VoxEU Column Financial Regulation and Banking International Finance

Regulatory arbitrage and the G20’s global derivatives market reform

Managing global financial risks requires coordinated policies and a firm commitment by national actors. In the absence of such commitment, risks are reallocated and concentrate where they are least effectively addressed. Using data on the staggered implementation of the G20’s global derivatives market reform, this column documents US banks’ response to reform progress. It finds that banks shift trading activities towards less regulated jurisdictions and adopt riskier portfolios overall. The effects are driven by agenda items – like the promotion of central clearing – that are costly and do not benefit banks directly.

In 2009, the G20 launched its global derivatives market reform to target the kind of over-the-counter (OTC) trading that had spread losses from the US housing market to the world economy (Acharya and Engle 2009, Tavares 2011). The reform was designed to improve transparency and regulatory oversight and consisted of five building blocks: (i) the introduction of trade repositories to facilitate surveillance, (ii) central clearing of standardised OTC derivatives to reduce counterparty exposure and facilitate resolution, (iii) exchanges and electronic trading platforms to reduce operational risk, (iv) higher capital, and (v) higher margin requirements for non-standardised OTC derivatives to buffer losses ex-post and align incentives ex-ante. That the different blocks had to be integrated into national or – in the case of Europe – regional regulation by participating governments raises concerns about heterogeneous standards and cross-jurisdictional arbitrage.

Opportunities and incentives for regulatory arbitrage

Participating governments have been integrating the different blocks of the reform into (supra)national regulation at different paces. By Q4 2016, all countries had fully implemented at least one block of the reform, but only five jurisdictions (Australia, Japan, Mexico, Switzerland, and the US) had fully implemented at least four agenda items, and only Japan had adopted all five blocks.

In addition, different governments also appear to have prioritised different dimensions of the reform. While Argentina had only implemented capital regulation by Q4 2016, the Republic of Korea had only adopted the regulation on trade reporting.

Finally, even if the reform has officially been adopted, discretion at the national level leaves room for loopholes and exceptions. Under the Dodd-Frank Act, the US implemented significant parts of the reform earlier than other countries. At the same time, it allowed US rules to apply to overseas branches but not to (de-guaranteed) overseas subsidiaries, which fall under the host country’s regulation.

On a global market with highly mobile assets and low transaction costs, such as the derivative market, differential standards create evident opportunities for cross-jurisdictional arbitrage. With one estimate for the European derivatives market predicting annual costs of €15.5 billion for compliance with the G20 reform (Deloitte 2014), the incentives to exploit these opportunities are strong.

US banks’ response to the Dodd-Frank loophole

In recent research (Gandre et al. 2020), we provide a first assessment of regulatory arbitrage during the implementation of the G20 reform. To this end, we combine hand-collected data on the staggered implementation of the reform with subsidiary-level information on the security holdings of American banks in distinct countries.1 Our data enables us to analyse the security holdings of US banks’ foreign subsidiaries, before and after the adoption of the reform in the US.

Figure 1 shows the evolution in the share of interest rate swap (IRS) activity – from before the adoption of the Dodd-Frank Act in Q1 2010 to Q4 2015 – that the largest US dealers (Bank of America, Citigroup, Goldman Sachs, JP Morgan, Morgan Stanley) operate through their foreign subsidiaries. With the exception of Bank of America, they clearly move their security holdings abroad; in the most extreme case (Citigroup) from no foreign holdings to over 60% of total consolidated IRS activities.

Figure 1 The shift abroad (Q1 2010 – Q4 2015)

Notes: The evolution in the share of interest rate swap activity operated abroad is shown for the top five US dealers (95% of total US activity): Bank of America, Citigroup, Goldman Sachs, JP Morgan, and Morgan Stanley. Fractions were calculated with data from the FED Financial Statement of Foreign Subsidiaries of US Banking Organizations and from the FED Consolidated Financial Statement for Holding Companies. They are equal to the sum of the interest rate swap activity by each of the banks’ foreign subsidiaries over total consolidated interest rate swap activities (resp. over total consolidated foreign exchange swap activities) of the bank taken from the calls reports.

Figure 2 suggests that this pattern extends to all US banks and illustrates the geographic reallocation. It shows, in particular, that US banks reduced their security holdings at home, while increasing them in Australia, Japan, the UK, Brazil, China, Hong Kong, and Mexico.

Figure 2 Change in the location of US banks’ interest rate swap activity between Q1 2010 and Q4 2015

Notes: Figure 2 presents the change (in percentage points) of US banks’ consolidated interest rate swap activity in each country of the world between Q1 2010 and Q4 2015. In a given country, this share is calculated as the interest rate swap activity of the top five US dealers in this country relative to their total interest rate swap activity multiplied by 100. Categories are based on quantiles for non-zero data. Source: FED Financial Statement of Foreign Subsidiaries of US Banking Organizations and FED Consolidated Financial Statement for Holding Companies and Calls reports.

