Understanding Expectations to Design Better Policies

Expectations about economic variables are a key driver of how households and firms react to changes in the economic environment, including policy interventions. When a central bank cuts interest rates, for instance, the economy will expand only if businesses and households expect the rates to remain low for some time. Only in this case, in fact, will they borrow, spend, and invest more, thus contributing to economic growth. If, on the other hand, they expect interest rates to rise again soon, they will not react to the cut and the expansion will not materialize.

Luigi Iovino, Associate Professor of Economics at Bocconi, obtained a €1.5mln ERC Consolidator Grant from the European Research Council for GEM (Great Expectations: Macroeconomic Implications of Forecasting Behavior), a research project aimed at advancing our understanding of expectations formation. Somewhat unexpectedly, the project could also be providing central banks with new tools to fine-tune the green transition.

Macroeconomic models have traditionally assumed that agents enjoy both full information and rational expectations (the FIRE paradigm), in the sense that they can correctly predict the implications of any change in the relevant variables. "The recent collection of extensive survey data has, however, led scholars to question the validity of the FIRE paradigm," Professor Iovino said. "Information frictions and limited rationality seem to both be present in the data. These powerful results have motivated empiricists and theorists alike to construct plausible alternatives to the FIRE paradigm."

In the first of the three parts of the project, Iovino will challenge recent theories, according to which expectations react sluggishly to shocks, while the opposite would be expected. Such a slow reaction may depend on the state of the economy or on the expectations' time horizon. Most of the data was gathered in developed economies, under stable economic conditions, and regards short-term expectations. Novel datasets including developing countries and long-term expectations could help solve the puzzle.

Agents' expectations also depend on the models they use to understand the economy. For example, a sophisticated agent understands that interest rates affect inflation through several channels (e.g., borrowing cost for firms and household demand). Such an agent will thus have more accurate expectations of the effects of an interest rate cut compared to a less sophisticated agent, who interprets the interest-rate policy through the lens of a simplified model with fewer channels. In the second part of the project, Iovino aims to provide policymakers with a better understanding of the subjective models that agents use to interpret policy interventions.

An interesting observation, at the center of the third part of the project, is that firms are not equally sensitive to changes in interest rates. Those that invest in projects that pay off in the distant future (i.e., "high-duration" projects) tend to be more exposed to variations in interest rates. Furthermore, a preliminary analysis shows that interest-rate-sensitive firms involved in high-duration projects are also greener. There seems to be a strong negative correlation between carbon intensity and duration. Intuitively, dirty firms (e.g., utilities) are "low-duration" because they tend to be older, with lower growth opportunities, and a cash-flow profile tilted towards the present, while the opposite is true for greener, innovative firms. "If these results are confirmed," Iovino concluded, "it means that central banks can fine-tune the green transition by adjusting interest rates. It does not only mean that they can cut them to stimulate the transition; they can also raise interest rates when the transition is too fast and causes, for instance, a large employment loss in brown industries that has not yet been balanced by a gain in employment in green businesses."

ERC Consolidator Grants are assigned to researchers with 7-12 years of experience since completion of their PhD, a scientific track record showing great promise and an excellent research proposal. Younger researchers can apply for a Starting Grant and more experienced scientists can bid for an Advanced Grant.