Bank deposits in tax havens potentially hide illegal tax evasion. The October 2016 release of bilateral locational banking statistics permits us to illuminate three open issues in this respect. We find that the intended effect of additional information-exchange-on-request treaties vanishes since about 2010. This also holds for bank deposits from tax havens in non-havens. In contrast, new treaties based on the automatic exchange of information show bite. This suggests that tax evasion changes its disguise: it adapts to established information exchange treaties while tax evaders seem (partly) surprised by, and thus react to, new treaty forms.

live interview on DW English
Handelsblatt (Germany)
Der Standard (Austria)
Neue Zuercher Zeitung (Switzerland)

Both the data and the code for this paper have been made available together with its publication here.

The political economy literature has put forward a multitude of hypotheses regarding the drivers of structural reforms, but few, if any, empirically robust findings have emerged thus far. To make progress, we draw a parallel with model uncertainty in the growth literature and provide a new version of the Bayesian averaging of maximum likelihood estimates (BAMLE) technique tailored to binary logit models. Relying on a new database of major past labor and product market reforms in advanced countries, we test a large set of variables for robust correlation with reform in each area. We find widespread support for the crisis-induces-reform hypothesis. Outside pressure increases the likelihood of reform in certain areas: reforms are more likely when other countries also undertake them and when there is formal pressure to implement them. Other robust correlates are more specific to certain areas — for example, international pressure and political factors are most relevant for product market and job protection reforms, respectively.

The data and a readme for this paper has been made available together with its publication here. I’m in the process of publishing R code on the model averaging routine that will allow you to replicate the paper and more.

This paper studies how national implementation shapes individual responses to global agreements by looking at the introduction of the multilateral standard for automatic information exchange on financial assets, i.e., the Common Reporting Standard (CRS). We utilize rich micro-level data on all bank transfers to Norway. This provides us with unparalleled detail on hidden ownership structures. These data show a significant increase in cash repatriation from tax havens post-CRS implementation. Yet, we document substantial heterogeneity in responses down to a null result if CRS enforcement is weak. Relying on macroeconomic data on crossborder bank deposits, we employ model averaging techniques to establish the most important characteristics of the receiving countries that make the CRS more effective. Our results suggest that a highly digitized tax administration triggers twice the drop in tax haven deposits compared to a tax administration relying on paper tax returns. These results have implications for global policy initiatives more broadly.

Ownership of real estate through corporations in offshore tax havens creates opportunities for tax evasion and money laundering and may have undesirable effects in housing markets. In this paper, we study offshore ownership of real estate in the United Kingdom by combining several data sources: administrative data from the land register, a comprehensive transaction database, a propriety database on corporate ownership links, and a handful of offshore data leaks. Our descriptive analysis shows that the market share of offshore corporations has increased over time and varies strongly across market segments: It currently stands at 1.25% in the overall residential market and around 15% for top-end properties. When data leaks allow us to trace ownership through offshore corporations to the beneficial owners, we find that around half have ties to Africa, Asia and the Middle East, but that the largest ‘foreign’ investor is the United Kingdom itself. Turning to causal evidence, we show that changes in tax incentives and ownership transparency induce strong responses in patterns of offshore ownership, suggesting that both taxation and secrecy are important motives for the beneficial owners. Finally, we show that the Brexit referendum was followed by a sharp increase in property sales by offshore owners and a large differential decrease in property prices in local areas with more offshore ownership, conditional on area and property characteristics. This suggests that the reduction in demand from offshore investors triggered by Brexit had a negative causal effect on property prices and, more broadly, that offshore ownership can have significant real effects in housing markets.

VoxEU column
Frankfurter Allgemeiner Zeitung (german)
OECD Global Forum Plenary Meeting 2023
Welt am Sonntag (german)

This paper studies financial service provision in offshore financial centers (OFCs). Exploiting the natural experiment of recurring hurricanes hitting small islands, I show that local conditions, captured by satellite data, deteriorate significantly for nine months on average. However, both the international bank sector and international investors do not react in OFCs while non-OFCs show strong negative reactions. Instead of being connected to local activity, OFC service provision declines during holidays in London, Tokyo, and New York. Thus, international regulation attempts requiring local enforcement could be targeted better, financial risks could be mis-assigned, and offshore finance is a doubtful development strategy.

