Profit Shifting: Drivers and Potential Countermeasures
Disclaimer: This article is a summary of other researchers' work. My objective on doing this is to understand and to keep up with international issues and updates in a more practical way. I hope it will help those who read this that certainly has already helped me. It's a pure summary, I didn't give any opinion on the article. Omission of some part could lead to some level of less comprehension, in that case please refer to the original article.
(Sebastian Beer et al.)
(Sebastian Beer et al.)
Abstract
In trying to explain the drivers of global profit shifting by MNEs authors investigate industry-specific variation in profit shifting and identify determinants thereof. Using the ORBIS database authors show that intangible asset endowment of subsidiaries and the complexity of MNE groups explain aggregate profit shifting trends and tend to drive industry specific results. Moreover, authors incorporate country-specific transfer pricing mitigation measures (documentation requirements) into our analysis.
1 Introduction
A growing body of evidence, comprehensively analyzed by Heckemeyer and Overesch (2012), indicates that MNEs do indeed minimize their tax obligations by shifting profits from high to low tax jurisdictions. Studies typically show that pre-tax profitability of affiliates is decreasing in a jurisdictions tax rate or tax differential with economies hosting other firms in the same MNE group.
While transfer pricing needs to be analyzed looking at both managerial and tax optimization objectives, opportunities to make use of the relocation of assets or aggressive transfer (mis)pricing in the case of R&D based intangibles (Grubert, 2003) and the ease of locating intangibles in low tax subsidiary jurisdictions (Dischinger and Riedel, 2011), the ownership structure of subsidiaries (Weichenrieder, 2009), and the location of parent companies (Dischinger and Riedel, 2010) have been shown to impact profit shifting behavior.
Another potential driver of an MNE’s profit shifting opportunities is the regulatory framework in host or parent countries. Particularly, the absence of transfer pricing regulations or lax enforcement of the arms length principle for related party transactions is likely to be associated with aggressive pricing strategies for profit shifting (Bartelsman and Beetsma, 2003; Beuselinck et al., 2009; Lohse and Riedel, 2012).
This paper aims to add to existing research on profit shifting behavior in two ways: First, it investigates industry-specific variation in profit shifting and identifies determinants thereof. Second, we incorporate country-specific transfer pricing mitigation measures (documentation requirements) into our analysis and find a significant, though lagged, dampening effect on aggregate shifting behavior.
2 Research Hypothesis and Background
In very general terms, our core hypothesis is that MNEs operating in industries and countries with more opportunities and incentives to shift profits post a lower share of earnings in their subsidiaries operating in high tax jurisdictions. To investigate this premise, we look at a range of potential industry- and country-specific determinants.
2.1 Identifying Industry Trends and Drivers
The first part of our analysis aims to uncover the determinants of profit shifting by exploiting industry characteristics. In particular, we test whether the share of intangibles of affiliates and the complexity of the production process (cross-industry activities covered by an MNE group) explain variations in profit shifting. Our emphasis on intangibles, a key source of growth and competitiveness for MNEs, is based on the insight that the valuation of trademarks, copyrights and patents for tax orother purposes is a dauntingly difficult task.
In an early study, using data on U.S. companies, Grubert (2003) identified an association of the volume of intercompany transaction and associated profit shifting opportunities with parent R&D intensity. In addition, Karkinsky and Riedel (2009) showed that patent allocation is affected by the corporate tax rate differential across European MNEs. Importantly, Dischinger and Riedel (2011) find that the allocation of intangibles at the subsidiary level can be partly explained by tax rate differences, and that this allocation corresponds to a larger sensitivity of affiliate profitability to the tax difference. Building on these findings, we test more generally whether the relative endowment of a subsidiary with intangibles provides an opportunity driving profit-shifting activities across key industries.
2.2 Capturing the Enforcement of the Arm’s Length Principle
Over the last two decades, governments have increasingly responded to the threat of corporate tax base erosion through transfer mispricing by introducing provisions to regulate transfer pricing based on the arms length principle and by strengthening the capacity of audit staff. Notably, from 1994-2011, the number of countries requiring detailed reporting of related party transactions has increased more than tenfold from 5 to more than 70 (World Bank 2013).
