Minimum Wage Policy: A Markov Decision Approach

9 Pages Posted: 14 Nov 2017

See all articles by Thanakorn Sornkaew

Thanakorn Sornkaew

Arnaud Cheron

University of Maine (France) - School of Law, Economics, and Business Administration

Date Written: October 13, 2017

Abstract

In this paper, the Markov decision process with multi-objective Q-learning algorithm is proposed for setting minimum wage. The policy objectives are maximizing economic growth and income equality, and minimizing unemployment. The indicators based on ILO 131 are selected to support these objectives. The states which are minimum wage rate, actions (% of rising), rewards based on OLS (Ordinary Least Square), 𝑄𝑄-values and 𝑆𝑆𝑄𝑄 -values are calculated to select appropriate action. Policy makers can define the weight vectors of objective s and define % of rising for the algorithm. The simulation which uses the data sourced by National Statistical Office, Bank of Thailand, and Ministry of Labour shows good results.

Keywords: Markov Decision Process, Multi-objective Q-learning Algorithm, Economic Growth, Income Equality, Unemployment, Ordinary Least Square

Suggested Citation

Sornkaew, Thanakorn and Cheron, Arnaud, Minimum Wage Policy: A Markov Decision Approach (October 13, 2017). Available at SSRN: https://ssrn.com/abstract=3069259 or http://dx.doi.org/10.2139/ssrn.3069259

Arnaud Cheron

University of Maine (France) - School of Law, Economics, and Business Administration

Le Mans
France

No contact information is available for Thanakorn Sornkaew

Register to save articles to
your library

Register

Paper statistics

Downloads
15
Abstract Views
185
PlumX Metrics