U.S. tech companies heavily laden with intellectual property have long taken advantage of opportunities to engage in cross-border tax planning. This is evidenced most plainly by their low effective tax rates and the large proportion of their profits outside the United States. The tax community understandably has been preoccupied with the new and complex international tax provisions that were supposed to put a damper on all that. But central to understanding the Tax Cuts and Jobs Act’s effect on tech companies is its 14 percentage point cut in the corporate tax rate.
Data presented here show that large U.S. tech companies generally have improved their tax position since passage of the reform. But in what is turning out to be a major reversal of fortune, many of those that were most aggressive previously at shifting profits have gained the least from the TCJA.
Company responses have varied. Early indications are that generally, profit shifting out of the United States has halted and even reversed. Figure 1 shows that the average foreign share of non-U.S. profits dropped from 66 percent in the three-year period from 2015 to 2017 to 61.2 percent during 2018 and 2019. That suggests that the compression of the differentials between U.S. and foreign corporate taxes — attributable to the reduction of the U.S. rate from 35 percent to 21 percent; the extra tax burden on (foreign) global intangible low-taxed income; and the new relief for (domestic) foreign-derived intangible income — may be having the expected effect on where companies locate their profits. But, as should be expected when so much depends on individual company facts and circumstances, some taxpayers have enhanced the benefits from the corporate rate cut by increasing their share of foreign profits outside the United States.
Big Money
Our data is from 20 U.S. tech companies that in their most recent fiscal year reported a total of $298 billion in worldwide before-tax profits and made $57 billion in cash payments of income tax to all governments.
The statistics in Table 1 combine the past five years of annual data for our 20 companies. The first three columns of percentages are effective tax rates. Like all the data presented in this article (all from company 10-K reports), they are computed from accounting data. Except for cash payments of tax, these financial statement data can and often do stray far from data that would be found on tax returns. At every juncture we must remind ourselves that data from financial reporting can sometimes approximate what is needed to answer tax policy questions, but at the same time it can be wildly off that mark.
The prime, but hardly only, example is effective tax rates reported by companies used to determine their all-important net income figures that are highlighted by Wall Street and the financial press. They are often a poor reflection of the pure effective tax rate by which some ideal measure of tax burden is divided by some ideal measure of profits.
A major shortcoming of book effective tax rates from a tax policy perspective is their short-term volatility due to large one-time charges to the income tax provision attributable to, for example, changes in valuation allowances; changes in unrecognized tax benefits; settlement with the IRS; and revaluation of net deferred tax assets with a change in the U.S. statutory rates. Such one-time changes, when reported in the tax rate reconciliation table of the 10-K report’s income tax footnote, are removed to arrive at the adjusted effective tax rate reported in the first column of Table 1.
The cash effective tax rate is reported in the second column of Table 1. Of these three measures, the adjusted effective tax rate seems the best indicator using data available from 10-K reports to measure a company’s tax burden. (Dell Technologies Inc. is dropped from any further calculations because its reported negative before-tax income in all years makes effective tax rates impossible to correctly interpret.) The average adjusted effective tax rate over the previous five years is 16.4 percent. There is a wide variance, from a low of 4 percent for Micron Technology Inc. to 28.8 percent for Amazon.com Inc. By this measure, out of 20 tech companies, Micron Technology has had the lowest tax burden.
Company | Latest Fiscal Year-End | Adjusted ETR | Cash ETR | Reported ETR | Effect of Foreign Rate on ETR | Foreign Percentage of Before-Tax Income |