In our empirical analysis, we relate this reallocation to the adoption of the G20 reform in the country of the foreign subsidiary. We find that US banks’ security holdings are indeed negatively related to a more advanced implementation of the reform. Even in our most conservative specification, we find the link to be statistically and economically strong: the adoption of an additional reform block in a given country reduces US foreign subsidiaries’ security holdings in this country by more than the sample average. Supporting this causal interpretation is the fact that we find no comparable reallocation for foreign currency swap holdings, which are much less affected by the reform.

Because the different blocks of the reform are not all equally costly for banks and because some – like the promotion of electronic trading – also produce direct benefits, we further investigate the role of the separate dimensions of the reform. This exercise reveals that the US traders are primarily concerned about the adoption of central clearing, presumably because it is particularly costly and generates the least direct benefits.

The most important challenges to identifying the effect of reform progress are (a) the concern that countries might actively delay the reform to attract business from the American traders and (b) that unobserved country characteristics simultaneously explain the traders’ reallocation and the slow progress of the G20 reform. To address these challenges, we show that the drivers of reform progress are primarily structural and broadly associated with institutional quality and the development of local derivatives markets.

This insight not only allows us to alleviate concerns about the endogeneity of reform progress by controlling for structural and time-invariant country characteristics, it also suggests that the geographic reallocation is problematic. US banks are not moving their trading activities because they target more advanced markets; much to the contrary, they instead move into jurisdictions that are less regulated, ceteris paribus. As a result, the arbitrage does not move risk around neutrally but instead increases the fragility of the global derivatives market.

Arbitrage beyond OTC derivatives

Despite the economically important effect on geographic arbitrage, banks might not be able to move all of their interest rate swap activity out of more tightly regulated jurisdictions. In this case, they might be inclined to compensate for the higher transaction cost related to the reform, for instance by increasing their risk-taking in other dimensions, to maintain a target return on equity. To test this, we investigate the link between the volatilities of the foreign subsidiaries’ overall trading portfolios and progress of the derivatives market reform.

Consistent with the risk-shifting hypothesis, we find that US subsidiaries indeed hold riskier trading portfolios in jurisdictions that have progressed further on the derivatives market reform. In other words, the reform not only redistributes IRS activity towards less regulated foreign subsidiaries, but it also redistributes risk to the broader trading portfolio once the regulation tightens.

Concluding remarks

Existing work on the reform of the OTC derivatives market has focused primarily on implications for market efficiency and systemic risk (Benos et al. 2016, Ghamami and Glasserman 2016, Duffie 2017, Faruqui et al. 2018).

Our paper complements this work by first analysing the factors that drive cross-border differences in the implementation of the reform and by then linking the resulting heterogeneity in reform progress to cross-jurisdictional arbitrage.

This allows us to identify the regulatory response to the crisis – to the extent that it is unevenly implemented – as a source of new and unchecked risks. At least during its implementation phase, the reform has likely contributed to increased cross-border risk-taking.

From a regulatory perspective, our results therefore highlight the global risks associated with loopholes in national regulation. From a governance perspective, they emphasise the importance of coordinated and swift action.


Acharya, V and R Engle (2009), “A Case for (even) more Transparency in the OTC markets”,, 29 August.

Benos, E, R G Payne and M Vasios (2016), “Centralized Trading, Transparency and Interest Rate Swap Market Liquidity: Evidence from the Implementation of the Dodd-Frank Act”, Bank of England Working Paper No. 580.

Deloitte (2014), “OTC Derivatives. The New Cost of Trading”, Report EMEA Centre for Regulatory Strategy.

Duffie, D (2017), “Financial Regulatory Reform after the Crisis: An Assessment”, Management Science 64:10.

Faruqui, U, W Huang and E Takáts (2018), “Clearing Risks in OTC Derivatives Markets: The CCP-bank nexus”, BIS Quarterly Review.

Gandré, P, M Mariathasan, O Merrouche and S Ongena (2020), “Unintended Consequences of the Global derivatives Market Reform”, CEPR Discussion Paper 14802.

Ghamami, S and P Glasserman (2016), “Does OTC Derivatives Reform Incentivize Central Clearing?”, Office of Financial Research Working Paper.

Tavares, C (2011), “Short Selling and OTC Derivatives Policy Options”,, 9 January.


1 To account for cross-country disparities in the implementation of the reform, we construct indices of regulatory progress from Financial Stability Board reports tracking the implementation of the OTC Derivatives Market Reforms. The subsidiary-level data on security holdings of US banks’ foreign affiliates comes from the FED Financial Statement of Foreign Subsidiaries of US Banking Organizations and from the FED Consolidated Financial Statement for Holding Companies.

1,575 Reads