Back to History

Fields Changed

Registration

Field Before After
Last Published April 30, 2026 11:25 PM April 30, 2026 11:30 PM
Intervention (Public) Firms receive official letters from the Secretaría de Hacienda de Medellín reminding them of their obligation to file their 2024 Industry and Commerce Tax (ICA) declaration. Letters vary in content across four groups: (1) neutral reminder, (2) deterrence message, (3) social norm message, and (4) combined deterrence and social norm message. All messages are sent via email and certified mail in April 2026, providing firms with a 15-day deadline to file voluntarily before potential enforcement actions. Firms receive official letters from the Secretaría de Hacienda de Medellín reminding them of their obligation to file their 2024 Industry and Commerce Tax (ICA) declaration. Letters vary in content across four groups: (1) neutral reminder (control), (2) deterrence message highlighting penalties and enforcement, (3) social norm message emphasizing peer compliance rates, and (4) combined deterrence and social norm message. Due to administrative and operational considerations in identifying eligible non-filers, the study is implemented in two phases: Phase 1 (May 2026): n≈1,000 firms. Messages sent May 5, 2026 via email and certified mail, with a 15-day deadline to file voluntarily. Phase 2 (July-August 2026): Approximately 6,000 additional firms receive messages using identical protocols. All messages are official communications from the Tax Authority. Outcomes measured at 90 days post-intervention from administrative records.
Experimental Design (Public) Design: 2×2 factorial randomized controlled trial Sample: 4,000 firms in Medellín, Colombia that failed to file their 2024 Industry and Commerce Tax (ICA) declaration Randomization: Stratified random assignment by firm size (3 categories) × sector (4 categories) × location (3 zones) = 36 strata. Firms are randomly assigned with equal probability to one of four groups (n=1,000 per group). Treatment Arms: Group 1: Control message (neutral reminder) Group 2: Treatment A Group 3: Treatment B Group 4: Treatment A + Treatment B (combined) Intervention Timing: Treatment letters sent April 16, 2026 via email and certified mail from the Secretaría de Hacienda de Medellín Primary Outcome Measurement: Tax filing and payment behavior measured at 90 days post-intervention (June 1, 2026) using Tax Authority administrative records Follow-up Period: The study includes an observational follow-up phase (June-December 2026) tracking firms that did not comply during the initial experimental phase. This follow-up phase is non-experimental and analyzes responses to the Tax Authority's standard enforcement procedures. Analysis: Primary analysis compares filing and payment rates across the four groups using ordinary least squares regression with stratification fixed effects. The factorial design enables estimation of main effects for Treatment A and Treatment B, as well as their interaction effect to test for complementarity or substitution. Statistical Power: With 1,000 firms per group and assumed baseline compliance of 2-5%, the study is powered at 80% (α=0.05) to detect main effects as small as 1.25 percentage points and interaction effects as small as 1.77 percentage points. Design: 2×2 factorial randomized controlled trial testing two behavioral mechanisms in tax compliance messaging Sample: Firms in Medellín, Colombia that failed to file their 2024 Industry and Commerce Tax (ICA) declaration. Implementation in two phases due to administrative process of identifying eligible non-filers on a rolling basis: - Phase 1 (May 2026): n≈1,000 firms - Phase 2 (July-August 2026): n≈6,000 firms - Total: ≈7,000 firms Randomization: Stratified random assignment by taxpayer type (2 categories) × sector (4 categories) × firm size (4 quartiles) × firm age (4 groups). Each phase uses independent randomization. Firms are randomly assigned with equal probability to one of four groups within strata. Treatment Arms: - Group 1: Control message (neutral reminder) - Group 2: Treatment A - Group 3: Treatment B - Group 4: Treatment A + Treatment B (combined) Intervention Timing: - Phase 1: Messages sent May 5, 2026 - Phase 2: Messages sent July-August 2026 All messages sent via email and certified mail from Secretaría de Hacienda de Medellín. Firms receive a 15-day deadline to file voluntarily. Primary Outcome Measurement: Tax filing and payment behavior measured at 90 days post-intervention using Tax Authority administrative records Follow-up Period: Observational follow-up (through December 2026) tracks firms that did not file during the initial 90-day window. This phase analyzes responses to the Tax Authority's standard enforcement procedures using propensity score matching (non-experimental). Analysis: Primary analysis pools both phases with phase fixed effects, comparing filing and payment rates across the four groups using ordinary least squares regression with stratification controls. The factorial design enables estimation of main effects for Treatment A and Treatment B, as well as their interaction effect to test for complementarity or substitution. Statistical Power: - Phase 1 only (n≈1,000): Main effects MDE ≈2.5pp, Interaction MDE ≈3.5pp - Both phases combined (n≈7,000): Main effects MDE ≈1.1pp, Interaction MDE ≈1.5pp Calculations assume baseline filing rate of 3-5% and α=0.05 (two-tailed).
