Tax compliance among SMEs has been the subject of extensive scholarship, drawing on economic, institutional, and behavioural paradigms to explain why firms comply ( Kirchler , 2007 ). In developing economies such as Zimbabwe, SMEs straddle the formal-informal divide, complicating measurement and policy ( Kanyoka, 2020 ; Ndhlovu, 2021 ). Globally, SMEs account for the majority of private sector employment yet face similar burdens like complex regulations, capacity constraints, and low trust in authorities which undermine voluntary compliance ( Beck et al., 2019 ; Ayyagari & Beck, 2007 ). In sub-Saharan Africa, empirical studies highlight that bureaucratic complexity and uneven enforcement are principal deterrents to compliance ( Chinomona & Sandada, 2013 ; Munos & Santos, 2019 ). Zimbabwe-specific research, though limited, confirms persistent under-compliance among SMEs, driven by procedural barriers (e.g., TaRMS integration challenges) and behavioural factors such as low tax morale and digital illiteracy ( Chifamba & Mkhari, 2023 ; Mushoriwa, 2022 ).

This study adopts an integrative theoretical framework that synthesises:

Institutional Theory, which emphasises how regulative, normative, and cognitive pressures shape organizational behaviour ( Meyer & Rowan, 1977 ; Scott, 2008 ). In the tax context, regulative elements (e.g., statutes, penalties) and normative expectations (e.g., social legitimacy of ZIMRA) jointly influence SME compliance decisions.

Procedural Fairness Theory, positing that perceptions of fairness in administrative processes (neutrality, transparency) bolster voluntary cooperation ( Wenzel, 2002 ). Fair, accessible procedures (e.g., user-friendly TaRMS) can thus mitigate resistance.

Theory of Planned Behavior (TPB) and Behavioral Economics, which foreground individual attitudes, subjective norms, perceived behavioural control, and cognitive biases in shaping compliant actions ( Ajzen, 1991 ; Torgler, 2007 ). Tax morale, an intrinsic willingness to comply emerges from these psychosocial factors.

Together, these theoretical lenses provide a robust platform for examining how institutional legitimacy, procedural efficiency, and behavioural motivations intertwine to influence SME tax compliance.

Institutional drivers encompass the formal structures and perceived legitimacy of tax authorities in Zimbabwe.

Regulatory Architecture: Despite Zimbabwe’s Vision 2030 emphasis on a dynamic SME sector, the statutory framework is marred by overlapping levies and frequent policy shifts, creating ambiguity that raises compliance costs ( Government of Zimbabwe, 2018 ; Ministry of Finance Zimbabwe, 2021 ).

Enforcement Consistency: While ZIMRA’s Strategic Plan 2021 to 2025 and outreach initiatives (e.g., provincial kiosks, Block Management) signal commitment to modernisation, audits and sanctions remain unevenly applied, diminishing deterrence and fuelling perceptions of arbitrariness ( Mutizwa, 2021 ; Maponga, 2022 ).

Institutional Trust: Trust in ZIMRA is critical, when SMEs doubt that revenues are used transparently, their willingness to comply erodes ( Fei & Chimhowu, 2021 ; Dube & Savage, 2021 ). Enhancing transparency and stakeholder engagement is thus essential to strengthen the social contract.

Procedural drivers refer to the administrative processes that shape the taxpayer experience.

Complexity of Compliance: SMEs face a labyrinth of forms, sector-specific levies, and recurrent deadlines, all of which impose significant time and financial burdens ( Mutizwa, 2021 ; Eccleston, 2022 ).

Digitalisation via TaRMS and E-Invoicing: The rollout of TaRMS and mandatory e-invoicing promises real-time filing and error reduction ( ZIMRA, 2023 ), yet uneven internet connectivity, technical glitches, and limited training hamper uptake especially among micro-enterprises ( Mupambwa , 2022 ).

Service Delivery and Support: Initiatives such as SME desks and mobile kiosks aim to improve taxpayer services ( Dube & Ho, 2023 ), but staffing shortages, long queues and procedural delays at walk-in centres continue to frustrate SMEs and may erode confidence in administrative efficiency ( Chingonzo & Dylan, 2020 ).

Knowledge of tax obligations and incentives underpins voluntary compliance ( Moyo, 2022 ). However, many SMEs report inadequate access to clear, sector-specific guidance and uneven outreach effectiveness ( Ndhlovu, 2021 ; ZIMFIRS, 2020 ). Despite ZIMRA workshops and online tutorials, micro-enterprises often remain unaware of filing thresholds, allowable deductions, and penalty structures, which exacerbates non-compliance ( Torgler, 2007 ). Tailored information campaigns and integration of interactive help feature into TaRMS could substantially reduce these informational barriers.

