Does the threat of being blacklisted change behavior?

Regression discontinuity evidence from the EU’s tax haven listing process

Activists stage a protest on a mock tropical island beach representing a tax haven outside a meeting of European Union finance ministers in Brussels, Belgium, December 5, 2017.  REUTERS/Francois Lenoir


In late 2016, the EU Commission began a detailed review of over 80 non-European jurisdictions to determine how compliant they were with international standards around tax transparency, fair taxation, and adherence to the OECD’s Base Erosion Profit Shifting (BEPS) minimum standards. The EU subsequently published both a “grey” and “black” list of jurisdictions to promote their cooperation with these standards. I investigate the impact that this process has had on several measures of international tax governance by employing a regression discontinuity design derived from the unique process the EU used to select jurisdictions for review. I find that although jurisdictions selected into review were substantially more likely to be grey or blacklisted, detectable improvements in tax governance are largely limited to (i) increases in transparency around the presence and removal of harmful tax regimes and (ii) increases in the effective implementation of exchange-of-information (EOIR) agreements. However, countries selected into the EU process were significantly more likely to join the Inclusive Framework, a forum dedicated to implementing the BEPS minimum standards and deliberating over changes to international tax rules. Back-of-the-envelope estimates suggest that the Inclusive Framework is roughly 15 percent larger thanks to the EU review process, although its composition in terms of representation by developing countries or jurisdictions traditionally thought of as tax havens remains roughly the same.


There has been a rapid shift in global tax governance in the past decade. This has been prompted by a recognition that there are two significant externalities driving the movement of financial assets and profits to offshore financial centers (OFCs). The first is financial secrecy, provided by offshore jurisdictions to clients who are able to obscure their ownership and potentially avoid taxation. Studies estimate the amount of wealth being held in offshore tax havens to be approximately 8 percent of all household wealth or 10 percent of global GDP, a significant portion of which goes unreported (Zucman 2013; Johannesen et al. 2018). The second externality is a set of corporate tax policies—a combination of rates, loopholes and lack of transparency—that create incentives for multinational enterprises (MNEs) to shift their profits away from high tax jurisdictions to lower tax ones where there is little economic activity of substance. In a recent study, Tørsløv, Wier, and Zucman (2019) estimate that up to 40 percent of global MNE profits are shifted to tax havens.

To collectively deal with these externalities, jurisdictions across the globe are in the process of committing to two separate OECD frameworks aimed at reducing international tax evasion and avoidance. The first of these is the OECD/G20 Base Erosion and Profit Shifting (BEPS) initiative, which is being taken forward by the international forum known as the Inclusive Framework (IF) on BEPS. The goal of the IF is to promote specific actions and standards that will help countries tackle tax planning efforts by multinationals which lead to an erosion of the corporate tax base. At the very least, members of the IF are expected to adopt four minimum standards, built around reducing harmful tax practices, combating tax treaty abuse, handling treaty disputes and arbitration and finally documenting transfer pricing. The last of these include country-by-country reporting (CbCr), the requirement for parent companies of multinationals to disclose significant details about their operations, profits and tax payments, which will then be exchanged between participating tax authorities. It should be noted that the BEPS Minimum Standards are only part of the entire package of reforms promoted by the OECD – and by themselves do not represent a sufficient set of policies for eliminating cross-border tax externalities. As of April 2020, 137 jurisdictions have joined the Inclusive Framework as members, committing to adopting its standards.

The second initiative is the adoption of the OECD’s Common Reporting Standard (CRS) for the Automatic Exchange of Information (AEOI). Jurisdictions that adopt this standard will require financial institutions to report account information for non-resident taxpayers and for that information to be automatically exchanged between tax authorities in participating jurisdictions. Recent studies suggest that bilateral AEOI exchanges lead to a shift of offshore assets out of tax havens, although it is unclear how much of this presumably untaxed wealth is repatriated (Beer, Coelho, and Leduc 2019; Casi, Spengel, and Stage 2019; O’Reilly, Ramirez, and Stemmer 2019; Menkhoff and Miethe 2019). Approximately 130 jurisdictions have committed to exchanging under CRS, over a third of which which began their first exchanges in 2017, another third in 2018 and the remaining third between 2019 and 2023. The CRS framework is seen as an improvement from an older system of information exchange, known as exchange-of-information on request (EOIR), where tax authorities must make active requests for information on specific taxpayers. Despite this, there are still efforts to ensure that EOIR is being adequately implemented, as jurisdictions are reviewed through the Global Forum on Transparency and Exchange of Information for Tax Purpose.

