StuW Sonderheft NeSt 2025

StuW Sonderheft NeSt 2025

Abhandlungen – Steuerwissenschaften

S74

Koch / Scheider – The Known Unknown: Tax Avoidance by European Multinationals

tax affiliates are above the profits of high-tax affiliates. Second, we assume that an affiliate ’ s profit depends also on the size of the local economy. For example, in the absence of profit shift ing, an affiliate from the U.S. is expected to be larger, on aver age, than an affiliate from Slovakia, since the economy is con siderably larger. We hence apply a second proxy of profit shift ing (TRD_res_gdp) , that is defined as the residual between TRD and TRD_subs_gdp (see equation 3b). We are aware that neither of our two proxies identifies profit shifting with perfect accuracy. However, our two estimates describe a reasonable range of the true effect. We have no reason to assume that the observed time trend in these two proxies is subject to systema tic bias. Figure 2 depicts the time trends of the sample averages for TRD and our two proxies for profit shifting, TRD_res_unw and TRD_res_gdp . This figure underlines clearly the importance of splitting TRD , which follows to some extent a contrary trend than our two proxies of profit shifting, at least after 2015.

have reduced their profit shifting in recent years to a level that is significantly below that found in other studies. We have also tested the decline in the average size of TRD_res_unw and TRD_res_gdp between the two time periods for statistical sig nificance. The respective t-tests indicate that the difference in the size of TRD_res_unw between 2011 – 2017 and 2018 – 2020 is statistically significant at the 5 percent significance level, for TRD_res_gdp it turns out statistically significant even at the 1 percent significance level.

Table 3: Foreign tax rate differentials and profit shifting

2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 -0.4% -0.4% 0.4% 0.0% -0.6% -0.1% 0.0% -0.9% -0.7% -1.2%

TRD

TRD_subs_ unw TRD_res_ unw TRD_subs_ gdp TRD_res_ gdp TRD_res_ unw/ETR TRD_res_ gdp/ETR

0.8% 0.8% 1.0% 1.0% 0.8% 1.3% 1.3% -0.3% -0.5% -0.1%

-1.2% -1.2% -0.7% -1.0% -1.4% -1.4% -1.3% -0.6% -0.3% -1.0%

1.5% 1.6% 1.8% 1.7% 1.5% 2.1% 2.1% -0.4% -0.6% -0.3%

-1.9% -2.0% -1.5% -1.7% -2.1% -2.2% -2.1% -0.5% -0.2% -0.9%

Figure 2: Profit shifting and overall effect of different tax rates

-4.7% -4.5% -2.5% -3.9% -6.0% -6.0% -5.6% -2.7% -1.1% -4.3%

-7.6% -7.4% -5.6% -6.8% -9.1% -9.3% -9.1% -2.0% -0.7% -3.7%

This table shows average effects of location choice (TRD_subs_unw, TRD_subs_gdp) and profit differences (TRD_res_unw, TRD_res_gdp) on effective tax rates. One potential concern with our analysis and the finding of less profit shifting from 2018 onwards is that the large tax rate cut in the U.S. may bias our findings. We therefore depict in fig ure 3 the average values for TRD_res_unw and TRD_res_gdp also for two additional sub-samples that only consider firms that have more than 80 percent (or more than 90 percent) of their affiliates located outside the U.S. We observe that our esti mates of profit shifting are smaller over the whole time period for the firms with a small share of U.S. affiliates. Removing firms with many U.S. affiliates, however, does not change the downward trend in profit shifting after 2017 to a substantial extent.

This figure depicts the sample average for TRD, TRD_res_unw and TRD_res_gdp. Table 3 reports the yearly sample averages for TRD and its two sub-components TRD_subs and TRD_res , defined in the two different described ways, in numbers. It also shows that TRD hides indeed two different effects. Before 2018, the statutory tax rates of affiliates are, on average, a magnitude higher than the tax rate in the headquarter ’ s country, as indicated by posi tive average values for TRD_subs_unw and TRD_subs_gdp. This changes from 2018 onwards are probably due to the large tax rate cut in the U.S. The average values for TRD_res_unw and TRD_res_gdp are ne gative in all years considered, indicating that firms allocate their profits to some extent according to tax considerations. On the other hand, however, we do not find any evidence that Eu ropean multinationals reduce their effective tax rates cosider ably through such practices. The average values for these two variables are -1.2 and -1.9 percentage points for the years 2011 to 2017, depending on the weighting scheme applied. This re duction in the effective tax rate corresponds to -4.7 and -7.8 percent of the average effective tax rate in the respective year. Our estimate of profit shifting is thus at the lower bound of the Johansson et al. (2017a) estimates and well below the Tørsløv et al. (2022) estimates. In the post-BEPS period (2018 to 2020), our estimates of profit shifting are significantly smaller, both in absolute terms (-0.5 or -0.6 percentage points) and in relative terms (-2.2 or -2.7 percent of the average effective tax rate). Overall, our results thus suggest that European multinationals

Figure 3: Profit shifting for firms with small share of US affiliates

This figure depicts the sample averages for TRD_res_unw and TRD_res_gdp for the full sample and two sub-samples that consider only firms where more than 80 or 90 percent of all affiliates are located outside the U.S.

4.2. Cross-Sectional Differences In this section, we analyze heterogeneity in our estimates of the size in profit shifting. In particular, we investigate to what ex