Randomization Method Method: Computerized stratified block randomization conducted by the Secretaría de Hacienda de Medellín Procedure: The Tax Authority will use a random number generator in statistical software (Stata or R) to assign firms to treatment groups. The randomization will be conducted on April 16, 2026 immediately before message dispatch using the following steps: Stratification: Firms are sorted into 36 strata based on: firm size (3 categories) × sector (4 categories) × location (3 geographic zones) Block randomization within strata: Within each stratum, firms are randomly assigned to one of four groups (G1, G2, G3, G4) with equal probability (0.25 each) using computer-generated random numbers Implementation: The Tax Authority executes the randomization algorithm and generates the treatment assignment list. Each firm receives a unique treatment code (1-4) which determines which message variant they receive. Documentation: The randomization seed and complete assignment list are stored securely by the Tax Authority. Researchers receive only anonymized data with treatment codes (no firm identifiers). Timing: Randomization occurs on the same day as message dispatch (April 16, 2026) to minimize time between assignment and treatment delivery. Quality checks: After randomization, balance checks will verify that observable characteristics (firm size, sector, location, prior compliance history, debt amounts) are statistically similar across the four treatment groups. Method: Computerized stratified block randomization conducted by the Secretaría de Hacienda de Medellín using R statistical software Procedure: The Tax Authority uses a random number generator to assign firms to treatment groups. Randomization is conducted on the message deployment date using the following steps: 1. Stratification: Firms are sorted into strata based on: - Taxpayer type (2 categories: natural person vs. legal entity) - Economic sector (4 categories: construction, restaurants, tourism/RNT, financial services) - Firm size (4 quartiles based on log of taxable base) - Firm age (4 groups: pre-2000, 2000-2009, 2010-2019, 2020 onwards) 2. Block randomization within strata: Within each stratum, firms are randomly assigned to one of four groups (1=Control, 2=Social Norm, 3=Deterrence, 4=Combined) with equal probability using computer-generated random numbers 3. Implementation: The Tax Authority executes the randomization algorithm and generates the treatment assignment list. Each firm receives a unique treatment code determining which message variant they receive. 4. Documentation: The randomization seed and complete assignment list are stored securely by the Tax Authority. Researchers receive only anonymized data with treatment codes (no firm identifiers). Fixed Seeds (for reproducibility): - Phase 1 (May 2026): seed = 12345 - Phase 2 (July-August 2026): seed = 123452 Timing: Randomization occurs on the same day as message dispatch to minimize time between assignment and treatment delivery. Unit of randomization: Individual firm (identified by NIT - Número de Identificación Tributaria) Level: Firm-level randomization. Each firm is independently assigned to one of the four treatment groups within its stratum. There is no group-level or cluster-level randomization. Each firm is randomized individually, even if multiple firms share the same address, ownership, or accountant. Stratification note: While randomization occurs at the firm level, it is stratified by firm characteristics to ensure balance across groups. Stratification does not constitute a separate level of randomization—all randomization occurs at the individual firm level within strata. Unit of analysis: Matches unit of randomization (firm-level). All outcome measures (filing, payment) are recorded at the firm level from administrative records. Phase 2 exclusions: Firms assigned in Phase 1 are explicitly excluded from Phase 2 randomization to prevent double-treatment. This is implemented by filtering the Phase 1 firm list (by NIT) from the Phase 2 eligible population before randomization. Quality checks: After randomization, balance checks verify that observable characteristics (firm size, sector, taxpayer type, age, prior compliance history, debt amounts) are statistically similar across the four treatment groups.
Randomization Unit Unit of randomization: Individual firm (identified by NIT - Número de Identificación Tributaria) Level: Firm-level randomization. Each of the 4,000 firms is independently assigned to one of the four treatment groups within its stratum. No cluster randomization: There is no group-level or cluster-level randomization. Each firm is randomized individually, even if multiple firms share the same address, ownership, or accountant. Stratification: While randomization occurs at the firm level, it is stratified by firm characteristics (size × sector × location) to ensure balance across groups. Stratification does not constitute a separate level of randomization—all randomization occurs at the individual firm level within strata. Unit of analysis: Matches unit of randomization (firm-level). All outcome measures (filing, payment) are recorded at the firm level from administrative records. Unit of randomization: Individual firm (identified by NIT - Número de Identificación Tributaria) Level: Firm-level randomization. Each firm is independently assigned to one of the four treatment groups within its stratum. No cluster randomization: There is no group-level or cluster-level randomization. Each firm is randomized individually, even if multiple firms share the same address, ownership, or accountant. Stratification: While randomization occurs at the firm level, it is stratified by firm characteristics (taxpayer type × sector × firm size × firm age) to ensure balance across groups. Stratification does not constitute a separate level of randomization—all randomization occurs at the individual firm level within strata. Unit of analysis: Matches unit of randomization (firm-level). All outcome measures (filing, payment) are recorded at the firm level from administrative records.