In a first attempt to capture country specific transfer pricing legislation effects, Bartelsman and Beetsma (2003) report country response coefficients, explaining some of the differences with the strength of transfer pricing regimes. The authors account for the existence of transfer pricing provisions, documentation guidelines and specific penalties linked to transfer pricing. However, limited variation among the observed countries’ enforcement practices in their panel limits the usefulness of these initial findings.
In a first attempt to capture country specific transfer pricing legislation effects, Bartelsman and Beetsma (2003) report country response coefficients, explaining some of the differences with the strength of transfer pricing regimes. The authors account for the existence of transfer pricing provisions, documentation guidelines and specific penalties linked to transfer pricing. However, limited variation among the observed countries’ enforcement practices in their panel limits the usefulness of these initial findings.
More recently Lohse and Riedel (2012) used a more sophisticated scoring system which is presented in Lohse et al. (2012) and captures the existence of transfer pricing regulation, including the extend of documentation obligations in the legislation. Similar to these earlier efforts we assess the effect of regulatory measures on profit shifting activities. We chose this narrow approach as a proper assessment of transfer pricing risks - guiding enforcement efforts - initially requires filling information gaps on transactions with associated parties. The introduction of documentation requirements is therefore commonly assumed to have a strong signaling and compliance effect on multinationals and can be interpreted as a proxy for actual administrative enforcement of the arms length principle.
3 Estimation Approach, Information Sources and Sample Selection
3.1 Sample Selection
Our firm-level micro data on MNEs is extracted from the ORBIS database, commercially offered by Bureau Van Dijk (BvD). The database offers administrative information for more than 50 million private companies globally. The data is collected by national institutions based on legal requirements, and compiled and standardized by BvD. The database provides information on parents and global subsidiaries, allowing for the construction of a comprehensive MNE panel. We use the following variables reported in ORBIS: Fixed assets, fixed intangible assets, total assets, sales, EBIT, number of employees, cost of employees.
We cover the time period from 2003-2011, providing nine years of firm specific information. In order to classify MNE groups, we start with the immediate ownership indicated in ORBIS dropping all parent observations (and associated subsidiaries) with consolidated accounts, as we require information on the specific activities of each parent and subsidiary.
We select all parents owning a foreign subsidiary with at least 50% of its shares, which amounts to around 42,048 immediate parents owning 138,115 subsidiaries (domestic and foreign). Our analysis is focused on subsidiaries only. Approximately 5000 of the companies in the subsidiary sample are also parents. We eliminate these from our sample but add them again as part of our robustness checks. We limit our sample to observations in OECD member states, though this still includes global information on the tax differences for each MNE group.
3.2 Industry Classification
We rely on the European Classification of Economic Activities (NACE) for the ordering of MNEs across sectors. The starting point for our classification of an MNE groups industry is the industry assignment of each affiliate provided in the ORBIS database. Generally, we aggregate industries at the first level of the alphabetical NACE codes. For sectors, such as manufacturing, where our sample includes more than 1000 MNEs at the first level, we also differentiate activities at the second and/or third level in the NACE hierarchy. Moreover, we group all affiliates designated as holding, head office activity, trust, funds and similar financial entities together in a single holding category and discard those industries with less than 10 multinationals in our sectoral analysis. For all affiliates across our sample we differentiate a total of 40 industries (see Annex 2). A simple descriptive analysis of our data presented in Figure 1 below, reveals noticeable differences in both intangible endowments of affiliates across different sectors as well as a major variance in the complexity of the MNE groups these affiliates belong to.
3.3 Documentation Requirements
Our information on the introduction of documentation rules is taken from several sources, including the Transfer Pricing Guide of the International Bureau of Fiscal Documentation (IBFD), transfer pricing country summaries prepared by PricewaterhouseCooper, Ernst and Young, Deloitte, and KPMG, as well as country assessments in transfer pricing journals. An overview of our country classification based on these sources is provided in the Annex. Reflecting the increasing importance of transfer pricing regulation, the share of observations in our sample that has been covered by documentation rules has risen from 12 percent in 2003 to about 80 percent in 2011.
