Learning to believe in simple equilibria in a complex OLG economy - evidence from the lab

Abstract

We set up a laboratory experiment within the overlapping-generations model of Grandmont (1985). Under perfect foresight this model displays innitely many equilibria: a steady state, periodic as well as chaotic equilibria. Moreover, there exists some learning theory predicting convergence to each of these equilibria. We use experimental evidence as an equilibrium selection device in this complex OLG economy, and investigate on which outcomes subjects most likely coordinate. We use two alternative experimental designs: learning-to-forecast, in which subjects predict the future price of the good, and learning-to-optimize, in which subjects make savings decision. We nd that coordination on a steady state or 2-cycle are the only outcomes in this complex environment. In the learning-to-forecast design, coordination on a 2-cycle occurs frequently, even in the chaotic parameter range. Simulations of a behavioral heuristic switching model result in initial coordination on a simple AR(1) rule though sample autocorrelation learning, with subsequent coordination on a simple second-order adaptive rule once the up-and-down pattern of prices has been learned. JEL codes: C90, D83, C62.

What to target? Insights from a lab experiment

“…In particular, if subjects use some form of adaptive expectations involving only the information of period t − 1, the target is unstable under AIT with more than two lags. This is particularly interesting because we know from previous learning-to-forecast experiments that subjects typically only look at a few lags, at best up to two, to form their expectations (see, e.g., Arifovic et al 2019). Whether seven independent subjects may coordinate on a forecasting rule involving more lags in the context of our experiment remains a policy-relevant but open question that Tr.…”

“…First, three lags are sufficient to nest the rational forecasts based on the MSV solution of each treatment. Second, previous forecasting experiments have demonstrated that subjects tend to quickly discard past information, and using information beyond two lags is not common (Arifovic et al, 2019).…”

Expectation Formation in Finance and Macroeconomics: A Review of New Experimental Evidence

“…(5) Traditionally, LtFEs mainly study questions related to asset pricing. Recent LtFEs pay more attention to monetary economics and the role of monetary policy in asset markets (e.g., Arifovic and Petersen, 2017;Arifovic et al, 2019;Assenza et al, 2019;Bao and Zong, 2019;Hommes et al, 2019a,b;Mauersberger, 2019;Ahrens et al, 2019). The findings of this literature show that higher interest rates and central bank communication are useful in stabilizing asset prices or inflation rates.…”

Behavioral and Experimental Macroeconomics and Policy Analysis: A Complex Systems Approach

“… Bernasconi and Kirchkamp (2000) combine the learning-to-forecast and learning-to-optimize designs in an OG experiment; see alsoArifovic, Hommes & Salle (2018).ECB Working Paper Series No 2201 / November 2018…”