|---|---|---|---|---|---|---|
Micron Technology Inc. | Aug. 2019 | 4% | 3.5% | 4% | -18.7% | 99.1% |
Thermo Fisher Scientific Inc. | Dec. 2019 | 6.2% | NA | 6.2% | -11.7% | 59.1% |
Nvidia Corp. | Jan. 2020 | 7.5% | 2.3% | 3.5% | -13.1% | 62.3% |
IBM Corp. | Dec. 2019 | 10% | 15% | 19.5% | -17% | 82.9% |
Applied Materials Inc. | Oct. 2019 | 10.8% | 8.7% | 18.1% | -17.2% | 86.3% |
Qualcomm Inc. | Sept. 2019 | 11.6% | 22.8% | 46.9% | -15.9% | 56.8% |
Adobe Inc. | Nov. 2019 | 12.5% | 13.5% | 13.5% | -10% | 67.1% |
Alphabet Inc. | Dec. 2019 | 14.3% | 17.4% | 21.9% | -9% | 56.6% |
Microsoft Corp. | June 2020 | 17.4% | 17.3% | 22.6% | -7.3% | 66.4% |
Oracle Corp. | May 2020 | 17.5% | 20.1% | 27.9% | -8.8% | 68.9% |
Netflix Inc. | Dec. 2019 | 18.7% | 16.8% | 5.5% | 1.3% | 28.3% |
Cisco Systems Inc. | July 2020 | 18.7% | 23.1% | 35.2% | -7.6% | 63.8% |
Facebook Inc. | Dec. 2019 | 19% | 14% | 21.3% | -9.4% | 66.1% |
Texas Instruments | Dec. 2019 | 20.6% | 19.5% | 24.5% | -1.7% | 17.2% |
Apple Inc. | Sept. 2019 | 21.7% | 18.1% | 22.1% | -4.4% | 67% |
HP Inc. | Oct. 2019 | 22.2% | 19.1% | -8% | -8.2% | 100.7% |
Intel Corp. | Dec. 2019 | 22.6% | 14.8% | 22.6% | -6.2% | 41.6% |
Cognizant Technology Solutions Corp. | Dec. 2019 | 26.8% | 27.8% | 30.7% | -3.3% | 66.6% |
Amazon.com Inc. | Dec. 2019 | 28.8% | 10.7% | 19.5% | 0.2% | -6.7% |
Dell Technologies Inc. | Jan. 2020 | --- | --- | 67.5% | --- | --- |
Average of above |
| 16.4% | 16.9% | 18.8% | -8.9% | 60.5% |
Source: Author’s calculation using company 10-K report data. | ||||||
The fourth column of Table 1 shows the five-year average reported impact of lower foreign tax rates on a company’s effective tax rate. So, for example, Microsoft Corp.’s reported average of -7.3 percent means its effective tax rate would be 7.3 percentage points higher if foreign profits were taxed at the U.S. statutory rate. The average of these foreign rate impacts on effective tax rates is -8.9 percent. They vary from -18.7 percentage points for Micron Technology to 0.2 percent for Amazon.
The last column of Table 1 shows the percentage of foreign profits as a percentage of worldwide profits over the past five years. Most companies have foreign shares well above 50 percent. The more prominent exceptions are Netflix Inc. (28.3 percent), Texas Instruments (17.2 percent), and Amazon (-6.7 percent). These relatively low amounts of foreign profits suggest the companies aren’t aggressive in their tax planning. (Amazon’s negative foreign share of profits means it has foreign losses.)
The Effect of the TCJA
Table 2A shows how the measures introduced in Table 1 have changed since passage of the TCJA. For each company, the weighted average of the most recent two years of available data is subtracted from the weighted average of the two fiscal years ending in 2015 and 2016. (The use of weighted averages is equivalent to merging two years of data and treating it as a single period.) The first numerical column of Table 2A shows change in adjusted effective tax rates. By this measure, in our sample of tech companies, Amazon was the biggest winner after passage of the TCJA with a reduction in its adjusted effective tax rate of 17.5 percent. Except for Thermo Fisher Scientific Inc. and Micron Technology, all our tech companies reduced their adjusted effective tax rates after passage of the TCJA.
Using Amazon as an example, let’s speculate about what factors might explain large beneficial changes after passage of the TCJA. We can see from the fifth column that all Amazon’s income is from U.S. operations. This makes it possible for it to get a large amount of benefits from a U.S. corporate rate reduction.
What about second-place finisher Adobe Inc.? Here we can see from the fifth column that only about one-third of its profits were domestic. Corporate rate reduction will help, but it won’t be anywhere near the magnitude of the benefit it can provide Amazon. Adobe, however, has an alternative path to tax relief. The fifth column of Table 2A shows us that after passage of the TCJA, Adobe increased its foreign profit share by a whopping 44 percentage points. If that increase in foreign profits is attributable to jurisdictions with tax rates that are significantly below the 21 percent U.S. tax rate, that could explain a large part of the company’s reduced effective tax rate.