Behavioural drivers capture psychological and social influences on compliance choices.

Taxpayer Education and Literacy: Targeted workshops and digital tutorials enhance SMEs’ understanding of obligations, incentives and deadlines; higher financial literacy correlates with more accurate and timely filings ( Moyo, 2022 ; Ndhlovu, 2021 ).

Perceptions of Fairness: Distributive fairness (perceived equity of tax burdens) and procedural fairness (transparency of processes) heavily influence voluntary compliance ( Alm, McClelland, & Schulze, 1992 ; Wenzel, 2002 ). Zimbabwean SMEs often perceive that larger firms receive preferential treatment, weakening trust in ZIMRA ( Dube & Savage, 2021 ).

Social Norms and Tax Morale: In contexts where informal practices predominate, peer norms can normalize under-reporting, sustaining a culture of non-compliance ( ZIMFIRS, 2020 ; Fei & Chimhowu, 2021 ). Elevating positive role models and publicizing compliance successes may strengthen collective morale.

Although prior studies elucidate individual driver categories, they too often examine them in isolation. Three critical gaps emerge: 1) limited exploration of interactions among institutional, procedural, and behavioural factors; 2) scarce empirical data focused specifically on Harare’s SME landscape; and 3) a dearth of integrated, multi-theoretical models tested in Zimbabwean contexts ( Chichava & Gumbo, 2023 ; Mhlanga & Tabitha, 2022 ). To address these gaps, the present study examines how institutional, procedural, informational and behavioural barriers collectively relate to SME tax non-compliance in Harare.

A descriptive, cross-sectional survey of 75 SMEs operating in Harare was used, an approach appropriate for mapping patterns and constraints at a single point in time in resource-constrained contexts ( Creswell & Creswell, 2018 ; Saunders, Lewis, & Thornhill, 2023 ). The target population covered firms in retail, services, manufacturing, ICT, and other activities to reflect the city’s SME mix. The study aimed for broad sectoral coverage rather than statistical representativeness, drawing respondents from busy trading clusters, industrial parks, and business association networks to ensure a practical spread of firm types and sizes. Inclusion criteria were: currently operating in Harare at the time of data collection and availability of an owner, manager, or accounting or administration officer able to complete the questionnaire. Firms that were temporarily closed or without a responsible person present were not included. Participation was voluntary, and respondents gave informed consent. No personally identifying information was collected beyond general firm characteristics.

For screening and eligibility purposes, we defined an SME as an independently owned and managed enterprise employing 1 to 100 full-time equivalent (FTE) workers and or reporting annual turnover typical of small and medium firms (turnover plus assets less than $1 million), excluding state-owned entities and NGOs. We included micro (1 to 9 FTE), small (10 to 49 FTE), and medium (50 to 100 FTE) firms provided they were actively trading in Harare and an owner or manager or accounting or administration officer could respond to the instrument. This operationalisation follows widely used international SME conventions while allowing for local turnover variability ( OECD, 2019 ; World Bank, 2021 ).

We used a non-probabilistic approach to achieve broad sectoral coverage in a context where 1) no reliable, up-to-date SME sampling frame exists for Harare; 2) many micro and small operators are spatially dispersed and mobile; and 3) access to owners is time-bound and irregular due to variable operating hours and gatekeeper constraints. Intercepting eligible firms within known trading clusters, industrial sites, and association networks is a recognised, practical strategy for hard-to-reach business populations and exploratory mapping ( Etikan , Musa, & Alkassim, 2016 ; Palinkas et al., 2015 ; Saunders et al., 2023 ). This approach suits our descriptive objective; surface patterns and constraints across institutional, procedural, informational, and behavioural domains without making population-level inferential claims.

We screened respondents using the SME definition specified in Section 3.1 (1 to 100 FTEs and or typical SME turnover) and confirmed active trading within Harare before administering the questionnaire.

