Stabilizing Macroeconomic Shocks: Experiments on the Interaction between Central Bank and Private Sector – StabEX

From a methodological point of view, our approach is novel, as laboratory experiments on macroeconomic models – which consists in applying the tools of experimental economics to macroeconomic topics – have hardly been used so far. Indeed, the assumptions and predictions of macroeconomic models have historically been tested using field data, giving rise to problems of reverse causality and identification. The alternative empirical approach that we followed uses laboratory experiments with controlled environments (thus, excluding the potential of reversed causality) and subjects paid according to incentive schemes that follow the theory to be tested. Since subjects’ information and actions can be directly monitored by the experimenter, identification is no issue. This approach has recently attracted increasing attention. It still shares some problems with pure theory, in particular that connotation of real economic variables may exert psychological effects on decision makers that are not accounted for by incentive schemes. Another potential problem is that the scale of macroeconomic phenomena is often perceived as being too vast to be tested by laboratory experiments with only a limited number of subjects – implying that one needs to be aware of the composition paradox. Hence, we need to be cautious with respect to the generality of experimental results in an aggregate macroeconomic framework. Nevertheless, experiments on targeted fields of macroeconomics can help evaluate individual and aggregate predictions of micro-founded macroeconomic models. Indeed, the use of micro-foundations in modern macroeconomics allows programming economies in which subjects take the role of various economic agents and testing the microeconomic modules that these models are composed of. Among this literature, a growing but yet relatively limited part of research papers addresses monetary policy and central banking issues.

The StabEX project contributes to this strand of the literature. Its aim has been to develop experimentation in macroeconomics as a tool for central bankers for bench testing alternative policy measures or rules. In fact, field experiments are difficult to pursue in central banking. Moreover, endogeneity of policy in the real world makes the analysis of real data and the formulation of correct inferences on changes that occurred complicated. By contrast, in the laboratory, the experimenter has the ability to control all the parameters of the tested model. Experiments allow to elucidate and approximate the different effects – anticipated and unanticipated – of alternative policy regimes or measures by comparing different treatments and offer a quick and cost effective way to identify possible consequences of a monetary policy initiative. Using this methodology, we show for example that a central bank that aims at stabilizing both inflation and output against shocks by communicating its inflation target can indeed reduce the volatility of inflation, interest rates and output, while leaving the level of these variables unaffected. The relevance of the target announcement is due to the reduction in subjects’ uncertainty about these objectives and to the coordination of different subjects’ expectations.

The StabEX project is a fundamental research project coordinated by Camille Cornand on the French side and Frank Heinemann on the German side. It associates the GATE L-SE laboratory and the Chair of macroeconomics at the Technische Universität Berlin. The project started in March 2013 and lasted for 48 months. It benefited from an ANR grant that amounted to 49000 €.

1) Camille Cornand and Cheick M’baye (2016a), «Does inflation targeting matter? An experimental investigation«, forthcoming in Macroeconomic Dynamics.

2) Camille Cornand and Cheick M’baye (2016a), «Band or Point Inflation Targeting? An Experimental Approach«, accepted for publication in Journal of Economic Interaction and Coordination.

3) Camille Cornand (2016), «Appraising sticky prices, sticky information and limited higher order beliefs in light of experimental data«, Festschrift in honor of Prof. G. Illing's 60th birthday, forthcoming, Springer.

4) Camille Cornand and Romain Baeriswyl (2016), «The predominant role of signals precision in experimental beauty contest«, B.E. Journal of Theoretical Economics, Research article, vol. 16, n°1, p.267–301.

5) Camille Cornand and Frank Heinemann (2015a), «Macro-expérimentation autour des fonctions des banques centrales«, Revue Française d’Economie, vol. XXX, n°2, p. 59-103.

6) Camille Cornand and Frank Heinemann (2015b), «Limited higher order beliefs and the welfare effects of public information«, Journal of Economic Studies, Special issue: Macroeconomic Experiments (ed.: C. Noussair), vol. 42, n°6, p. 1-26.

7) Camille Cornand and Frank Heinemann (2014a), «Measuring agents’ overreaction to public information in games with strategic complementarities«, Experimental Economics, vol. 17, n°1, p. 61-77.

8) Camille Cornand and Frank Heinemann (2014b), «Experiments on monetary policy and central banking«, in “Experiments in Macroeconomics”, John Duffy (eds.), Vol. 17, Research in Experimental Economics, Emerald Group Publishing.

Macroeconomists often assume that fluctuations of inflation and employment are associated with social costs. A central bank can use monetary policy to contain fluctuations in a country’s inflation and employment rates. In an economy affected by supply shocks, however, the central bank faces a tradeoff between stabilizing inflation and employment. The private sector, on the other hand, can theoretically absorb supply shocks by proper adjustments of wages and prices. Private responses to supply shocks should be more efficient, because they can address asymmetric shocks and relieve the central bank from stabilizing employment, so that it can achieve an efficient stabilization of inflation. But, price and wage adjustments are associated with a coordination problem, because prices in different sectors are strategic complements and the optimal response to an exogenous shock depends on the responses of other private agents. Central banks have the ability of solving the coordination problem by affecting product demand via the interest rate. However, monetary policy can just address aggregate shocks and may crowd out private responses that are, however, necessary for convergence to equilibrium after asymmetric shocks.

We, thus, identify two sources of interactions:
- Strategic substitution between central bank’s stabilization policy and private sector’s reaction: there is a conflict between the central bank and the private sector, since active stabilization is costly, although both parties benefit from stabilizing employment.
- Strategic complementarity between actions of private sector’s agents: wage or price adjustments to macroeconomic shocks in one sector raise the incentives to adjust wages or prices in other sectors of the economy.

In this research project, we will use laboratory experiments to generate data on a game that follows a modern macro model. The advantage of this approach is that we can implement different policy regimes in treatments that are otherwise equal and, thus, isolate the effects of transparency, cheap talk, and commitments to rules. Our specific research question is which policy regimes are best-suited to resolve the conflict of interests between central bank and a decentralized private sector in stabilizing employment and minimize the welfare costs of exogenous shocks.