Why statistics can be controversial and how important they are

Will Davies, Professor of Political Economy at Goldsmiths University of London, questions the role of statistics in our lives.

Will Davies

Professor in Political Economy

View Author ->

Will Davies

Professor in Political Economy

02 Jul 2021

Key Points

Statistics can be seen as too dispassionate to accurately represent people’s real-life problems.
One of the only scientific ways to understand various aspects of society (demography, public health, for example) and to track progress and policy impact is to use statistics.
Statistics started becoming important in politics and policymaking in the late 17th century, gaining in influence in wartime politics.

Why statistics can be controversial

I think statistics are at the heart of many of our dilemmas in politics at the moment, because they are seen as somewhat cold, somewhat technocratic, like they apply a rather objective and dispassionate view of society. For that reason, many political movements that have wanted to rebel against the elites, the technocrats and the experts across liberal democracies of recent years have declared that statistics don’t capture real people or how people are actually feeling in their lives and the forms of suffering that are going on in their society. The idea that GDP – gross domestic product – is something that matters to ordinary people is something that movements and political events such as the Brexit referendum have rejected.

But one of the great challenges and one of the great difficulties that we face at the present moment is that while statistics don’t have very much democratic legitimacy because the people who produce statistics are unelected experts – they work for independent NGOs and technocratic bodies that are some way removed from the ordinary lived realities of many people’s lives – they also are one of the only ways that modern societies have to weigh the benefits and costs of different courses of action. They are one of the only ways of knowing about things like inequality in a scientific fashion rather than just on the basis of a hunch, on what certain people are able to communicate via the media or on who is the loudest.

What statistics offer modern societies

So, what statistics offer modern societies and have done for several hundreds of years is the promise that there is a way of establishing good and bad policies, which is not simply determined by whoever happens to control the media or whoever happens to be elected into government at that moment. That is also why statistics can actually be used to resist power sometimes or to resist the claims that particular politicians are trying to make, that particular media agencies are trying to make, and also perhaps even to resist the more emotive claims that certain sections of society make.

For example, in the context of recent populist uprisings, we’ve seen people try to make the claim that they are uniquely and unjustifiably oppressed, when in fact they might actually, according to statistical measures, be doing better than average. So, statistics in some ways allow us to dampen some of the more emotive and more misleading aspects of democratic movements, which, of course, also opens them up to the charge that they are politically or democratically illegitimate.

The birth of statistics

Photo by Vladimir Mulder

Looking back over the history of statistics, we find that they originate in the late 17th century, around the time when the scientific revolution in the natural sciences was really gathering momentum. After the religious wars that tore Europe apart through the middle decades of the 17th century, statistics offered decision-makers and rulers and monarchs the opportunity to understand their populations and to understand the resources at their disposal in ways that were more scientific. At least that was the promise of statistics: that it would be possible to apply a similarly mathematical or quantitative perspective to human activities as natural sciences, such as astronomy, physiology and the study of botany, had already applied to nature. This meant that it would be possible to start to view economic life or the health of the population in terms of processes that had a certain kind of mathematical pattern to them, and could be predicted in certain ways.

By the end of the 18th century, this had given birth to the field that we now recognise as political economy, and a hundred years after that, what we now recognise as economics, which involves a certain degree of mathematical modelling to understand the outcomes of various different public policies, of individuals taking decisions in marketplaces and what that will do to prices and so on.

Applications for early statistics

Of course, statisticians were the clients of early modern rulers – and those rulers wanted to know what was in it for them. They weren’t just simply interested in science for science’s sake. They were interested in wanting to know how they could become wealthier, how they could grow their populations, because population growth was considered to be a good thing for economic prosperity: developing populations that would be capable of waging wars meant accumulating more territory and more wealth. So, there was a particular reason why statistics were considered to be useful at that particular time.

There were also efforts in early demography – the study of populations – to try and understand the passages of disease. Early modern European cities were repeatedly savaged by epidemics. Statistical science and statistical reasoning held out the possibility that, rather than seeing these things as acts of God or punishments for moral misdemeanours, we could actually start to apply a more rational and more scientific lens to movements of infections through society. Within the context of a new pandemic, we’re seeing the importance of statistics and how statistics suddenly become an absolutely crucial device and a crucial way of viewing the world when uncertain and unexpected events start to course through populations.

Photo by annaliberty111

The political importance of statistics

There are various moments when statistics have increased in their influence and become more politically decisive over the course of history. In their origins they were something quite experimental and niche, so it did not have a huge amount of political influence. But when applied to different types of social and political problems, their capacity to solve problems is what grants them greater political and public influence.

Now, the birth of political economy in the late 18th century, with the work of Adam Smith and others, although that wasn’t a branch of statistics, created the framework within which it became possible to start thinking about national economies and national economic activity in terms of aggregates and in terms of predicted benefits of economic growth, increase in trade, increase in wealth and so on.

In the late 19th century, there was the birth of what we now recognise as social statistics and the science of sociology. There were early attempts to try and understand poverty, or what was sometimes called the social question, dealing with delinquency and inner city slums and poverty and the problems for the very early welfare state in the late 19th century.

Statistics during times of war

One of the other interesting things about where statistics suddenly get their grip upon the State and upon policymakers is that they’ve also been very important during times of war. During World War One, the ability to manage the domestic economies became absolutely crucial. The capacity for governments to plan the production of munitions and uniforms and food and to operate the running of agriculture in their domestic economies became key to how the war effort was fought. That meant that policymakers needed far more statistics than they otherwise would have done.

Similarly, during World War Two, when our contemporary notion of gross domestic product was born, it was widely believed amongst policymakers that the capacity to beat Germany was dependent on how productive the Allies’ domestic economies were. New statistical techniques needed to be developed and implemented to capture how much production was occuring within different national units so as to check that it was increasing at a fast enough pace so as to outcompete Germany in an economic and industrial sense, not just in a military sense.

The example of the European Union

If we think about political and democratic developments in Europe over the last 40 years or so, I think one of the difficulties the European Union has faced over that period is that it has what many people would recognise as a democratic deficit. It often sees itself in a rather technocratic sense. It depends heavily on statistics for understanding different levels of regional development and competitiveness. It uses various scoreboards to try and encourage different member states to become more competitive, to become more prosperous. In some sense, the European Union has been over-dependent on statistics as a way of trying to impose some kind of legitimacy and transparency upon its own activities, partly because that makes up for some of the democratic deficit that exists in the European Union.

Photo by Anikin Denis

Rather than people being represented as they might be at a national level via parliament, political parties, a national media, decisions at the EU level have been far too dependent on a statistical logic. This is where the problem of technocracy or the technocratic elite comes in: there are these very influential decision-making bodies, such as the European Central Bank and the European Commission, which really are grappling with decisions that apply to such a vast number of people that they can’t possibly be negotiated politically or democratically through all of those different public institutions and member states and parliament and so on. Reliance on statistics is often an attempt to try and turn policymaking into a branch of science, much as statistics often promised might be possible, but in ways that can seem like an affront to democracy.