure them with solid figures, for example by means of a cost-benefit analysis.’
At the same time, Mügge cautions against making communication about economic figures too complex. ‘There is a fundamental tension between the desires to keep things simple and to do justice to the complexity of the economy and society. On the whole, policymakers and citizens alike need to become wise to the uses of figures and better understand the limitations of figures and purchasing power pictures and the complexity of the calculations.’
The FickleFormulas study was made possible by an NWO Vidi Grant and an ERC Starting Grant.