VAT is a key source of tax revenue in many low and lower-income countries, but evasion at the point of sale remains a challenge. In Tanzania, a VAT-receipt lottery piloted in Tegeta Tax Region in Dar es Salaam, boosted reported sales and VAT payments among registered firms, with no effect on unregistered ones. Strategic business behavior limited overall gains. The findings suggest that customer-focused incentives such as VAT lotteries can improve compliance but need stronger enforcement and broader tax system reforms to be fully effective.

Background and policy challenge

VAT systems rely on firms to issue receipts that record taxable transactions. In practice, the self-enforcing nature of VAT breaks down at the point of sale to final consumers. Unlike business-to-business transactions earlier on in the value chain, final consumers have no direct incentive to demand receipts, which weakens enforcement. While customers have a legal obligation to obtain receipts, this obligation is not enforced. This “last-mile” problem is particularly acute in countries with limited administrative capacity and widespread informal economic activity.

In Tanzania, between 2000 and 2021, the African Development Bank (2024) estimates that only about 24 percent of potential VAT revenues were collected. Many firms remain outside the VAT net entirely by underreporting sales and staying below the registration threshold.^[1] For firms already registered, the incentive to conceal part of their turnover persists, especially in a context where detection and penalties are rare. Previous efforts to enforce compliance, such as the introduction of Electronic Fiscal Devices (EFDs) - electronic tills which transmit information directly to the tax authority when used to register transactions - have had limited impact in the absence of consumer demand for receipts.

The lottery intervention

The Tanzania Revenue Authority (TRA), in collaboration with CMI-researchers, launched a receipt lottery in Tegeta Tax Region from July to September in 2022. Customers who obtained EFD-generated receipts and submitted their phone numbers were automatically entered into weekly and monthly prize draws. The intervention aimed to create a consumer incentive to request receipts, thus indirectly inducing firms to record sales accurately.

The lottery applied to both VAT and non-VAT-registered firms using EFDs. All firms with an annual turnover exceeding TZS 14 million (approximately USD 5,400) are required to register their transactions with an EFD. While transaction details from VAT-registered firms directly impact tax revenues, those from non-VAT firms may contribute to increase recorded sales above the VAT threshold, potentially leading to increased VAT registration.

To advertise the lottery, public awareness campaigns including announcements via loudspeakers, posters, and promotional materials such as branded T-shirts and caps were carried out before and during the lottery.

Methodology

The evaluation of the lottery combined two primary sources of data. Administrative data from the TRA covered all Tanzanian firms using EFDs from January 2021 through October 2022. This allowed for a difference-in-differences analysis comparing Tegeta with other tax regions. The study focused on reported sales, number of receipts, and VAT liability for registered firms, and changes in turnover and registration threshold breaches for non-registered firms.

To complement the administrative data, three waves of customer surveys were conducted during the implementation period. These surveys assessed consumer awareness of the lottery, receipt-requesting behavior, and the type of receipts received. The empirical methods and the use of administrative as well as survey data enabled the researchers to evaluate both the statistical outcomes of the lottery and the behavioral mechanisms driving them.

Key findings

The lottery generated measurable improvements in recorded sales and VAT due among VAT-registered firms. On average, firms in Tegeta reported a 61 percent increase in monthly sales and a 66 percent rise in VAT liability compared to control regions. These are substantial shifts, suggesting that the increased customer demand for receipts caused by the lottery was effective in encouraging more complete reporting.

Findings from the survey furthermore show that customer awareness of the lottery rose steadily over time. Initially, only 10 percent of respondents had heard about it. By the end of the intervention, this figure had risen to nearly half. Furthermore, among those aware of the lottery, demand for receipts increased significantly, and the likelihood of obtaining a TRA receipt in this group rose by 13 percentage points. Customers who requested a receipt were more likely to receive one as the lottery progressed.

However, while customers became more likely to request receipts - particularly those aware of the lottery - firms reduced their voluntary compliance and responded by reducing the number of receipts issued when customers did not ask for it. Moreover, there was a marked increase in issuing unofficial, non-TRA receipts, meaning that the transaction details are not transmitted to the TRA.

These strategic responses partially offset the intended effects of the lottery. Overall, 77 percent of transactions captured in the survey were non-compliant, indicating the depth of the challenge despite the lottery’s influence. This underlines the risk of relying solely on consumer incentives without reinforcing mechanisms to constrain evasive firm behavior.

The intervention had no statistically significant impact on non-VAT-registered firms. Sales, receipt issuance, and the frequency of surpassing the VAT registration threshold remained unchanged. This may be attributed to customers’ limited understanding of which firms are obligated to use EFDs, as well as the typically small value of purchases from non-registered vendors.

Conclusion

The receipt lottery in Tanzania demonstrates that consumer-side incentives can improve compliance at the point of sale, particularly in settings where enforcement capacity is weak. The increase in reported VAT liabilities shows the potential for this tool to complement traditional enforcement strategies.

However, the intervention also reveals its limitations. Firms adapted quickly to avoid the full implications of customer demand, and without sufficient monitoring they were able to neutralize some of the gains. The lottery had no effect on firms operating below the VAT threshold, underscoring the need for broader strategies to address informal sector compliance. Thus, to be effective at scale, receipt lotteries should be embedded in a comprehensive compliance strategy that includes improved enforcement, public education, and digital tools to track firm behavior.

The Tanzanian receipt lottery experiment underscores the value of aligning citizen behavior with government enforcement systems. By motivating consumers to participate in tax enforcement, the intervention helped close a key compliance gap. However, the strategic responses of firms highlight the need for coordinated efforts that combine incentives, monitoring, and enforcement.

If scaled and strengthened, receipt lotteries could become a vital tool in the broader effort to improve tax capacity in low-income settings, helping governments mobilize domestic resources more effectively and equitably.

Further Reading

African Development Bank (2024). African Economic Outlook 2024. https://www.afdb.org/en/documents/african-economic-outlook-2024

Buluba, K., Fjeldstad, O.-H., Nawab, O., Sjursen, I.H. and Somville, V. (2025). Improving VAT-compliance by incentivizing customers: Evidence from Tanzania. CMI Working Paper WP 2: 2025. Chr. Michelsen Institute.

Fjeldstad, O.-H., Kagoma, C., Mdee, E., Sjursen, I.H. and Somville, V. (2020). The customer is king: Evidence on VAT compliance in Tanzania. World Development 128.

Naritomi, J. (2019). Consumers as tax auditors. American Economic Review 109(9).

Pomeranz, D. (2015). No taxation without information. American Economic Review 105(8).

How to cite this policy brief

Buluba, K., Fjeldstad, O.-H., Nawab, O., Sjursen, I.H. and Somville, V. (2025). The power of receipts: Enhancing VAT compliance in Tanzania through customer incentives. CMI Policy Brief. Chr. Michelsen Institute.

This CMI Brief is part of the project Tax Compliance, VAT Revenues and Business Development in Tanzania, funded by the Research Council of Norway (#287105/H30).