A series of trials with the Tax Administration of Kosovo demonstrate the power of behaviorally informed messaging for increasing tax compliance, and the importance of institutional capacity building The Project for efficient administration Between February and October 2018, the team of tax collection. supported TAK in the design, implementation, and evaluation of three experimental trials to improve tax compliance in Kosovo, with the goal of evaluating how innovative approaches to communication with taxpayers using evidence and adaptive learning could improve timely and honest compliance behavior at low or zero cost. These interventions targeted taxpayers subject to As in many countries, tax collection is a development the personal income tax (PIT) and value added tax (VAT) challenge in Kosovo, one of the poorest and youngest across the country, and tested the impact of reminder countries in Europe in terms of gross domestic product messages on tax compliance outcomes. Initial research (GDP) per capita, demographics, and statehood. in Kosovo honed in on the success of the model that Limited tax revenues hamper the government’s views the taxpayer as a client, informing the decision ability to address economic cycles. Between 2011 to promote positively framed messages, such as the and 2017, total government revenue amounted to benefits of paying taxes, rather than negatively framed about 14 percent of GDP, below the average of 19 messages, such as deterrence. percent among countries in Europe and Central Asia. Unlike other countries that collect significant nontax The first trial targeted letter reminders at PIT taxpayers revenues (for example, from natural resources), the who were late in filing their 2017 annual declarations. Kosovar government relies on taxes for more than Taxpayers who, by four days after the due date, had not 85 percent of its revenues. Mobilizing tax revenues is yet submitted their PIT declarations for the 2017 fiscal therefore critical from both an efficiency perspective year were identified in the tax registry and were eligible (to ensure that taxes are collected at the lowest to receive a letter in the mail. The second trial targeted cost for the public administration) and an equity all firms required to submit June VAT declarations; it perspective (to ensure that all taxpayers comply by involved the use of e-mail–based reminders on two declaring and paying their due taxes). occasions, the first well in advance of the monthly due date. All firms in the VAT registry, excluding those that The Tax Administration of Kosovo (TAK) requested had already submitted declarations by July 3, 2018, assistance from the World Bank and the Deutsche or for which no e-mail addresses were on file, were Gesellschaft für Internationale Zusammenarbeit (GIZ) considered eligible to receive an e-mail reminder from to address this challenge using an evidence-based the TAK Information Technology Department, first on approach. Together, the World Bank and GIZ applied July 4, then again on July 20, the due date of the June behavioral insights to promote tax compliance among declaration. The third trial targeted all firms required to specific groups of taxpayers. Three experiments were submit August VAT declarations; it involved the use of designed, implemented, and evaluated in 2018 that SMS-based reminders a few days in advance of the due involved sending behaviorally informed reminders using date. All firms in the VAT registry, excluding those that letters, e-mails, and short messaging service (SMS) had already submitted their declarations by September messages to various groups of taxpayers to induce 17, 2018, or for which there was no cell phone number timely and honest declarations and payments. The short- on file, were considered eligible to receive an SMS term objective of these trials was to increase the number reminder, which was sent through a telecommunication and timeliness of tax declarations. firm in Kosovo. The Results All three communication trials succeeded in Implementation challenges created unexpected producing statistically significant impacts on outcomes. the declaration rate or the payment rate in Only 48% of letters made it to their intended recipients and the short term, within four to six weeks. less than one of four e-mails was actually opened. Moreover, The behavioral messages managed not only to induce the messages themselves were not always interpreted the more people to submit their declarations on time, but way we intended: the augmented behavioral reminders did also to encourage more people to submit declarations. no better (and in fact, were less successful) than the simple behavioral reminders. These implementation challenges arose as a result of outdated (or incomplete) contact information, highlighting the need to establish processes that help to integrate tax collection functions and data systems; an unbalanced timing of message delivery, highlighting the sensitivity of communication strategies in tax trials; and an imperfect interpretation of messages, highlighting the need Letter reminders to taxpayers late in submitting their to carefully evaluate language in communications. annual declarations resulted in small, but statistically significant impacts on the declaration rate. The PIT reminders were successful in inducing more people to submit declarations closer to the due date. Among those who were randomly assigned one of the two letters (the intent-to-treat group), there was an increase of 2 to 4 percentage points in the declaration rate within one Collaboration and capacity building are essential. month of the transmission of the letters. But despite the Close coordination between the World Bank, GIZ, and TAK significant impacts on the declaration rate and the payment demonstrated that a behaviorally informed nudge might rate, no effects were identified on payment amounts – encourage a population of interest toward a common goal, perhaps because the population of interest in this trial but also create positive externalities by nudging the policy plausibly consisted of individuals who did not actually have makers. By participating in the design and implementation anything to declare. Indeed, among those who were late of the trials through brainstorming sessions with the in submitting declarations, but who did end up submitting TAK Director General’s Office and the TAK Information in the absence of the reminder, only 18 percent claimed Technology Department, TAK staff were able to come they owed something in taxes. These results were similar in to understand how behavioral science tools can change the e-mail and SMS trials – declaration and payment rates the mindsets of taxpayers and tax administrators by improved, but payment amounts did not. In this case, the highlighting nuances in the communication with taxpayers, lack of an impact on payment amounts was likely due to the and underscored how TAK data systems could be pushed fact that nearly 8 of 10 firms on average declare no VAT to the limit to identify new ways of doing business and taxes are owed. interacting with taxpayers. Policy Implications As with other behaviorally informed interventions to Crucially, the support of local partners is indispensable improve tax compliance, our trials in Kosovo demonstrate for successful implementation and long-term capacity the potential impact for improving taxpayer behavior building. Supporting the implementation of tax trials without onerous traditional policy tools. But despite the requires a hands-on approach by the project team, successes of the three trials, implementation challenges which is feasible only with local support. Because created bottlenecks in design and, ultimately, evaluation; decisions are generally made in a tight time frame challenges that should be addressed in future trials. A lack and data are shared on the fly, daily communication of reliable contact information with the tax authority during and accurate taxpayer data Our trials in Kosovo demonstrate implementation is crucial resulted in un-delivered letters, for success in these types the potential impact for improving low email open rates, and other of interventions. By closely obstacles to implementation. For taxpayer behavior without onerous collaborating during the tax authorities in Kosovo, and traditional policy tools. trials, the World Bank and other policy makers looking to the GIZ were able to foresee learn from these trials, these hiccups provide important the challenges posed by the project and make quick lessons for greater effectiveness as they continue to adjustments when necessary. Ongoing cooperation integrate behavioral science into tax messaging and between these two institutions is essential. Partners policy. For instance, while TAK authorities initially can learn from this implementation success, especially preferred email messages to SMS, the latter were more if the collaboration with a government counterpart is successful in delivery, opening, and reading rates. Given communication intensive. Close collaboration between cultural considerations, SMS messages might be the most the teams creates mutual learning and effective effective in future trials simply because of the reliability of adaptation, instilling a behavioral mindset among the contact information and accessibility (nearly universal partners and ensuring that future policies and programs use of smartphones). incorporate behavioral design at the outset. About eMBeD The Mind, Behavior, and Development Unit (eMBeD), the World Bank’s behavioral science team in the Poverty and Equity Global Practice, works closely with project teams, governments, and other partners to diagnose, design, and evaluate behaviorally informed interventions. By collaborating with a worldwide network of scientists and practitioners, the eMBeD team provides answers to important economic and social questions, and contributes to the global effort to eliminate poverty and enhance equity. Stay Connected eMBeD@worldbank.org #embed_wb worldbank.org/embed bit.ly/eMBeDNews Last Update: March 4th, 2019