We used a structured questionnaire organised into eight sections that mirror the study’s domains: firmographics; structural factors; procedural/administrative complexity; information and education (including two knowledge checks on VAT filing frequency and the purpose of presumptive tax); perceptions of trust and fairness; digital access and technology (including TaRMS); self-reported compliance behaviours; and open-ended suggestions. Most attitudinal items used a five-point Likert-type scale (Strongly Disagree to Strongly Agree), consistent with best practice in survey measurement ( Joshi, Kale, Chandel, & Pal, 2015 ). We piloted the instrument with 15 SMEs for clarity and flow and made minor wording adjustments, reflecting standard recommendations for pretesting survey tools ( Presser et al., 2004 ). Questionnaires were completed either self-administered or with enumerator assistance. We emphasised confidentiality and neutrality to minimise social-desirability bias common in compliance-related surveys ( Grimm, 2010 ; Krumpal, 2013 ).

Analysis focused on descriptive statistics, consistent with the study’s design and purpose in exploratory settings ( Hair, Black, Babin, & Anderson, 2019 ). We report frequencies and percentages for categorical and Likert-type items and used simple cross-tabulations selectively to illuminate patterns (e.g., registration status and reported behaviours) without making inferential claims. For clarity in the narrative, some Likert responses were summarised as “agree/strongly agree” vs other categories when presenting headline proportions ( Joshi et al., 2015 ). Missing responses were minimal; item-level percentages were calculated using the non-missing denominator for each item. Open-ended answers were reviewed using brief thematic coding to identify recurring suggestions (e.g., digital reliability, training needs, clearer communication), following a pragmatic reflexive thematic analysis orientation ( Braun & Clarke, 2019 ). Figures and tables visualise key distributions, and a short synthesis linking the empirical patterns to the theoretical lenses is provided. These choices are consistent with our coverage-oriented, non-probability sampling design, and results are interpreted descriptively rather than as population estimates.

This section presents the empirical findings of the study, based on responses from 75 SMEs in Harare drawn from the retail, services, manufacturing, ICT, and other sectors. The results are discussed in the context of the research objectives and framed against the institutional, procedural, informational, and behavioural factors identified in the literature review. Where applicable, the discussion draws on the theoretical lenses of Institutional Theory ( Meyer & Rowan, 1977 ; Scott, 2008 ), Procedural Fairness Theory ( Tyler & Lind, 1992 ; Wenzel, 2002 ), and the Theory of Planned Behavior ( Ajzen, 1991 ) to interpret findings in light of Zimbabwe’s current tax environment.

The three theoretical lenses informed both instrument design and how the results are interpreted. Items on regulative burden, stability, and institutional trust reflected Institutional Theory; items on clarity, predictability, voice, and proportionality reflected Procedural Fairness Theory; and items on knowledge/skills, peer norms, and future intentions reflected Theory of Planned Behaviour’s constructs of perceived behavioural control, subjective norms, and attitudes. While the study is descriptive rather than hypothesis-testing, interpretation is guided by directional propositions consistent with these frameworks: higher perceived burden or volatility and lower fairness coincide with weaker intentions and lower self-reported compliance; conversely, greater perceived behavioural control (knowledge and usability) and stronger pro-compliance norms align with better intended and reported behaviours ( Ajzen, 1991 ; Tyler & Lind, 1992 ; Scott, 2008 ).

Sectoral Distribution: The sample achieved broad sectoral coverage, with services constituting the largest proportion (25.3%), followed by retail (20.0%), manufacturing (18.7%), ICT (17.3%), and other sectors (18.7%). This distribution reflects a broad spread of SME activities in Harare and supports comparisons across different operational contexts. See Table 1 .

Registration Status: The findings reveal that only 26.7% of respondents were fully registered with ZIMRA, while 36.0% were registered but operating informally, and 37.3% were not registered at all (see Table 2 ). This aligns with previous research highlighting the persistent informality of SMEs in Zimbabwe ( Kanyoka, 2020 ; Mushoriwa, 2022 ). Institutional Theory suggests that where regulative structures are perceived as burdensome or inequitable, compliance incentives diminish ( Scott, 2008 ).

The 36.0% identified as registered yet informal typically exhibit hybrid compliance behaviours common in Harare’s SME ecosystem. In our data and open-ended responses, this status frequently meant selective or episodic filing (e.g., submitting returns when prompted or when a large client requires an invoice), delayed or partial remittance even after filing, and cash-heavy record-keeping with limited use of fiscal devices or e-invoicing. Firms may transact on-book with formal counterparties but off-book with retail customers, under-utilise online portals due to usability or connectivity constraints, and rely on third-party intermediaries ad hoc. This nuance clarifies that formal registration alone does not guarantee continuous, rule-consistent compliance, helping to explain gaps reported in knowledge, process usability, and perceived fairness.