Planned Number of Clusters Not applicable - no cluster randomization. Randomization occurs at the individual firm level. Total sample: 4,000 firms. Not applicable - no cluster randomization. Randomization occurs at the individual firm level. Total sample: ≈7,000 firms across two phases.
Planned Number of Observations 4,000 firms (1,000 firms per treatment group) ≈7,000 firms across two phases (≈1,750 firms per treatment group)
Sample size (or number of clusters) by treatment arms Group 1 (Control): 1,000 firms Group 2 (Treatment A): 1,000 firms Group 3 (Treatment B): 1,000 firms Group 4 (Treatment A + B): 1,000 firms Total: 4,000 firms Total: ≈7,000 firms - Phase 1 (May 2026): ≈1,000 firms - Phase 2 (July-August 2026): ≈6,000 firms By treatment group (expected): - Group 1 (Control): ≈1,750 firms - Group 2 (Treatment A): ≈1,750 firms - Group 3 (Treatment B): ≈1,750 firms - Group 4 (Combined): ≈1,750 firms
Power calculation: Minimum Detectable Effect Size for Main Outcomes Assumptions: Sample size: 1,000 firms per group (4,000 total) Significance level: α = 0.05 (two-tailed) Power: 80% Baseline compliance rate: 2-5% (assumed 3.5% for calculations) Standard errors clustered at stratification level (36 strata) Minimum Detectable Effects: Primary Outcome: Tax Filing (Binary) Main effects (Treatment A alone, Treatment B alone): 1.25 percentage points Interaction effect (complementarity/crowding-out): 1.77 percentage points Unit: Percentage points change in filing rate Baseline: 3.5% filing rate Effect size as % of baseline: 36% increase for main effects Secondary Outcome: Tax Payment (Binary) Main effects: 1.25 percentage points Interaction effect: 1.77 percentage points Unit: Percentage points change in payment rate Baseline: 2-3% payment rate Secondary Outcome: Amount Paid (Continuous) Main effects: 0.14 standard deviations Interaction effect: 0.20 standard deviations Unit: Standard deviations of amount paid (COP) Note: Exact COP amounts depend on distribution skewness; will report both absolute amounts and standardized effects Power Calculation Method: Following Muralidharan, Romero & Wüthrich (2025) methodology for factorial designs: For main effects: MDE = 2.8 × √[(p(1-p)/n)] For interaction effects: MDE = 2.8 × √[2×p(1-p)/n] Where p = baseline compliance rate, n = sample size per group Interpretation: The study is powered to detect effect sizes consistent with the tax compliance literature. For reference: Hallsworth et al. (2017): 2-5pp effects in UK Castro & Scartascini (2015): 5pp effect in Argentina Del Carpio (2013): 10-20pp effects in Peru Brockmeyer et al. (2019): 3.4pp effect in Costa Rica Our MDEs (1.25pp for main effects, 1.77pp for interaction) are smaller than most published effects, providing adequate power to detect policy-relevant impacts. Assumptions: - Significance level: α = 0.05 (two-tailed) - Power: 80% - Baseline filing rate: 3-5% (assumed 3.5% for calculations) - Standard errors clustered at stratification level SCENARIO 1: Phase 1 Only (n≈1,000, ≈250 per group) Primary Outcome: Tax Filing (Binary) - Main effects (Treatment A alone, Treatment B alone): 2.5 percentage points - Interaction effect (complementarity/crowding-out): 3.5 percentage points - Unit: Percentage points change in filing rate - Baseline: 3.5% filing rate - Effect size as % of baseline: 71% increase for main effects Secondary Outcome: Tax Payment (Binary) - Main effects: 2.5 percentage points - Interaction effect: 3.5 percentage points - Unit: Percentage points change in payment rate - Baseline: 2-3% payment rate Secondary Outcome: Amount Paid (Continuous) - Main effects: 0.28 standard deviations - Interaction effect: 0.40 standard deviations - Unit: Standard deviations of amount paid (COP) SCENARIO 2: Both Phases Combined (n≈7,000, ≈1,750 per group) Primary Outcome: Tax Filing (Binary) - Main effects: 1.1 percentage points - Interaction effect: 1.5 percentage points - Unit: Percentage points change in filing rate - Baseline: 3.5% filing rate - Effect size as % of baseline: 31% increase for main effects Secondary Outcome: Tax Payment (Binary) - Main effects: 1.1 percentage points - Interaction effect: 1.5 percentage points - Unit: Percentage points change in payment rate - Baseline: 2-3% payment rate Secondary Outcome: Amount Paid (Continuous) - Main effects: 0.11 standard deviations - Interaction effect: 0.15 standard deviations - Unit: Standard deviations of amount paid (COP) Power Calculation Method: Following Muralidharan, Romero & Wüthrich (2025) methodology for factorial designs: For main effects: MDE = 2.8 × √[(p(1-p)/n)] For interaction effects: MDE = 2.8 × √[2×p(1-p)/n] Where p = baseline compliance rate, n = sample size per group Interpretation: The study is powered to detect effect sizes consistent with the tax compliance literature. For reference: - Hallsworth et al. (2017, UK): 2-5pp effects - Castro & Scartascini (2015, Argentina): 5pp effect - Del Carpio (2022, Peru): 10-20pp effects - Brockmeyer et al. (2019, Costa Rica): 3.4pp effect Scenario 1 MDEs (2.5pp main effects, 3.5pp interaction): Adequate power to detect effects typical in literature Scenario 2 MDEs (1.1pp main effects, 1.5pp interaction): Excellent power to detect policy-relevant impacts, smaller than most published effects
Intervention (Hidden) Phase 1 Treatment Messages (April 16, 2026): Firms are randomly assigned to one of four groups (n=1,000 each): Group 1 - Control (D=0, N=0): Neutral reminder letter stating that the firm failed to file their 2024 ICA declaration, legal references (Decree 350/2018, Articles 88-91), invitation to file within 15 days, filing instructions, and contact information for questions. Group 2 - Deterrence Only (D=1, N=0): Control message PLUS section titled "CONSECUENCIAS ECONÓMICAS DE NO DECLARAR" detailing: 5% monthly penalty on tax owed for late filing (Article 242, Acuerdo 066/2017; Article 351, Acuerdo 093/2023) 1.5% monthly interest on unpaid tax Minimum penalty of 10 UVT, maximum up to 200% of tax owed Potential coercive collection (cobro coactivo) with additional procedural costs Numerical example: "If your firm owes $1,000,000 in tax and files 3 months late, the penalty would be $150,000 (5% × 3 months), plus approximately $45,000 in interest, totaling $195,000 in avoidable costs" Group 3 - Social Norm Only (D=0, N=1): Control message PLUS section titled "CUMPLIMIENTO TRIBUTARIO EN SU SECTOR" stating: "[X]% of firms in sector [SECTOR_NAME] in Medellín have already filed their 2024 ICA declaration" (personalized by firm's CIIU sector) Benefits of compliance: contribute to Medellín's development, finance infrastructure/security/public services benefiting businesses, strengthen municipal tax system, consolidate as responsible firms committed to the city Specific uses of ICA revenue: road maintenance in commercial zones, security in business corridors, public space improvements in commercial areas, business development programs Invitation to "join the majority of compliant firms in your sector" Group 4 - Combined (D=1, N=1): Social norm section (condensed version) PLUS deterrence section (condensed version). Social norm presented first, followed by deterrence information. Message Delivery: Primary: Email to registered address from official @medellin.gov.co domain Secondary: Physical certified mail on official letterhead Subject: "Declaración Impuesto de Industria y Comercio 2024 - Secretaría de Hacienda" Signature: Gloria Shirley Hincapié Ospina, Líder Proyecto Fiscalización All messages include firm name, NIT, filing link, and contact information Randomization: Stratified by firm size (3 categories based on revenue) × sector (4 CIIU categories) × location (3 geographic zones) = 36 strata Conducted by Tax Authority on April 16, 2026 Letters sent same day as randomization Phase 2 - Observational Enforcement (June-December 2026): Firms that did not file during Phase 1 become eligible for formal enforcement notices (emplazamientos). The Tax Authority applies standard prioritization based on: Outstanding debt amount Firm size and payment capacity Administrative enforcement capacity Approximately 50% of non-compliant firms receive emplazamientos between June-August 2026. This phase is observational (non-randomized). We use propensity score matching to control for selection into enforcement, analyzing whether Phase 1 message type affects response to enforcement. Follow-up period (September-December 2026): Observe long-term compliance persistence, delayed responses, and payment behavior. Data Collection: All outcome data extracted from Tax Authority administrative records. No surveys or additional contact with firms beyond standard communications. PHASE 1 TREATMENT MESSAGES (May 5, 2026): Firms are randomly assigned to one of four groups (target n≈250 each, actual Wave 1 n=982 total: Control=285, Deterrence=232, Social Norm=252, Combined=213): GROUP 1 - CONTROL (D=0, N=0): Neutral reminder letter stating that the firm failed to file their 2024 ICA declaration, legal references (Decree 350/2018, Articles 88-91), invitation to file within 15 days, filing instructions, and contact information for questions. GROUP 2 - DETERRENCE ONLY (D=1, N=0): Control message PLUS section titled "CONSECUENCIAS ECONÓMICAS DE NO DECLARAR" detailing: - 5% monthly penalty on tax owed for late filing (Article 242, Acuerdo 066/2017; Article 351, Acuerdo 093/2023) - 1.5% monthly interest on unpaid tax - Minimum penalty of 10 UVT, maximum up to 200% of tax owed - Potential coercive collection (cobro coactivo) with additional procedural costs - Numerical example: "If your firm owes $4,000,000 in tax and files 3 months late, the penalty would be $600,000 (5% × 3 months), plus approximately $180,000 in interest, totaling $780,000 in avoidable costs" GROUP 3 - SOCIAL NORM ONLY (D=0, N=1): Control message PLUS section titled "CUMPLIMIENTO TRIBUTARIO EN EL DISTRITO" stating: - "77.19% of registered firms in Distrito de Medellín have already filed their 2024 ICA declaration" (city-wide statistic, not personalized by sector) - Benefits of compliance: contribute to Medellín's economic and social development, strengthen public finances, finance essential public goods and services, strengthen trust in tax system, consolidate as responsible firms committed to regional development- Invitation to "join the majority of compliant firms" GROUP 4 - COMBINED (D=1, N=1): Social norm section (CONDENSED: 77.19% statistic + 3 key benefits) PLUS deterrence section (CONDENSED: penalty rates + numerical example). Social norm presented first, followed by deterrence information. Total message length comparable to Groups 2 and 3 to avoid confounding by message length. MESSAGE DELIVERY: - Primary: Email to registered business address from official @medellin.gov.co domain - Secondary: Physical certified mail on official letterhead - Subject line: "Declaración Impuesto de Industria y Comercio 2024 - Secretaría de Hacienda" - Signature: Gloria Shirley Hincapié Ospina, Líder Proyecto Fiscalización - All messages include firm name, NIT, filing link, and contact information - Delivery date: May 5, 2026 - Deadline: 15 calendar days from receipt RANDOMIZATION (PHASE 1): - Stratified by: Taxpayer type (2: natural person vs. legal entity) × Sector (4: construction, restaurants, tourism/RNT, financial) × Firm size (4 quartiles based on log taxable base) × Firm age (4 groups: pre-2000, 2000-2009, 2010-2019, 2020+) - Conducted by Tax Authority on May 5, 2026 using fixed seed randomization (seed=12345) - Letters sent same day as randomization - One auditor assigned per treatment arm to minimize operational variation PHASE 2 (July-August 2026): Approximately 6,000 additional eligible firms deployed using: - Identical treatment messages (same four groups) - Independent stratified randomization (seed=67890) - Explicit exclusion of Phase 1 firms to prevent double-treatment - Same message delivery protocols - Same measurement protocol (90-day outcome window) OBSERVATIONAL FOLLOW-UP (Non-filers): Firms that did not file during the 90-day outcome window become eligible for formal enforcement notices (emplazamientos). The Tax Authority applies standard prioritization based on outstanding debt, firm size, and administrative capacity. This phase is observational (non-randomized). Propensity score matching will be used to analyze whether message type affects response to enforcement. DATA COLLECTION: All outcome data extracted from Tax Authority administrative records. No surveys or additional contact with firms beyond standard tax communications. IMPLEMENTATION RATIONALE: The two-phase implementation reflects administrative and operational constraints in the Tax Authority's process of identifying eligible non-filers. The Tax Authority identifies and processes non-compliant firms on a rolling basis as administrative records are updated. This phased approach is standard practice in tax compliance field experiments (e.g., Hallsworth et al., 2017; Brockmeyer et al., 2019) and does not compromise internal validity, as each phase constitutes an independent randomized experiment that can be pooled with phase fixed effects.
Back to top