Real-effort laboratory experiment on the effects of production uncertainty on agent effort allocation among two inputs.

Participants use a web browser to engage in addition tasks while a proctor controls the flow of the experiment in an administrative dashboard. The basic innovation is to present participants with a choice of two possible tasks (inputs). In the allowed time, they attempt to successfully complete either task as many times as possible. The two tasks have different difficulty and financial payoffs.

Participants use a web browser to engage in the tasks while a proctor controls the flow of the experiment in an administrative dashboard. The basic innovation is to present participants with a choice of two possible tasks (inputs). In the allowed time, they attempt to successfully complete either task as many times as possible. The two tasks have different difficulty and financial payoffs.

Multi-task Session Description
Participants answer easy or hard addition problems. Easy questions require participants to add up three two-digit numbers, while hard questions require adding six two-digit numbers. There is no penalty for wrong answers, and participants can end a round at any time. Participants are told they can quietly visit other websites while they wait for the next section to begin, which is intended to simulate a leisure outside option. Participants can also leave the experiment if they finish early.

The activity is broken into three main sections -- Fixed Wage, Treatment (Production Uncertainty), and Control (Input Incentive) -- each consisting of a set of rounds. Sections have different payment schemes, round lengths, and number of rounds. During each round, participants can see their time remaining and how many questions of each type they have attempted. They do not see their results until the end of the round where a summary table displays all available information on questions attempted, time to completion, and payoff per question.

Part 1: Tutorial
To ensure that all participants understand the mechanics of the activity, they are led on an interactive guided tour. The tour individually highlights each element of the display and requires participants select questions and answer them for practice.

Part 2: Fixed Wage
As an introductory session, participants complete a Fixed Wage session. Here participants are told they will earn a fixed wage regardless of their performance. This session consists of three 3-minute rounds. Participants are also told that for this session, answering easy and hard questions will be valuable to the researchers and that hard questions are even more valuable. This is intended to provide non-monetary incentives to do well. This also is intended to imitate the vague notion that harder questions are more productive than easy questions.

Part 3: Treatment (Production Uncertainty) and Control (Input Incentive)
In the final two sessions, participants are subject to two different incentive schemes that imitate employment contracts with and without production uncertainty. Each session consists of eight 3-minute rounds. In order to compare an individual's performance between both contract types, all participants receive both treatments. The order in which the treatments are administered is randomized across participants to account for the possibility that participants apply information gained in the first section to the second (sequence effects).

In these rounds, participants are paid per successful completion of easy and hard questions. In addition, participants are paid a small amount for each minute remaining on the clock at the end of a round. This fixes a monetary value to leisure time and acts to assure participants that ending a round early is acceptable. The ability to visit other websites helps make leisure time less boring.

Treatment: Production Uncertainty -- In the treatment session, the payoff amount per question type is not known until the end of each round when it is randomly drawn. This simulates an output-based incentive where the marginal payoff of each input is known only after the output is measured. Because the distribution of payoffs remains constant across rounds, participants can optimally mix their inputs between easy and hard tasks as described in Equation~\ref{eq:simplified}. Payoff coefficients are never negative, and participants are informed of the range of possible payoff values.

Control: Input Incentives -- In the control session, the value of easy and hard questions is displayed prominently at the top of the screen. This simulates an input-based incentive where the marginal payoff is known precisely. Participants should optimize their earnings by identifying which input has the highest marginal productivity and dedicating all their time to this task. Specifically, participants identify their earnings per minute for easy and hard tasks, and then dedicate all their time to the task with the highest earnings per minute.

2016-10-17

2018-08-01

I measure how participants allocate their time between the two inputs. Based on their performance, I know which input is most effective for a participant. I can then see if the treatment (production uncertainty) induces them to inefficiently switch away from their effective input.

Each participant is presented both a control session and a treatment session (in randomized order). In treatment, the payoffs for each input are randomized. There are two treatment arms: one in which the easy question's payoffs have high variance, and the other in which the hard question's payoffs have high variance.

Randomization done by a computer at the beginning of each laboratory session. Randomization ensures equal split between treatment arms and the control/treatment session order. Math problems are randomly generated prior to the experiment so that all participants have the same problems.

Randomized at the individual level within each experimental session. No clustering required because all participants experience treatment and control.

No

60 individuals.

60 individuals.

30 with treatment Easy High Variance
30 with treatment Hard High Variance
All 60 participants have control and treatment.