The global minimum tax represents the most ambitious international effort in decades to curb profit shifting to tax havens. But how much profit shifting actually occurs, and how is it affected by taxation? This column argues that ignoring the large number of multinationals that shift all their profits out of high-tax countries understates the true scale of profit shifting by roughly half, and revenue projections based solely on responses to tax rate changes by partial shifters are too conservative. Profit shifting doesn’t just affect tax revenues – it fundamentally alters investment incentives, creating fundamental trade-offs that cannot be ignored.
President Trump’s second-term tax agenda has already reshaped the global fiscal landscape – most notably through the passage of the “One Big Beautiful Bill Act” (OBBBA) and a sharp, if legally contested, escalation in tariff use. But arguably the most consequential change for the structure of the international tax system has been his decision to withdraw the US from the multilateral agreement, negotiated by more than 140 countries under the OECD’s Inclusive Framework, to introduce a 15% global minimum tax (GMT) on multinational enterprises (MNEs).
The GMT represents the most ambitious international effort in decades to curb profit shifting to tax havens. At the time of writing, 57 countries have already implemented measures to introduce the tax, and ten more have implementation in progress. 1 But what, exactly, is at stake – for countries which have introduced the GMT, and for those that have not? How much profit shifting actually occurs? How is real activity affected by taxation, and how does that change when multinationals move paper profits across borders? And how costly are the elaborate tax-avoidance structures that make such shifting possible?
Economists have long debated the magnitude of profit shifting. Empirical estimates date back to Hines and Rice’s (1994) work that assumed that the more an MNE grows its profit shifting activities, the more and more costly it becomes to shift the next dollar of profit to tax havens. Meta-analyses of firm-level studies (Beer et al. 2019, Heckemeyer and Overesch 2017) suggest that reported taxable profits fall by a little over 1% when the tax rate rises by one percentage point – a modest semi-elasticity. Yet macroeconomic analyses using country-level data (Clausing 2016, Tørsløv et al. 2022) find far larger responses. Why do the two literatures disagree?
Our recent research (Bilicka et al. 2025) helps resolve this puzzle by identifying what has been missing from firm-level studies, namely, a more realistic treatment of global MNEs’ tax avoidance practices and acknowledging the large number of multinationals that shift all their profits out of high-tax countries. These ‘full shifters’ behave fundamentally differently from firms that shift only part of their profits, and ignoring them systematically understates the true scale of profit shifting by roughly half.
To explore this behaviour, we build a structural model in which each multinational decides not only how much capital to invest in each country, but also how much to invest in a tax-avoidance asset: the firm’s internal capability to move profits to low-tax locations. This ‘tax avoidance asset’ acts as a public good within the multinational: once established, it reduces the variable cost of shifting profit for all subsidiaries worldwide.
The idea reflects companies’ real-life policy choices. A company that owns valuable intellectual property might establish a cost-sharing agreement that allocates patent ownership to its Caribbean affiliate. Setting up such an arrangement might be expensive, but once in place, the marginal cost of routing income through that structure is small. Firms without such capacity face much higher marginal costs and may remain taxable in high-rate countries. Recognising the initial investment into avoidance and its impact on the variable profit-shifting cost changes the predicted responses to tax reforms.
Our study distinguishes between fixed and variable costs of profit shifting. Using corporate tax return data from the UK, we estimate that the fixed costs (of setting up the structure) are 1.5% of the true tax base. The variable costs (maintaining it and shifting each additional pound) add another 3%. These costs vary across firms depending on their business models and opportunities for shifting. For firms with low costs, full shifting is optimal. For others, partial shifting makes sense. And for some, the costs outweigh the benefits entirely. These estimates highlight the significant resources MNEs allocate to tax avoidance – lawyers, transfer-pricing specialists, and complex intra-group arrangements.
The standard economic model of profit shifting treats avoidance as a smooth, continuous response: a firm shifts a little more profit when tax differentials widen. Empirical studies using microdata typically estimate this relationship by regressing reported profits on tax rates, implicitly assuming that every firm adjusts at the margin.
In practice, however, many firms report zero taxable profits year after year. There are, of course, business-as-usual reasons for such patterns alongside profit shifting, including true company losses that can be carried over to multiple tax periods. But even when we account for these ‘true’ losses, there remains a substantial share of MNEs reporting zero taxable profits persistently. This is because, once a multinational has established the structure that allows them to move profits to low-tax jurisdictions, the marginal cost of shifting an additional dollar may be small.