4 Empirical Results
4.1 Aggregate findings
Our results are depicted in Tables 2 and 3 below. The first specification in table 2 illustrates the extent of profit-shifting in our sample. In line with findings reported by Heckemeyer and Overesch (2012) in a recent meta-study, we find a significant negative coefficient with a semi-elasticity of 1.02. This implies that an increase in a jurisdictions Corporate Income Tax rate by one percentage point results in an average decrease of subsidiary EBIT by 1.02 percent. As expected, our subsidiary input variables (fixed assets and cost of employees) as well as the aggregate assets held by the whole MNE group are explaining changes in EBIT across our sample.
In specification 2 and 3 we display our findings on the drivers of profit shifting. The second column presents the effect of the ratio of intangibles over total assets held by a subsidiary. The interaction term is negative with a magnitude of 3.34 and significant at the five percent level. This finding indicates a higher semi-elasticity of profits derived from intangibles at the subsidiary level. In the third specification we try to approach supply-chain complexity by constructing a variable on the industry coverage within the MNE group. The main assumption being that the more complex an MNE structure is, the more opportunities there are to take advantage of profit shifting strategies. This measure is obviously strongly correlated to firm size, but by including total aggregate MNE group assets into our specifications we isolate the effect of complexity.
4.2 Mitigation Results
The first specification re-estimates the sensitivity of EBIT to CIT changes. The second specification introduces an interaction of the tax difference with the variable capturing the existence of documentation rules. Taking the non-monotonicity of its effects into account, we find that the introduction of documentation rules has initially a negative, but not significant, effect on subsidiary profitability. The second-order term, however, is positive and statistically significant with a p-value of 0.003. This indicates that documentation rules become effective in mitigating profit shifting with a time lag.
A little more than two years after their introduction, we observe that documentation starts reducing overall profit shifting in the economies in our sample. The semi-elasticity of subsidiary EBIT is reduced by 0.64 within 4 years following enforcement through documentation rules, translating into mitigation of about 60 percent of the observed profit shifting among MNE subsidiaries in our sample.
5 Conclusion
In trying to explain the drivers of global profit shifting by MNEs, we analyzed the role of intangible assets and the complexity of MNE group structures. We find that subsidiaries with a high intangible to total asset ratio have a semi-elasticity of 1.2 compared to 0.78 for low intangible affiliates, suggesting a significantly larger sensitivity to CIT rate changes. Similarly, subsidiaries belonging to more complex MNE groups have a higher semi-elasticity (1.11) than those that are part of less complex entities (0.81).
Further differentiation for 40 industries confirmed important differences in the magnitude of shifting behavior across sectors, largely following our predictions. Additionally, our findings on the documentation rules support and extend the observations of earlier studies highlighting the importance of domestic enforcement efforts. Focusing on the introduction of documentation requirements, a narrow but straightforward proxy for the enforcement of transfer pricing provisions at the national level, we find significant non-linear mitigation effects. On average, the estimated profit shifting among MNE subsidiaries in our sample is reduced by 60 percent four years after the introduction of mandatory documentation requirements. Our findings suggest, however, that documentation rules do not help address profit shifting risks of intangible intensive subsidiaries. Taking the perspective of host country tax administrations, the findings of our research provide initial insights on the relative profit-shifting risk associated with different sectors of MNE activities.
Further differentiation for 40 industries confirmed important differences in the magnitude of shifting behavior across sectors, largely following our predictions. Additionally, our findings on the documentation rules support and extend the observations of earlier studies highlighting the importance of domestic enforcement efforts. Focusing on the introduction of documentation requirements, a narrow but straightforward proxy for the enforcement of transfer pricing provisions at the national level, we find significant non-linear mitigation effects. On average, the estimated profit shifting among MNE subsidiaries in our sample is reduced by 60 percent four years after the introduction of mandatory documentation requirements. Our findings suggest, however, that documentation rules do not help address profit shifting risks of intangible intensive subsidiaries. Taking the perspective of host country tax administrations, the findings of our research provide initial insights on the relative profit-shifting risk associated with different sectors of MNE activities.
Comments
Post a Comment