Major project to create more effective economic policy

The programme ‘Makroekonomi och högdimensionell heterogenitet på mikronivån’ (The macroeconomy and high-dimensional heterogeneity on the micro level) is led by Mikael Carlsson, Professor of Economics at Uppsala University. It is a collaboration with the University of Copenhagen, Oslo University and the Norwegian Business School in Oslo that will run for eight years.

The researchers are to study how differences in behaviour between households and companies interact with economic processes and outcomes at macro level. The research is based on a combination of economic theory, recently developed machine learning methods and the use of microsimulation models, and examines how individuals and companies have different levels of vulnerability to various types of stimulus measures.

“This includes propensity for consumption and to work, for example,” notes Carlsson.

Wealth differences have impact

Researchers have previously looked at how differences in wealth affect vulnerability to stimulus measurses. The amount of money you have in the bank and your creditworthiness affect aspects such as how much you consume if you receive financial assistance.

By using machine learning, the researchers now want to study other differences – and how these differences affect economic behaviour.

“In Scandinavia, we have excellent register data on both individuals and companies, including age, gender, education and profession. We will use machine learning as a clever method of pinpointing differences in large data sets.”

One question the project wants to answer is: How can we design an effective economic policy that exploits these differences?

“If, based on this, we can map out who is more or less sensitive to stimulus measures, that could be used to design economic policy when policymakers need to quickly stimulate the economy”, explains Carlsson.

Minimise pressure on state budget

One example is all the political investments made to kick-start consumption and keep companies afloat during the COVID-19 pandemic. On a global level, USD 9 million were put into stimulus packages.

“Our research results can be used to design fiscal incentives that minimise the burden on the state budget by stimulating the houesholds and companies that are most sensitive to stimulus measures. In light of the huge stimulus packages seen in recent times, even small improvements in efficiency could entail very large societal gains,” adds Carlsson.

Annica Hulth