Enhancing Tax Compliance with Administrative Nudges: Experimental Insights from Indonesia

How to cite

Satyadini, Agung; Yogama, Eko; Samodra, Bagas; Sutantri, Octa; Dularif, Muh. 2025. "Enhancing Tax Compliance with Administrative Nudges: Experimental Insights from Indonesia." AEA RCT Registry. AEARCTR-0016876. https://doi.org/10.1257/rct.16876-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Agung Satyadini* , Eko Yogama , Bagas Samodra , Octa Sutantri , Muh Dularif

* Primary investigator

Registry ID
AEARCTR-0016876
DOI
https://doi.org/10.1257/rct.16876-1.1
First registered
Last updated
Status
In development
Trial dates
2025-10-01 – 2025-12-31
Countries
Private
JEL codes
H20, H24
Registry record
http://www.socialscienceregistry.org/trials/16876

Abstract

This study examines the effectiveness of administrative nudges in enhancing tax compliance among small businesses in Indonesia. Employing a field experimental approach, we assess the impact of low-cost, behaviourally informed interventions conveyed through administrative letters, which vary in their framing to include simplified information, deterrence-oriented messaging, and social norm appeals. Tax compliance is operationalised through observable filing and payment behaviours, allowing for a systematic evaluation of the causal effects of the interventions. The experimental design further enables an exploration of both sustained compliance outcomes and the dynamic patterns of taxpayer responses over time..

Intervention

The experimental design focused on improving small businesses' tax compliance, particularly related to responses to administrative nudges which affecting their tax monthly payment and tax return submission compliance.

This trial employs three treatment letters: (1) a simplification letter, which reduces complexity in the existing correspondence, improves visual design, and adopts a less formal style; (2) a deterrence letter, which emphasises administrative penalties for non-compliance, the potential application of computerised audits, and subsequent enforcement actions; and (3) a public goods provision letter, which highlights taxpayers’ contributions to the national budget, with particular reference to public expenditure. Supplementary informative flyers are appended to the literacy and public goods provision letters. The use of colour schemes and graphical elements in the literacy and social norm letters is informed by the literature on colour psychology, with the objective of enhancing cognitive and affective engagement.

For the length of the experiment, the treatments will be contrasted with a control group comprised of individuals who did not receive the letters. The treatment impact of each intervention will be determined by comparing the average outcomes of the intervention group to those of the control group. There will be no interaction testing between treatments.

Experimental design

This experimental design is structured to evaluate the effectiveness of taxpayer communication strategies through a carefully staged field trial. The intervention will be implemented in collaboration with Directorate General of Taxes (DGT) offices, with taxpayers randomly assigned to treatment and control groups to ensure comparability. The treatment consists of letters emphasising different behavioural prompts, allowing the trial to test which messages are most effective in encouraging compliance. By embedding randomisation at the taxpayer level, the design minimises bias and enhances the credibility of the causal inference that can be drawn from the results.

The trial proceeds in phases, beginning with a pilot in November 2025, during which approximately 16,000 letters will be distributed. This phase will primarily assess operational feasibility, logistics, and the responsiveness of taxpayers to the interventions, while also providing early insights into their impact. Longitudinal data collection will extend into early 2026, capturing taxpayer behaviour over multiple reporting cycles. To promote transparency and facilitate independent validation, the trial will be registered in the AEA RCT Registry, with de-identified data made available by the DGT on a restricted basis, ensuring both research integrity and strict confidentiality safeguards.

Design details

Randomization method
To ensure credible stratified randomisation, taxpayers with comparable baseline characteristics were grouped into strata. Within each stratum, cases were randomly allocated to one of the treatment groups or to the control group, thereby achieving a balanced distribution of baseline characteristics across groups. Randomisation was implemented in \texttt{Stata\textsuperscript{\textregistered}} using a random variable generator with randomly selected seeds. For stratified randomisation, the user-written command \texttt{randtreat} (version~1.4) was employed.
Randomization unit
regional tax office
Clusters
7 regional tax offices
Observations
16,000 taxpayers
Arms
4000 taxpayers control, 4000 taxpayers T2, 4000 taxpayers T2, 4000 taxpayers T3
Primary outcomes
Taxpayer's inbound communication (responses)
Secondary outcomes
Tax payment, tax filing

Cited by

No citation data available for this registration yet.

Keywords