Revenue Mobilisation Strategies in Digitised Tax Administrations: The Case of Uganda

充足的税收对于低收入和中等收入国家缩小资金缺口、资助公益投资和加强国家建设至关重要。然而，许多中低收入国家，特别是非洲国家，受到税收薄弱的影响，主要是由于普遍存在的非正规性和税收合规性差。乌干达作为研究对象也不例外。在这方面，乌干达税务局（URA）对大规模登记和正规化方案进行了大量投资。更具体地说，市建局依靠来自其他政府机构的第三方数据信息，强制登记显示经济活动活跃的纳税人。虽然这种从传统的亲自登记和执法策略到更便宜的数据驱动方法的转变对税收限制的税收当局很有吸引力，但人们对其有效性知之甚少。如果说有什么不同的话，那就是在乌干达占多数的强迫登记比自愿登记更不可能纳税，这表明在遵守方面存在严重问题。在这种背景下，至关重要的是要了解不同的登记策略如何转化为更高的税收收入，它们如何塑造纳税人的观念和态度，并最终改善合规行为。此外，还需要更多的证据来说明税务当局如何鼓励这些本质上不太遵守税法的纳税人遵守税法。我的博士工作计划使用乌干达的情况下，阐明在国家能力薄弱的情况下，使用第三方数据进行正规化和执法的潜力。为此，我将从2，000名纳税人的代表性样本中收集和分析调查数据，这些样本分为强制和自愿登记。调查数据将侧重于纳税人的登记经历，以及他们的税务态度和看法，以了解登记过程与纳税人意见之间的相关性。最初结合调查和行政数据，我也打算连接的看法与现实生活中的合规行为。然后，我的目标是测量，充分的计量经济学工具，通过分析丰富的纳税人层面的URA行政数据，强制登记行为的因果影响。最后，我将严格测试强制注册背后的合规驱动因素，通过以随机对照试验的形式对大约50，000个强制注册建立一个廉价，易于扩展的大规模短信活动。通过试验，我将确定不同的推动信息（包括提供第三方信息）对纳税人合规行为的因果影响，从申报和付款数据中衡量。通过我的研究，我计划通过提供第三方数据的实际潜力，为加强税务管理和调动收入的核心功能提供基于证据的建议，直接为URA和其他税收当局的政策制定提供信息。

Adequate tax revenue is crucial for low- and middle-income countries to close their financing gap, fund investments in public goods, and strengthen State-building. Such needs are even more urgent after the COVID pandemic.However, many LMICs, especially in Africa, are affected by weak tax collection, mostly due to widespread informality and poor tax compliance. Uganda, the country under study, is no exception. In this context, the Uganda Revenue Authority (URA) has heavily invested in mass registration and formalisation programmes. More specifically, the URA is relying on information from third-party data from other government institutions to forcedly register taxpayers showing active economic activity. While such shift from traditional in-person registration and enforcement strategies to a cheaper, data-driven approach is appealing to budget-constrained revenue authorities, littleis known about their effectiveness. If anything, forced registrations, which became the majority in Uganda, are much less likely to pay taxes than voluntary registrations, indicating deep issue with compliance. Against this background, it is crucial to understand how different registration strategies translate into higher tax revenues, how they shape taxpayers' perceptions andattitudes, and eventually better compliance behaviour. In addition, more evidence is needed on how the revenue authority can encourage these intrinsically less compliant taxpayers to abide to the tax law. My doctoral work plans to use the case of Uganda to shed light on the potential of the use of third-party data in for formalisation and enforcement in a weak State capacity context. I will do so by collecting and analysing survey data from a representative sample of 2,000 taxpayers, split by forced and voluntary registrations. Survey data will focus on the registration experience of taxpayers, as well as on their tax attitudes and perceptions, so to capture correlations between the registration journey and taxpayers' opinions. Originally combining survey and administrative data, I also intend to connect perceptions withreal life compliance behaviour. Then, I aim to measure, with adequate econometric tools, the causal impact of forced registration on behaviours by analysing rich taxpayer-level URA administrative data. Lastly, I will rigorously test the drivers of compliance behind forced registrations, by setting up a cheap, easily scalable, mass SMS campaign on about 50,000 forced registrations in the form of a randomised control trial. Through the trial, I will identify the causal impact of different nudging messages - including the provision of third-party information - on taxpayers' compliance behaviour, measured from filing and payment data. With my research, I plan to directly inform the URA's and other revenue authorities' policymaking, by providing evidence-based recommendations on the actual potential of third-party datafor strengthening the core function of tax administrations and mobilising revenue.