Ownership Structure: Cooperatives (30.7%) emerged as the dominant form, followed by partnerships (20.0%), private companies (18.7%), sole proprietors (13.3%), and other arrangements (17.3%). See Table 3 .

This mix reflects the collaborative and resource-sharing nature of many Zimbabwean SMEs ( Chingonzo & Dylan, 2020 ). Together with the average firm age and headcount reported below, this mix suggests many respondents have sufficient organisational maturity to comply if administrative and informational barriers are eased.

Business Longevity and Size: Respondents had operated for an average of 16.3 years, employing an average of 25 people. These figures suggest a mature SME base, potentially capable of meeting compliance requirements if administrative and trust-and-fairness barriers are addressed ( Beck et al., 2019 ).

The findings reveal that 51.7% of respondents agreed that the total tax burden, when combining national and local levies, is high. Overlapping taxes such as municipal levies in addition to national VAT and PAYE were cited as sources of confusion. This corroborates findings from Mutizwa (2021) and World Bank (2022) , which note that multiplicity and lack of harmonisation in tax systems create inefficiencies and erode compliance motivation.

Dual-currency complexities (ZiG and USD) were frequently reported as complicating tax reporting, especially in payroll management. These challenges mirror earlier accounts by Chibaya (2022) of SMEs struggling to reconcile multi-currency transactions in fiscal records.

About half of respondents (49.3%) felt that tax filing procedures are clear, but many still struggle once they move beyond basic filing. Around 38.7% said frequent changes in the rules make compliance difficult, and only 34.7% felt audits and follow-ups are predictable and fair. Access to professional help was limited, with just 30.7% saying they can easily get support from tax consultants or accountants. Overall, 44.0% agreed that the time and money spent on compliance are burdensome, suggesting that the process remains costly and time-consuming for many SMEs. See Table 4 .

The above is consistent with Eccleston’s (2022) argument that administrative complexity imposes disproportionate burdens on resource-constrained SMEs and with procedural-justice perspectives that uncertainty and opacity depress cooperative, voluntary compliance ( Tyler & Lind, 1992 ; Wenzel, 2002 ).

Knowledge gaps are evident. Only 34.7% felt they fully understand all their tax obligations, 42.7% said they know what incentives are available to SMEs, and 44.0% said they know how to calculate and claim input VAT credits correctly. Formal tax education in the past year was reported by 36.0% of respondents, while 42.7% felt they know where to find reliable tax information. Communication from the tax authority about changes was described as timely and understandable by 38.7% of respondents. Two quick knowledge checks reinforce these gaps: just 25.3% correctly chose “Monthly” as the standard VAT filing frequency for typical VAT-registered SMEs, and only 20.0% correctly identified presumptive tax as a measure to simplify compliance for qualifying small businesses. See Table 5 .

These findings mirror Ndhlovu’s (2021) observation that knowledge gaps and the absence of targeted, SME-appropriate training depress compliance, and they reinforce behavioural insights that perceived behavioural control (i.e., “knowing how”) shapes intention and action ( Ajzen, 1991 ; Torgler, 2007 ).

Views on trust and fairness were mixed. Overall, 44.0% said they trust the revenue authority to use tax revenues responsibly, and 49.3% felt small businesses are treated fairly compared to larger firms. At the same time, only 41.3% believed most SMEs in their sector comply with tax obligations, and 40.0% felt enforcement (audits and penalties) is applied evenly. A third (36.0%) believed corruption affects how taxes are administered, while 37.3% felt that paying taxes gives their business legitimacy and access to benefits. Perceived detection risk was moderate, with 40.0% saying there is a high chance of being caught if they under-report or delay payments. See Table 6 .

This ambivalence is consistent with the literature: when taxpayers question fairness and probity, tax morale weakens ( Dube & Savage, 2021 ; Fei & Chimhowu, 2021 ), and deterrence loses traction unless paired with procedural justice ( Alm, 2019 ; Tyler & Lind, 1992 ).

Digital readiness is uneven. Reliable internet access for using online tax platforms was reported by 37.3% of respondents, and 41.3% said they have suitable devices. Ease of use for TaRMS or similar portals was reported by 41.3%, while 37.3% said digital filing has reduced the time and cost of compliance. Almost half (48.0%) had received support or training on electronic systems, yet 42.7% still rely on third parties to do their digital filing. Technical issues remain common: 48.0% said system glitches or downtime hinder their ability to comply. In the open-ended suggestions, the most common request was to improve the reliability of digital systems (29.3%), followed by more regular training for SMEs (22.7%) and clearer communication on tax changes (17.3%). See Table 7 .

