Debate over a shake-up of the international corporate tax system has been ongoing for many years now, with few concrete outcomes (Figure 9). But this time looks different. The pandemic has had a profound effect on public finances around the world. And with governments scrabbling for additional revenues, corporate taxes are an attractive source.
Moreover, public dissatisfaction with aggressive tax planning by multinationals has grown, and with the pandemic accelerating the trend towards digitalisation, the need to address the tax challenges posed by the digital economy is becoming more acute. In this context, it’s no surprise that the new US administration’s desire to change the system has been backed by some of the key countries. Indeed, this month’s G7 deal marks a shift in the momentum toward global tax reform.
In recent months a series of at times competing proposals have been put forward. But the OECD/G20 Inclusive Framework on Base Erosion and Profit Shifting (BEPS 2.0) launched in 2016 remains the main vehicle for global reform. To date 139 countries and jurisdictions have joined to date this large forum, which follow a two-pillar approach:
- Pillar One expands the taxing rights of countries where there is an active and sustained participation of a business.
- Pillar Two is designed to ensure that large multinationals pay a minimum level of tax regardless of where they are headquartered or operate.
The OECD is targeting a successful conclusion of negotiations by mid-2021, meaning a lot is at stake in the coming weeks as a number of design elements and parameters still need to be agreed (Figure 10). In particular, there is still no consensus on the scope of Pillar One in terms of the categorisation of business activity and revenue threshold, and on how much residual profit can be reallocated. The level of the global minimum tax rate in Pillar Two is also yet to be decided, though political discussions appear to be moving toward a 15% rate, which is substantially lower than the 21% recently proposed by the US.
Winners and losers under BEPS 2.0
For the European countries for which data is available, we have identified the winners and losers under BEPS 2.0 based on five key metrics: (1) reliance on corporate tax revenues; (2) corporate tax rates; (3) measures of allocated profits and customers; (4) foreign direct investment positions; and (5) the contribution of multinationals to the real economy.
The key takeaway from Figure 2 is that, with a few notable exceptions, corporate taxes are a modest source of revenue for most countries. In 2019, corporate taxes generated around 15% of Luxembourg’s total tax revenue and 14% in Ireland (which rose to nearly 21% in 2020). The Netherlands had a smaller, but still important, 9% share of tax in 2019.
In terms of identifying the countries that will be particularly impacted by Pillar One, we examined the rich US IRS dataset that includes where US multinationals book profits in Europe, and compare these results with the location of Facebook users and e-commerce revenues – two business areas which will likely be in scope for BEPS 2.0. Figure 4 shows that in 2018 over one-third of multinational profits were booked in Netherlands, Ireland, and Luxembourg, but in aggregate these countries amount to less than 5% of Facebook users and e-commerce revenues. BEPS Pillar One would result in a loss of tax base for these countries. Meanwhile, Germany, France, Italy, and Spain stand to be the main winners. These countries’ combined shares of US multinational profits booked in Europe stood at 7.6%, while their combined shares of Facebook customers and e-commerce revenues were above 40%.
Analysis of foreign investment also provides important insight, with Luxembourg, Ireland, and the Netherlands standing out again as having accumulated particularly large FDI stocks relative to the size of their respective economies (Figure 5). Moreover, these countries record outward FDI positions that roughly match, and in the case of Luxembourg and the Netherlands surpass, their inward FDI positions, consistent with the idea that a considerable amount of investment is driven by multinational tax planning strategies.
However, overlooking the large contributions of foreign-owned multinationals to the economies of the Netherlands, and particularly Luxembourg and Ireland would be unfair (Figure 6). These companies accounted for 44% of Luxembourg’s overall employment in 2018, 18% in Ireland, and 14% in the Netherlands, in all cases comfortably above the EU average of 11%. In the same year, foreign-owned multinationals accounted for nearly 26% of Luxembourg’s gross value added (GVA), around 43% in Ireland, and nearly 18% in the Netherlands.
