Skip to main content
Log in

A randomized intervention to gauge preferred tax rates and progressivity

  • Published:
International Tax and Public Finance Aims and scope Submit manuscript

Abstract

A key concern in high-inequality countries is whether greater tax revenue and progressivity could help to decrease inequality. Studies have shown that while people care about inequality, providing them with information about current levels of inequality or about specific policies has little impact on their preferences for redistribution. These preferences have usually been proxied by their preferences for government spending without any reference to taxation. In this study in Mexico, we take a novel approach by carrying out a randomized intervention in which we provide information related to (i) corruption, (ii) public health, or (iii) public safety. We then ask individuals about their willingness to pay higher taxes, and we ascertain their support for tax progressivity by asking about their preferred tax rates for the rich, the middle-income, and the poor. We find that willingness to pay higher taxes is affected only by the information on corruption and public health (2.56 and 2.89% points in additional taxes, respectively). We find that this information does not increase their preference for tax progressivity.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+
from €37.37 /Month
  • Starting from 10 chapters or articles per month
  • Access and download chapters and articles from more than 300k books and 2,500 journals
  • Cancel anytime
View plans

Buy Now

Price includes VAT (Netherlands)

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Data availability

Data is available upon request.

Notes

  1. Campos-Vazquez et al. (2018) and del Rovira (2017) account for the underrepresentation of the rich and their income in household surveys and adjust that data using disposable income from national accounts. They estimate a Gini coefficient of around 0.70.

  2. We built and embedded our experiment on an app and then ran it on Prolific. For a more detailed discussion of this platform see Peer et al. (2017). Previous studies using this platform include: Wiwad et al. (2021), which is especially relevant to our study; Khenfer et al. (2020); MacGregor et al. (2020); Moynihan et al. (2018); Schild et al. (2019); Shepherd and Fitzsimons (2020); and Stanovich and Toplak (2019).

  3. This is a step before consideration of tax compliance, and it is more aligned with the nature of our study. It remains an open empirical question whether the mechanisms affecting the willingness to pay taxes are the same as those explaining tax compliance.

  4. It is also worth mentioning that recent non-experimental research has found that individuals prefer higher tax progressivity but lower rates of taxation (Barnes, 2015; Berens & Gelepithis, 2019; Roosma et al., 2016).

  5. Prolific is a crowdsourcing platform for recruiting human subjects for behavioral research that enables large-scale data collection by connecting researchers and participants worldwide. It is similar to Amazon’s Mechanical Turk: both platforms allow requesters or researchers to offer monetary incentives to workers to perform cognitive jobs. However, Prolific offers fewer restrictions on the types of tasks that researchers can request from participants, more transparent information regarding its pool of workers, and it promotes an environment of ethical payments. For more details, see Peer et al. (2017).

  6. We have data for 1244 individuals. However, some of them did not answer the questions about sex (n = 6), educational level (n = 1), or household income in the previous quarter (n = 3). We exclude these individuals from our analysis.

  7. To further enhance the validity of our results, we implemented additional measures beyond those provided by the Prolific platform. To encourage participants to engage attentively, we incorporated brief messages prompting them to answer all questions and backtrack if necessary. Additionally, we minimized potential psychological biases by designing sliders that mitigated anchoring effects. To gauge the consistency of responses, we included questions aimed at this purpose. Moreover, our experiment was designed for seamless accessibility across devices (e.g., computers, cellphones, tablets), with mechanisms in place to prevent duplicate participation. Once a participant completed the experiment on a device, they were unable to access it again from the same device, ensuring the exclusion of potential robot participation or multiple entries from the same individual using different accounts on Prolific.

  8. We divide the sample into the following regions: South: Campeche, Chiapas, Guerrero, Oaxaca, Quintana Roo, Tabasco, Veracruz, and Yucatán. Center: Aguascalientes, Colima, Estado de México, Guanajuato, Hidalgo, Jalisco, Michoacán, Morelos, Puebla, Querétaro, San Luis Potosí, and Tlaxcala. North: Baja California, Baja California Sur, Chihuahua, Coahuila, Durango, Nayarit, Nuevo León, Sinaloa, Sonora, Tamaulipas, and Zacatecas. We consider Mexico City as an independent region.

  9. The “high-income household” variable is constructed by asking people about their household income in the previous quarter according to specific ranges. If the person answered that their household had received more than MXN $30,000, they are included in this category.

  10. Our approach is inspired by the traditional models of preferences for redistribution and their focus on the median voter as the decisive voter (e.g., Meltzer & Richard, 1981) and echoes the recent empirical work of Campos-Vazquez et al. (2023).

References

Download references

Author information

Authors and Affiliations

Authors

Contributions

RC-V and SR-O designed the study, SR-O implemented the randomized intervention, RC-V and SR-O wrote and reviewed the manuscript. Both the authors read and approved the final manuscript.

Corresponding author

Correspondence to Raymundo M. Campos-Vazquez.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The content of this study does not necessarily reflect the opinions and positions of the World Bank or El Colegio de México. We especially thank an anonymous reviewer and the editor for their insightful comments and suggestions. All errors and omissions are the sole responsibility of the authors.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1 (download PDF )

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Campos-Vazquez, R.M., Restrepo-Oyola, S.D. A randomized intervention to gauge preferred tax rates and progressivity. Int Tax Public Finance 32, 782–804 (2025). https://doi.org/10.1007/s10797-024-09852-x

Download citation

  • Accepted:

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1007/s10797-024-09852-x

Keywords

JEL

Profiles

  1. Raymundo M. Campos-Vazquez