These patterns closely track Mupambwa’s (2022) assessment that technology roll-outs must be matched with capacity-building and reliability investments, and they echo ZIMRA’s own modernisation agenda that stresses end-user support alongside system upgrades ( ZIMRA, 2023 ; Pagero, 2021 ).

The survey points to clear compliance gaps. Only 20% of respondents said they always filed returns on time, while 21.3% said they never did. About 26.7% admitted they always under-reported income, compared with 20% who said they never under-reported. A further 22.7% said they always delayed remitting taxes even after filing. Looking ahead, only 25.3% strongly intended to fully comply in the coming year, whereas 44.0% disagreed or strongly disagreed. When asked why they do not comply, respondents most often mentioned a low perceived risk of detection, the cost of compliance, and procedural complexity. These patterns are consistent with the earlier sections: when rules feel complex, processes unpredictable, and support uneven, self-reported compliance falls. They also align with deterrence thinking, where the likelihood of detection is an important driver of behaviour ( Allingham & Sandmo, 1972 ).

The findings align with and interconnect the three guiding lenses:

Institutional Theory (rules, legitimacy, and alignment). High combined burdens (51.7%) and reported rule volatility (38.7%) signal weak regulative alignment and strained perceived legitimacy of the tax system ( Scott, 2008 ). Dual-currency complications further increase institutional complexity, raising transaction costs and encouraging work-arounds. In this environment, even “registered” firms may adopt hybrid practices (the 36% “registered but informal”); selective filing, delaying or partially remitting, and cash-heavy record-keeping, consistent with institutional responses to misfit between rules and operating realities ( Meyer & Rowan, 1977 ; Scott, 2008 ).

Procedural Fairness Theory (neutrality, consistency, transparency, voice). Only 34.7% felt audits and follow-ups are predictable and fair, and 44.0% viewed compliance as time- and cost-burdensome. These perceptions map directly to procedural justice elements; neutrality, consistency, and respect, that shape cooperative behaviour. Where processes feel opaque or uneven, voluntary compliance and tax morale weaken, and deterrence alone underperforms ( Tyler & Lind, 1992 ; Wenzel, 2002 ; Alm et al., 1992 ).

Theory of Planned Behavior or behavioural lens (attitudes, norms, perceived behavioural control). Knowledge gaps; only 34.7% feeling they fully understand obligations; 25.3% correctly identifying standard VAT filing frequency; 20.0% correctly identifying presumptive tax’s purpose, indicate diminished perceived behavioural control. Weak subjective norms are suggested by just 41.3% believing most SMEs comply, potentially normalising under-reporting. Attitudes are shaped by perceived burden and fairness; meanwhile, perceived detection risk is only moderate (40.0%), which, together with low control and norms, helps explain reported behaviours (e.g., 26.7% “always under-report,” 22.7% “always delay remittance”) and modest future intentions (25.3% “strongly intend to fully comply”) ( Ajzen, 1991 ; Torgler, 2007 ; Allingham & Sandmo, 1972 ).

Taken together, institutional misfit (burden or volatility), procedural justice deficits (predictability or fairness), and behavioural constraints (low control, weak norms, moderate deterrence) provide a coherent explanation for the hybrid and uneven compliance patterns observed, particularly for firms that are registered but functionally informal. These results speak directly to the study’s descriptive objective and the guiding propositions outlined in Section 4.1: improving institutional alignment and procedural fairness, and raising SMEs’ perceived behavioural control, are likely prerequisites for strengthening intentions and behaviours in this context ( Ajzen, 1991 ; Tyler & Lind, 1992 ; Scott, 2008 ).

Structural Factors. We conclude that high and overlapping tax burdens, especially the layering of national taxes with municipal levies, discourage voluntary compliance. Ongoing dual-currency complexities (ZiG and USD) add accounting friction, reinforcing non-compliance tendencies. These patterns echo prior evidence on tax multiplicity and policy inconsistency as deterrents in Zimbabwe and the region ( Mutizwa, 2021 ; World Bank, 2022 ) and reflect institutional misalignment that weakens perceived legitimacy ( Scott, 2008 ).

Procedural Complexity. Frequent rule changes, time-consuming filing, and unpredictable audits raise the cost of compliance and erode procedural justice (clarity, predictability, and neutrality), thereby dampening cooperative behaviour ( Tyler & Lind, 1992 ).