Beyond the impact on public finances, itis clear that the reform process could affect these countries’ economies and employment, particularly if multinationals relocate profits and investments as a result. It is also worth drawing attention to Hungary, which has the lowest effective average corporate tax rate in our sample. While Hungary has a relatively low reliance on multinationals for tax revenue, foreign-owned companies make important contributions to its economy, meaning that the impact of BEPS will overall be negative.
The costs to the losers will be high
Still, the ultimate economic and fiscal effects of the BEPS reforms are far from clear, as some of the key parameters still need to be agreed. Importantly, the impact will also depend on how individual companies and countries respond to the changes.
Either way, a plausible scenario (detailed in the methodology section below) suggests that the BEPS 2.0 process will have relatively large impacts on the public finances of Luxembourg, Ireland, and the Netherlands. Relative to a no-change scenario, the BEPS 2.0 scenario will push these countries’ debt ratios higher (Figure 7), with a cumulative impact on debt-to-GDP of over 6ppts by 2028 in Luxembourg, below 5ppts in the Netherlands, and 2.4ppts in Hungary. For Ireland, the hit to the debt-to-GDP ratio is below 5ppts. But using modified gross national income (GNI*) – a more appropriate metric in this case, given the known multinational-related distortions to GDP in Ireland – the hit to the debt-to-GNI* ratio is larger – at nearly 9ppts by 2028.
Based on the GNI* metric, our scenario shows Ireland will also be one of the most highly indebted countries in Europe (Figure 8), with debt-to-GNI* standing at over 120%. What’s more, we think that worse outcomes not just for Ireland, but also for Luxembourg and the Netherlands, are possible. The economic performance of these countries is especially dependant on their ability to attract foreign investment, which could be undermined by the tax reforms. A larger-than-anticipated shock to economic growth, while hugely significant in itself, would also have severe knock-on
Beyond the impact on new financial flows, what happens to the stock of prior investments will be just as important, if not more so. For the timescale of our scenario, we haven’t assumed widespread relocations away from Ireland, the Netherlands, Luxembourg and Hungary. Relocation costs can be large and we expect multinationals already located in these countries will still face similar, and in some cases more favourable, tax treatment than in other European countries. But there are huge risks here. For example, corporate tax receipts in Ireland are highly concentrated – the top ten foreign-owned multinationals accounted for 56% of all corporate tax revenue in 2019. Analysis by the Irish Fiscal Advisory Council highlights that if half of these firms relocated, Ireland’s debt ratio would be 7ppts higher by 2025 compared to 4ppts in our scenario. If the downside risks were to materialise, the Irish Government would likely have to consider difficult trade-offs, such as raising the corporate tax rate to the new global minimum to compensate for the loss of the tax base, or tightening fiscal policy more generally.
Meanwhile, for the winners the first observation is that they easily outnumber the losers. Overall, the impact for the eurozone will be net positive. But the largest gains will accrue to France and Germany, followed by Italy and Spain (Figure 1). Relative to a no-change scenario, we estimate that by 2028 debt-to-GDP ratios for Germany and France will be 1.4ppts lower, for Italy 1ppt, and Spain 0.8pps. These are above average gains, considering the debt ratio for the eurozone will be 0.4ppts lower in our scenario.
And in the case of the winners, better outcomes are possible. We used the invaluable OECD economic impact assessment to calibrate the revenue increases. But a more optimistic take on the revenue potential from corporate tax reform can be found in the literature, including a recent report from Zucman et al (2021).
Next month’s G20 meeting will be pivotal in setting the basis for a global tax agreement, although a fully-fledged deal by then looks ambitious at this stage. Once a deal is struck, the focus will then turn to the ratification process. The European Commission intends to propose a new directive to ensure uniform implementation of Pillar One, while according to the European economy commissioner Paolo Gentiloni Pillar Two will require changes to existing directives and new legislation. So despite the positive momentum, final implementation could take some time.
Figure 9: Recent reform proposals have had limited success
Figure 10: Key elements of BEPS Pillar One and Two