Entrepreneurship at Odds: The Influence of Online Betting on Business Creation

The online betting market is still relatively new in Brazil, which makes it difficult to ensure a stable regulatory environment, effective enforcement, and widespread understanding of how the market works. It is also important to note that very few people are aware of the existing laws. With this in mind, we propose the following experiment. We aim to test whether knowledge of Brazilian legislation - which sets technical parameters for the Return-to-Player (RTP) - or information about the negative expected returns of online betting affects individuals’ willingness to invest in betting.

2025-09-18

2026-02-28

Firm closure, Intentions to start a business, and financial allocations (experiment)

Firm closure is a binary variable equal to one if the individual reports having owned a company that was closed, and zero otherwise. We will compare the prevalence of company closures between bettors and non-bettors. In addition, we will ask individuals why their company closed.
For the intentions to start a business, we rely on self-reported entrepreneurial aspirations. Respondents are asked whether they plan to open a business in the near future.
For financial allocations, we use the experimental arm of the survey, where participants are randomly assigned to receive information about the legal framework surrounding betting. We then measure reported intended allocation of discretionary income across categories, including betting, leisure, and investment. The outcome of interest is the share of income allocated to betting/saving/business/other goods or a dummy indicating a positive value allocated to betting/savings/business/other goods.

Debt, Savings, Financial Literacy, Risk Attitudes and Expected return from online betting.

Our secondary outcomes are directly related to the mechanisms we aim to test. For each mechanism, we include a specific question in the survey. To capture debt and saving behavior, we ask respondents whether they have saved any portion of their income in the past six months. To measure financial literacy, we use a standard World Bank question on financial knowledge. Finally, to assess risk attitudes, we include questions on the respondent’s risk profile and their expected return from bets.

We conduct a nationally representative online survey experiment in Brazil to study how exposure to online betting affects entrepreneurship and financial decision-making. The survey combines descriptive questions on betting behavior and entrepreneurship with a randomized information treatment.

Not available

Randomization was performed automatically by the Qualtrics platform at the individual level using computer-generated random numbers, ensuring equal probability of assignment to treatment and control groups.

Individual

No

3200 individuals

3200 individuals

1066 approximately

Our simulations, based on 1,000 Monte Carlo replications for each outcome, show that with 3,200 respondents, the design achieves approximately 80% empirical power to detect effects ranging from 0.08 to 0.11 in probability units, depending on the outcome. Specifically, the estimated empirical power and corresponding minimum detectable effects (MDE) are as follows: 0.795 power for High-value Bet (0.085), 0.790 power for High-value Business (0.105), 0.811 power for Low-value Bet (0.084), and 0.786 power for Low-value Business (0.112). These MDEs correspond to effects between 8 and 11 percentage points, which represent realistic magnitudes for behavioral responses in survey experiments of this type.