How research informs policy

Dr Martin Lukavec, Senior Lecturer at the London School of Business and Finance, discusses how research informs policy, illustrated by the examples of randomised trials, behavioural economics and economic incentives

Research informs policy – or at least it should – provided our politicians forget their ideologies for a minute or two, and our scientists do the same. In this editorial, I would like to forget the state of world affairs, concentrate on the positives, and write a few paragraphs about a few interesting cases where everything worked close to how it should, and nobody got overly upset by anything.

Randomised trials

The UK is a world leader in randomised trials and their use in informing policy decisions. Randomised trials are the gold standard of (not only) social research. Their alternative, casual correlations, provide a skewed picture of cause and effect because we live in a world where everything, to some degree, correlates with everything else, making it difficult to determine causality.

On the other hand, if you randomly select two groups, change something for one of them (treatment), and do nothing for the other (control), it allows you to pinpoint the effect. Randomised trials are hugely important, and their findings are considered as close to reality as you can get. But they are also very expensive.

Moreover, randomised trials cannot be used to measure adverse effects. Imagine the ethical implications of, for example, making a group of well-off, educated people homeless to see what impact it has on them. Wait a minute – that is the plot of the 1983 movie Trading Places. Well, never mind. If you watched the movie, you know that the ethical problems of the experiment foiled the whole thing.

So, randomised trials are mainly used for positive interventions. In the UK, for example, the Education Endowment Foundation (EEF) tests literacy and numeracy interventions in schools to determine the most effective approaches. Because of the randomised nature of these trials, their research is hugely important and informs policy better than anything else.

The Department of Education references some of EEF’s findings in its national strategy. EEF has also done much important work on teaching strategies – examining what educational interventions make better teachers, which teaching techniques work best, and many other areas. This research is hugely influential worldwide, and policymakers worldwide look to these findings for inspiration.

Behavioural economics

Another area where a lot of important work is done is behavioural economics. Suppose you took an economics class in college. In that case, you might have heard some absurd-sounding assumptions, such as the idea that everyone behaves rationally and is perfectly informed. It is not much of a surprise that this is not the case, and behavioural economics examines patterns of actual human behaviour.

Richard Thaler, an intellectual giant in the field, illustrated this by showing that when employees have to opt-out, rather than opt-in, to voluntary pension schemes, many of them end up with better pensions. The necessity of completing just one extra administrative act can lead people to economically inadequate situations. So, when trying to “nudge” people to participate in voluntary schemes, an opt-out system always works better, for example.

In the UK, the “Nudge Unit”, officially known as the Behavioural Insights Team (BIT), has tested some behavioural economics concepts. They have run trials of text messages sent to individuals owing fines to the UK HM Courts & Tribunals Service before the fine due date. A simple reminder doubled the payments received. Considering how much money the government spends on chasing overdue payments, this proved to be an extremely cost-effective method.

Another Nudge Unit intervention employed some gentle social shaming: in letters to taxpayers, it informed them that most people in their area pay on time. This simple statement increased tax collection – not by much, but the effect was there.

Economic incentives

Yet another area where social research informs policy is economic incentives. Excessive alcohol consumption is one such area in which this research is applicable. People tend to drink less when alcohol becomes pricey, but knowing the exact size of the effect – as well as its wider ramifications – is hugely important.

In economic terms, the demand elasticity of heavy drinkers is very low, as they try to maintain their consumption even when alcohol prices rise. This can lead to increased poverty levels, spillover into crime, and the use of illegal drugs as alcohol substitutes. The impact on occasional drinkers is smaller and comes with fewer side effects.

In 2018, Scotland introduced Minimum Unit Pricing (MUP) for alcohol, setting a base price per unit of alcohol. The University of Sheffield’s Alcohol Research Group conducted extensive modelling of alcohol consumption patterns, hospital admissions, and societal costs to assess its impact.

All of the above are examples of research aimed at making a better world – finding empirical effects within statistical noise and presenting them to policymakers for use. And sometimes, it happens. Let’s be glad for that.