In late 2016, the EU Commission began a careful review of 80-90 non-European jurisdictions to determine how compliant they were with international standards around tax transparency, fair taxation and adherence to the OECD’s Base Erosion Profit Shifting (BEPS) minimum standards. After the review and some dialogue with non-compliant countries, the EU released a “grey” list of jurisdictions who were non-compliant with these standards, but had committed to make improvements, as well as a “black” list of jurisdictions who were non-cooperative, who were to be subject to a number of EU countermeasures.

In this paper, I investigate the impact that this process has had on the standards that the EU intended to enforce. I rely on the process the EU used to select countries for consideration in its listing process to compare jurisdictions who scored just high enough to be considered with those that did not. Using a regression discontinuity specification, I find that countries that were selected into the EU’s review process were substantially more likely to be grey or blacklisted, but that there is mixed evidence that, to date, the process has affected policy adoption. On average, index measures of global tax governance, based off of the EU’s own goals, do not show large improvements. The main exception is for Fair Taxation, where jurisdictions selected by the EU saw a large increase in the probability that their tax regimes had been inspected by the EU or the OECD and, as of the time of writing, that they no longer had any harmful regimes present. There is weak evidence that the EU review process increased the the number of BEPS minimum standards adopted. There is also some evidence that the EU process led to a sharp increase in the probability that the Global Forum rated a jurisdiction as “largely compliant” or better on its implementation of EOIR.

The most robust and striking result from the analysis is the fact that countries selected into the process were substantially more likely to join the Inclusive Framework, thus committing themselves to implementing the BEPS minimum standards. This means that even if the EU review process has not improved international tax governance by much in the medium term, it might do so in the long term as these commitments become more binding. It also has implications for the future of deliberation over new international tax rules, as it shows that unilateral involvement of regional unions can influence participation in international standard setting. Using a difference-in-difference strategy, I show that, on average, the EU review and listing process increased the probability of IF membership by approximately 30 percent for selected jurisdictions, translating into an increase in total IF membership by around 15 percent. While the composition of the IF is not radically different due to the EU’s involvement, its impact on the participation of developing countries might have been stronger had it set lower thresholds for the review, or had it not excluded least developed countries (LDCs) from the review process.

The other result worth emphasizing is that the EU review process appears to have had positive spillovers on the probability a jurisdiction was reviewed by the OECD Forum on Harmful Tax Practices. This may have been driven purely by the fact that IF members are subject to these reviews, but indicates how unilateral efforts by one entity can have spillovers onto others. While I am unable to identify the net impact the EU review and listing process has had on the total number of harmful tax regimes that have been struck down, the results in this paper are consistent with the EU review process having a sizable impact.

This paper makes several contributions. First, it is the first rigorous test of the impact of the EU’s efforts to improve tax governance worldwide. While it is easy to observe how countries included in the listing process have improved, we would not normally know how these countries would have improved without the EU’s intervention, particularly because there is ongoing pressure from a multitude of institutions (e.g. the OECD, US Government, IMF, World Bank) to improve tax governance.

Second, this paper adds to a nascent empirical literature on the effect of listing exercises on institutional behavior and outcomes. For example, Morse (2019) shows that those added to the Financial Action Task Force (FATF)’s “greylist” of countries that lack compliance with international anti-money laundering (AML) standards are significantly more likely to criminalize money laundering. Kelley and Simmons (2015) find that countries listed on the U.S. State Department’s annual Trafficking in Person’s report are more likely to subsequently criminalize human trafficking. These empirical studies are backed up with case study evidence that jurisdictions are nudged into compliance by the threat of blacklisting (in both the space of AML/CFT and in tax transparency), even when there are no explicit sanctions (Sharman 2009).

The rest of the paper proceeds as follows: Section 2 discusses the recent history of the EU review and listing process. Section 3 discusses the empirical approach I take in this paper, Section 4 presents the main regression discontinuity results as well as results exploiting changes across time. Section 5 discusses the implications these results have for international tax governance as well as reasons the EU blacklisting process may not have a powerful effect on state behavior. I conclude the paper with Section 6.

Download the full working paper


  • Footnotes
    1. The term tax haven is occasionally a contentious one, but they are generally thought of as jurisdictions that employ financial secrecy, low corporate tax rates, and/or preferential tax regimes to attract financial assets and investments from foreign entities and persons.