As digitalization of the economy accelerates, tax avoidance by multinational enterprises (MNEs) becomes a more serious and sophisticated issue to address. The issue can be attributed to several characteristics of firms in the digital economy, these being: discrepancy between the permanent establishment where the corporate income tax is levied and the location where substantial business activity is performed, high reliance on intangible assets in production that are movable across jurisdictions, and the intricacy of applying transfer pricing rules to new business activities. Due to these features, MNEs have been able to easily shift earned income from a country with high corporate income tax (CIT) rates to one with low CIT rates and lessen their tax burdens by taking advantage of extant principles of the international tax system and bilateral tax treaties between countries.

In order to tackle the base erosion and profit shifting (BEPS) issue of MNEs under the digital economy, the OECD and G20 launched an inclusive framework (IF) joined by 141 countries and has conducted discussions on a new international tax system. As a result, in October 2021, two major revisions on the current international tax system – the Two-Pillar solution on BEPS – were agreed upon by 139 IF member countries, including G20 members, to take effect in 2023.

The IF’s final agreement on the Pillar 1 and Pillar 2 is as follows. Pillar 1’s scope is limited to a group of MNEs whose annual revenues exceed 20 billion euros with above 10% profitability. This allows market countries to levy a portion of MNEs’ profits net of the 10% of the revenues, in proportion to each country’s size of revenues sources. The tax base alloted to eligible countries is called “Amount A.” Pillar 2 intends to impose global minimum CIT rates above 15%, under which MNEs’ affiliates are obliged to pay top-up taxes up to 15% if affiliates in other jurisdictions pay low taxes that amount to the effective CIT rates in short of 15%. Pillar 2 applies to a group of MNEs whose annual revenues exceed 750 million euros. The introduction of this new international tax system will clearly have a major impact on MNEs’ investment and value chain decisions.

This study mainly focuses on the Two-Pillar solution on BEPS, exploring how this affects the investment and production decisions of MNEs, and provides implications on Korea’s FDI policy.

In Chapter 1, we briefly explain the background of the OECD BEPS projects and the main schemes of the Two-Pillar solution finally agreed upon by the IF.

In Chapter 2, we identify which MNEs are encompassed within the scope of Pillar 1 and Pillar 2, and look at the firms’ locations and distributions across industries. Using Orbis financial data on MNEs, we find that Pillar 1 applies to MNEs whose headquarters and affiliates are located mostly in the US, China, and the UK. Also we find that the MNEs in the scope of Pillar 1 are distributed among varied industries uniformly. It also features that more consumer facing businesses are included in the scope of Pillar 1 than automated digital services companies. In the case of Pillar 2, we find that MNEs in the scope of Pillar 2 are located in broader jurisdictions including the US, China, Japan, the UK, Cayman Islands, British Virgin Islands, Hong Kong, and the UAE. We also show that MNEs in the scope of Pillar 2 are distributed among various industries uniformly.

In Chapter 3, we document the evidence of MNEs’ profit-shifting through intangible assets by testing empirical models with Orbis financial data. Our findings indicate that, firstly, as subsidiary companies face higher CIT rates in their jurisdictions, they tend to have less amounts of intangible assets; secondly, in jurisdictions with low CIT rates, the subsidiaries are more likely to have a higher level of profits than those with high CIT rates; thirdly, companies’ profits are likely to be more responsive to the difference between the CIT rates of a parent company and its subsidiary when the industry relies more on intangible assets in production; and lastly, MNEs’ profit-shifting occurs more in the service industry and tax havens than in the manufacturing industry and non tax havens.

In Chapter 4, inspired by Wang (2020)’s profit-shifting model, we newly develop a multi-country general equilibrium model featuring firms’ decisions on profit-shifting to minimize tax burden, as well as making location choices for multinational production and exports given the Two-Pillar solution on BEPS. We consider three scenarios where either Pillar 1 or Pillar 2 are introduced or where they are both introduced to disentangle the economic impacts of Pillar 1 and Pillar 2, separately. From these exercises, we find that CIT rates and an increase in IFDI (OFDI) due to introducing the Two-Pillar solution are positively (negatively) correlated. Korea is expected to benefit from introducing Pillar 1 because MNEs with conduit firms located in other jurisdictions with low CIT rates for tax-planning might relocate their production sites to Korea when the tax advantages are partially offset by Amount A. However, these benefits may be diminished when Pillar 2 is introduced because setting global minimum CIT would increase the overall tax burden of MNEs and thereby reduce the return on R&D investment, FDI incentives for production, and exports. For the case of Korea, the former effect slightly dominates the latter effect; thus, the IFDI in Korea increases after the Two-Pillar solution is introduced.

Based on these results, in Chapter 5, we suggest unleashing regulatory interventions in FDI policy could be more effective than reducing CIT rates to attract foreign investment. Also we suggest the Korean government may play a supportive role in MNEs’ reorganizing value chains when Pillar 2 is introduced and the tax burden imposed on MNEs harms their competitiveness.