Policies should never be implemented without a clear idea as to what they are supposed to achieve. This clear idea tends to be an underlying model of the world which suggests that this policy intervention will lead to a desired outcome. That requires deep analyses, fundamental research, and an objective look at the problem. Without doubt, that’s the job of academia.

Politicians then have a tough job, because most interventions suggested by academic researchers come from stylized views of the world that ignore potential side effects of the intervention. These side effects are sometimes positive, sometimes negative impacts on special interest groups.

Let’s take the economists’ price on carbon as an example. The idea that the correct price on carbon would save us from climate change stems from a highly stylized model of the world, the model of public goods and externalities. In that model the price on carbon works perfectly. In practice, specific interest groups such as the ‘gilets jaunes’ disagree strongly. It’s then the problem of the politicians to figure out how much importance they should place on these special interest groups and thus how far the policy will shift away from the economists’ first best suggestion. What tends to be forgotten is that this process is an entirely political one.

Is that a reason for economists to ignore this political process? Definitely not. If a suggested policy works wonders in a lab setting but in practice is not implementable, then economists have to understand that this is a limitation and it should be controlled for in their models.

At the same time, politicians cannot ask economists to do their job for them. There is a reason for which economists are economists and not politicians. Economists should give politicians an objective view of the world whereas politicians have to juggle around and somehow find ways to accomodate the subjective positions of those that are affected by the policies.

So where does this lead us in the debate on the policy-academia gap? Politicians have to understand better that economists are well equiped to tell them about the best solution and they need this information. That information is the reference point, the status quo from which any concessions made to interest groups will be evaluated upon. If the economist would already include some concessions into the model, then this would change the reference point (most likely reduce the bar) and thus the costs of giving in to some special interest groups would seem lower. This would only drive the policy that finally is implemented even further away from the optimum and thus lead to higher social costs.

Economists should accept that the ideal solution is very unlikely to be implemented, but instead help politicians in evaluating the costs of giving in to some interest groups. In that way, the politicians have some guidance as to how costly these concessions are and it would help to choose between various concessions to different groups.

Very often though the political program has its own unpredictable dynamics. Due to this it is of utmost importance that policies not only get evaluated before but also after they have been implemented. There is little to none overall ex post evaluation in Europe and it would be a good idea that politicians give incentives for outside reviews of the policies that they introduced.