Handelsblatt (german)\ finanzen.net

Assuming that not everyone cares about small islands as much as I do, Mark Toth and me wrote an R package based on the code of this paper that allows you to build plots and moonthly or yearly time series of nightlight intensity for (almost) any geospatial unit on the planet. Find it on github.

In this paper we provide evidence of a tangible advantage of being multinational when weathering external shocks. We combine multiple sources of entity level panel data and compare the borrowing and lending patterns of affiliates of multinational enterprises to the patterns of local firms at the onset of a period of bank credit shortage. We first document that multinational affiliates have access to substantially more debt than similar domestic firms after a shock to external credit markets. Second, we show that part of this capital advantage comes from access to the group internal capital market of the multinational. Third, we show that two groups of affiliates in particular drive these differences: the ones that are likely financially constrained when the credit shock hits and the ones that appear especially profitable. We interpret this as MNEs bailing out the members that are in trouble as a result of less access to bank finance as well as supporting especially profitable members to take over larger market shares following bank credit shortage in an area. Lastly, we document that this capital advantage results in a significant performance difference with multinational affiliates growing substantially faster than local firms after the external credit shock. Unlike domestic firms, multinational enterprises have access to a group-internal capital market that spans countries. We show that this internal capital market increases resilience of multinational affiliates during economic downturns. We combine multiple sources of entity level panel data and compare the borrowing and lending patterns of affiliates of multinational enterprises to the patterns of fully domestic firms at the onset of a period of bank credit shortage. We first document that multinational affiliates take on 10-15% more debt than similar domestic firms after a shock to external credit markets. Second, we show that this additional debt is driven by financing coming from within the multinational group i.e. via the internal capital market of the multinational. Third, we show that two groups of affiliates in particular drive these differences: the ones that are likely financially constrained when the credit shock hits and the ones that appear especially profitable. We interpret this as MNEs bailing out the members that are in trouble as a result of less access to bank finance as well as supporting especially profitable members to take over larger market shares following bank credit shortage in an area.

Offshore banking enables money creation outside of national regulation, yet its full scale remains opaque. This article provides the first comprehensive estimate of Eurodollar banking, the unregulated creation of the United States’ currency under another jurisdiction’s law. Drawing on economic history and interviews with market participants, we conceptualize the Eurodollar system as comprising loans, bonds, and foreign exchange derivatives. Applying a novel measurement method to Bank for International Settlements (BIS) data, we estimate that in 2023, $11.4 trillion in loans and bonds and $75.8 trillion across all instruments constituted unregulated offshore dollars, exceeding the volume of regulated cross-border U.S. dollar instruments. Geographically, Europe – particularly the City of London – remain central to the system. Offshore banking allows privileged private actors to circumvent public rules. It underpins other offshore financial services such as tax planning or sanctions evasion. Yet the recent regulation of Eurodollar futures suggests a latent capacity for public control.

We evaluate the impact of regulatory attempts by German authorities to combat international tax evasion. Monthly cross-border deposits from tax havens in German banks show a 32-34% reduction in reaction to bilateral information exchange. These findings are comparable in magnitude to a number of reference studies which confirms Germany as a valid case study for international tax evasion. We show disaggregated reactions for a list of tax havens and find large reactions to information exchange for Guernsey, the Bahamas, and Jersey. We then employ a manually constructed narrative databases collecting changes in the German tax code and information leaks from tax havens. Tax evaders do not react to changes in tax rates and leaks show consistent signs but are rarely significant. These results confirm information exchange as the main focus in the analysis of international tax evasion.

We wanted to use German bank data with relatively high frequency to say more about the anticipation and delays in tax haven transfers in reaction to international policy changes. While we were well able to establish the general patterns in German data that are documented globally elsewhere, we did not find anything particularly surprising in the timing.