Information Awareness and Education. Notable knowledge gaps persist around obligations, incentives, presumptive tax, and input VAT credits. Limited, tailored tax education reduces SMEs’ perceived behavioural control; the capacity necessary for intention and action ( Ajzen, 1991 ; Ndhlovu, 2021 ).

Perceptions of Trust and Fairness. Mixed views on fairness, probity, and evenness of enforcement continue to strain institutional legitimacy and tax morale ( Scott, 2008 ), with implications for willingness to comply.

Digital Access and Technology. Digital readiness is uneven. Platform reliability issues and limited embedded support constrain the realisation of e-filing benefits, particularly for smaller firms ( Mupambwa , 2022 ).

Compliance Behaviour. Self-reports indicate under-reporting, delayed remittance, and low future compliance intentions. Explanations centre on moderate perceived detection risk, high compliance costs, and procedural complexity ( Allingham & Sandmo, 1972 ), but also with TPB: low perceived behavioural control and weak subjective norms help explain the observed patterns.

Registered but Informal in Practice. A significant share (36%) is formally registered yet operating informally in day-to-day practice (e.g., selective or episodic filing, delayed or partial remittance, cash-heavy records). This hybrid posture underscores that registration does not equal compliance and that usability, fairness, and capability gaps must be addressed to translate formal status into continuous, rule-consistent behaviour.

Streamline national-local tax interfaces to reduce overlap and remove redundant levies. Publish stable, annual or biannual change cycles with consolidated guidance. Issue clear, practical rules for multi-currency reporting to reduce reconciliation errors. These steps improve regulative fit and perceived legitimacy ( Scott, 2008 ; World Bank, 2022 ).

Adopt predictable audit calendars with published risk criteria, standard notices, and checklists. Provide one-stop SME service windows and scheduled virtual help sessions to shorten resolution times. Time-bound grace periods when rules change can prevent inadvertent non-compliance. Greater predictability and neutrality support voluntary cooperation ( Tyler & Lind, 1992 ).

Deliver sector-specific micro-modules (10 to 15 minutes) via workshops and WhatsApp or SMS links covering: obligations, VAT credits, presumptive tax, and common filing flows. Embed step-by-step help inside TaRMS (tooltips, checklists, prompts) so users can complete tasks without intermediaries. Run local “tax clinics” with SME associations near markets or industrial parks to raise hands-on capability. Improving know-how elevates perceived control and, by TPB, intentions and behaviour ( Ajzen, 1991 ; Ndhlovu, 2021 ).

Publish easy-to-read reports on revenue use and anonymised enforcement statistics by sector and size. Provide a tracked feedback or appeals channel with time-bound responses. Co-design forms and notices with SME user groups before roll-out. These measures strengthen perceived fairness and institutional legitimacy, which condition tax morale and cooperation ( Scott, 2008 ; Tyler & Lind, 1992 ).

Guarantee platform uptime with clear outage notices; offer mobile-first, low-bandwidth pages and offline “save and sync” for common returns. Add guided wizards for VAT/PAYE, a safe sandbox or demo to practise, and quick-resolution chat or email support. Provide recurring, targeted training for micro-enterprises with short videos and checklists. These features convert digital access into usable capability.

Use risk-based audits informed by e-invoicing, third-party data, and simple cross-checks (e.g., municipal licences, payroll submissions). Pair short amnesties and compliance windows with visible, even-handed penalties thereafter. Link access to public procurement or permits to demonstrated compliance to shift subjective norms and keep deterrence credible ( Allingham & Sandmo, 1972 ).

Create a transition pathway for registered-but-informal firms: simplified quarterly filing for the first year, default behavioural nudges (deadline reminders, pre-populated forms), and small on-time filing credits that auto-apply within TaRMS. Provide a “first-error fix” protocol (quick correction without penalty) to reduce fear-based avoidance. This bundle directly addresses capability, predictability, and cost frictions that keep registered firms informal in practice.

Future studies should track whether procedural and digital reforms (predictable audit calendars, embedded TaRMS help) raise perceived behavioural control and improve sustained behaviour. Sector-specific analyses (e.g., cash-flow seasonality in retail, invoicing patterns in services) could clarify how industry dynamics shape tax morale and norms. Evaluations of procedural pilots (e.g., one-stop SME desks) using before-and-after or experimental designs would help identify high-impact levers.

Limitations. The study is descriptive, uses a coverage-oriented, non-probability sample of 75 Harare SMEs, and does not make population-level causal claims. Results should be read as indicative patterns. Future work can extend beyond Harare and test targeted interventions with experimental or quasi-experimental designs.