This pattern – some firms shifting intensively, others completely – creates a wedge between micro-level and macro-level estimates of the semi-elasticity of profit shifting. The latter capture both the intensive response among partial shifters and the extensive decision to exit the tax base entirely. Our goal was to model both margins jointly and quantify the underlying costs that generate such heterogeneity.
The model reconciles the conflicting evidence from previous studies. Using UK tax data, we estimate that a one percentage point increase in the tax rate reduces reported profits by 2.4%. This matches the range of ‘macro’ estimates, which typically find elasticities of between 2.5 and 5.2 (Clausing 2016, Hines and Rice 1994). But if we replicate the approach of typical firm-level studies – focusing only on firms with positive taxable income – we get an elasticity of just 1.3, closely matching the consensus from micro studies of around 1.1 to 1.4 (Beer et al. 2019, Heckemeyer and Overesch 2017). The gap therefore stems from ignoring firms that have exited the tax base altogether.
Profit shifting doesn’t just affect tax revenues – it fundamentally alters investment incentives. Our model recognises that profit shifting reduces the effective tax rate on capital, and so in turn reduces the cost of capital for investment in real productive assets. As a result, multinationals that engage in aggressive profit shifting respond much less to statutory tax rate changes when making investment decisions. A one percentage point tax increase reduces their capital investment by just 0.8%. Firms that don’t shift profits respond more than twice as strongly, cutting investment by 1.8%. This happens because profit shifting dampens the tax impact, keeping the effective tax rate low even when statutory rates rise.
Policies that successfully curb profit shifting will raise the effective tax rate multinationals face, potentially reducing investment. We gain revenue and reduce wasteful avoidance costs, but at the price of higher costs of capital and potentially lower investment.
We test our model through a quasi-natural experiment: the 2002 Italian Controlled Foreign Company (CFC) reform, which tightened rules on profit shifting to tax havens. CFC rules operate, for the purpose of our model, in a very similar way to the way in which current minimum tax rules work. Using a difference-in-differences approach, we find that the reform significantly reduced profit shifting by Italian MNEs operating in the UK. This led to higher reported UK taxable income but had limited short-term effects on investment. These results align with our model’s prediction that curbing profit shifting increases the cost of capital for productive assets, potentially dampening investment in the long run.
Our framework also allows us to conduct counterfactual analyses of tax policy scenarios, offering insights into the trade-offs faced by governments:
Our findings highlight that the optimal tax policy depends on balancing these trade-offs. For instance, while a GMT creates revenue gains by discouraging profit shifting, its broader impact on productive investment varies across countries. In Figure 1, we demonstrate that there are efficiency gains from introducing a GMT at moderate levels, and that the optimal rate varies depending on the implementing high-tax country’s own tax rate. The initial gains for low levels of GMT thresholds arise from the elimination of costly tax avoidance investments. The dashed lines in the figure show the net present value of MNE cash flows (denoted V) and the smooth lines indicate global welfare. Based on our simple measure of welfare as a tax revenue-investment trade-off, for a country with a 30% home corporate tax rate (the black lines), the optimal GMT rate is 18%, while this rate is 13% for a country with a 20% home corporate tax rate (the blue lines).
Figure 1 Impact of global minimum tax at varying threshold rates on welfare
As countries implement the global minimum tax and debate further reforms, two lessons emerge. First, the scale of profit shifting may be larger than many firm-level studies suggest – perhaps twice as large – and revenue projections based solely on intensive-margin responses will be too conservative. Second, the connection between profit shifting and investment creates fundamental trade-offs that cannot be ignored. In implementing policies aimed at eliminating profit shifting, policymakers need to balance revenue needs, efficiency gains from reduced avoidance costs, and the effects on real economic activity.
Recognising the importance of extensive margin helps reconcile conflicting evidence, explains why multinationals respond so strongly to tax changes in aggregate, and clarifies the true trade-offs facing policymakers as they reshape international taxation for the 21st century.
Source: cepr.org
While globalisation has increased productivity and incomes around the world, extensively linked value chains have…
Textbook monetary economics views inflation as fundamentally driven by aggregate shocks, such as money supply…
Artificial intelligence differs from other technological advancements in finance, such as the initial adoption of…
Industrial raw materials such as nickel, cobalt, and rare earths are critical inputs in countless…
As European governments scale up investment, bond market stability is more critical than ever. This…
Economists have long warned of the negative consequences of excessive US public debt (e.g. Friedman…