The researcher -Felipe Aguiar- is at the front of the participants giving instructions during the experiment, and two assistants are standing at the right side of the computer laboratory.

Abstract

This research project tests Laffer's economic theory (1981) and contributes to scientific knowledge with new evidence. Based on a controlled laboratory experiment, it was possible to cover an economic problem, applying a different methodology to the traditional study scheme. The experiment was designed and executed, following the guidelines of experiments and previous scientific contributions. It was worked with a sample of economically active subjects. A software for the experiment was designed, and a post-experiment survey was conducted. The results obtained corroborate that the expected behavior of individuals, in the arithmetic and economic effect of the Laffer phenomenon, is not met and ratify the guidelines of other authors. Tax rates of 12%, 28%, 50%, 73% and 87% were used. Average total tax revenue was $ 1.05 at the 12% tax rate, while at the 87% rate it was $ 7.36. Contrary to what Laffer's theory points out, subjects paid their taxes correctly and the production of their work increased even in the face of high tax rates. Average total production was 716.00 ‘!’ keys entered at the 12% tax rate, while at the 87% rate it was 836.00 ‘!’ keys entered. It was found that certain individual behavioral factors can have an impact on the incentives that subjects have related to their work and to paying taxes.

Keywords: Experimental Economics, Behavioral Economics, Laffer curve, Tax Pressure, Tax Revenue and Labor.

Introduction

The Laffer curve is the most common way to explain the relationship between tax rates and tax revenues. The basic idea behind the relationship between tax rates and tax revenue is that changes in tax rates have two effects on revenue: the arithmetic effect and the economic effect (Laffer, 2004). The arithmetic effect indicates that a percentage deduction in the tax rate will reduce tax revenues by the same percentage and in the same way an increase in the tax rate.

On the other hand, the economic effect recognizes the positive impact that low tax rates have on work, performance, and employment, by providing incentives to increase these activities. Therefore, an increase in tax rates has an opposite effect on the behavior of agents and discourages work, performance, and employment. However, these two effects are very far from reality and use extreme assumptions to validate its theoretic point. This is the main reason why several authors have criticized Laffer’s theory. For example, according to the economist Hal Varian (1993), for the economic effect to be fulfilled even “[…] in the simplest model, a marginal tax rate of 50% requires an elasticity of labor supply of 1 to obtain the Laffer effect"(Varian, 1993, p. 7). In this research project, I tested the Laffer curve and its implications through an economical experiment.

Research questions

The general research question of this project was:

1. What is the behavior of taxpayers in the face of changes in the tax pressure, is it possible to obtain evidence that proves the existence of a Laffer curve phenomenon in a controlled experiment?

The specific research questions were:

1. Is the arithmetic effect, the expected behavior of taxpayers in the economic effect, and the characteristic shape of the Laffer curve fulfilled, in the face of variations in the tax pressure?

2. How does the labor supply adjust to changes in tax pressure?

3. How do after-tax income needs, leisure preferences over work, expectations of future well-being, and anticipated well-being of tax redistributions have an impact on taxpayer incentives and tax payments?

Experimental Methodology

The experimental design of the present investigation replicated the computerized task proposed by Swenson (1988) in a redistributive state. The experiment was carried out on nineteen postgraduate students of the Master's degree in Developmental Economic Research, from the Latin American Faculty of Social Sciences (FLACSO, Ecuador headquarters). The experiment took place on December 5th, 2019, in FLACSO Computer Laboratory, with a total duration of forty minutes.

Graph 2: Picture taken in the Computer Laboratory the day of the Experiment.

The experimental design consisted of two stages. In the first stage, a computerized task was carried out and in the second section, a post-experiment survey was made. Instructions were read to the participants and a preliminary round of practice was carried out before carrying out the experiment. As you can see in Graph 3, to ensure participants' understanding, the instructions for the computerized task were shown at the beginning of the computerized program. The computerized task consisted of 10 work periods and 10 rest periods. Each work period lasted one minute and was followed by one minute rest period.

Graph 3: Screenshot of the start of the Computerized Program.