Company | Change in Adjusted ETR | Change in Cash ETR | Change in Reported ETR | Change in Effect of Foreign Rate on ETR | Level of Foreign Percentage of Before-Tax Income | Change in Foreign Percentage of Before-Tax Income |
|---|---|---|---|---|---|---|
Amazon.com Inc. | -17.5% | -4.4% | -29.3% | -0.3% | -6.7% | 26.5% |
Adobe Inc. | -16.3% | -10.2% | -14.5% | -0.5% | 67.1% | 44.1% |
Texas Instruments | -14.6% | -15.1% | -13.5% | 4% | 17.2% | -6.9% |
Facebook Inc. | -12.8% | 9.9% | -6.6% | 3% | 66.1% | 20.9% |
HP Inc. | -12.6% | -3.1% | -65.6% | -3.5% | 100.7% | 23.5% |
Apple Inc. | -9.9% | 0.8% | -8.9% | -5.5% | 67% | 0.3% |
IBM Corp. | -9.7% | 4.6% | 4.7% | 8.8% | 82.9% | 32.4% |
Nvidia Corp. | -9.6% | 2.4% | -14.9% | 5.1% | 62.3% | -8.3% |
Intel Corp. | -8.9% | -3.4% | -8.8% | 6.1% | 41.6% | -2.1% |
Oracle Corp. | -8.2% | 6.4% | -7.7% | 4.9% | 68.9% | 1.5% |
Netflix Inc. | -6.1% | 2.1% | -16.7% | -1.3% | 28.3% | -7.5% |
Alphabet Inc. | -5.6% | 6.5% | -5.5% | 6.8% | 56.6% | 3.1% |
Cognizant Technology Solutions Corp. | -5.2% | -4% | -4.6% | 8.2% | 66.6% | -2.2% |
Applied Materials Inc. | -4.3% | -2.5% | 10.8% | 9.2% | 86.3% | 13.4% |
Cisco Systems Inc. | -3.9% | -0.1% | 0.7% | 10.9% | 63.8% | -32% |
Microsoft Corp. | -3.6% | 10.3% | -3.5% | 9.1% | 66.4% | -19.7% |
Qualcomm Inc. | -0.4% | -6.5% | 23.6% | 17.4% | 56.8% | -61.9% |
Thermo Fisher Scientific Inc. | 3.9% | NA | 10.7% | 8.3% | 59.1% | -15.3% |
Micron Technology Inc. | 6% | 4.1% | 5.9% | 9.6% | 99.1% | 3.5% |
Average of above | -7.3% | -0.1% | -7.6% | 5.3% | 60.5% | 0.7% |
Source: Author’s calculation using company 10-K report data. | ||||||
Table 2B shows each company’s ranking for each of the statistics in Table 2A. Although there are exceptions, there is a strong correlation in the rankings of these measures. Of particular note is that companies, as in the cases of Amazon and Adobe, that have done particularly well after passage of the TCJA have either (1) started with large U.S. shares of worldwide income, or (2) shifted a large amount of their profit out of the United States after passage of the TCJA.
Drilling Down
Now let’s follow up a bit on these ideas. First, we will try to estimate the amount of change in an adjusted effective tax rate attributable to the reduction in the U.S. corporate tax rate. We will call this the “rate cut effect.” Second, we will try to estimate the amount of change in the adjusted effective tax rate attributable to post-TCJA profit shifting (either into or out of the United States) after passage of the TCJA. We will call this the “profit-shift effect.” After we’re done, the math will reflect what is self-evident (once you think about it): (1) Companies with a larger share of U.S. income benefit more from the reduction in the U.S. corporate tax rate, and (2) an increase in income shifting into (out of) the United States will push the effective tax rate up (down).
Company | Change in Adjusted ETR | Change in Cash ETR | Change in Reported ETR | Change in Effect of Foreign Rate on ETR | Level of Foreign Percentage of Before-Tax Income | Change in Foreign Percentage of Before-Tax Income |
|---|---|---|---|---|---|---|
Amazon.com Inc. | 1 | 4 | 2 | 5 | 1 | 3 |
Adobe Inc. | 2 | 2 | 5 | 4 | 14 | 1 |
Texas Instruments | 3 | 1 | 6 | 7 | 2 | 13 |
Facebook Inc. | 4 | 17 | 10 | 6 | 10 | 5 |
HP Inc. | 5 | 7 | 1 | 2 | 19 | 4 |
Apple Inc. | 6 | 10 | 7 | 1 | 13 | 10 |
IBM Corp. | 7 | 14 | 15 | 14 | 16 | 2 |
Nvidia Corp. | 8 | 12 | 4 | 9 | 8 | 15 |
Intel Corp. | 9 | 6 | 8 | 10 | 4 | 11 |
Oracle Corp. | 10 | 15 | 9 | 8 | 15 | 9 |
Netflix Inc. | 11 | 11 | 3 | 3 | 3 | 14 |
Alphabet Inc. | 12 | 16 | 11 | 11 | 5 | 8 |
Cognizant Technology Solutions Corp. | 13 | 5 | 12 | 12 | 12 | 12 |
Applied Materials Inc. | 14 | 8 | 18 | 16 | 17 | 6 |
Cisco Systems Inc. | 15 | 9 | 14 | 18 | 9 | 18 |
Microsoft Corp. | 16 | 18 | 13 | 15 | 11 | 17 |
Qualcomm Inc. | 17 | 3 | 19 | 19 | 6 | 19 |
Thermo Fisher Scientific Inc. | 18 | 19 | 17 | 13 | 7 | 16 |
Micron Technology Inc. | 19 | 13 | 16 | 17 | 18 | 7 |
Source: Data from Table 2A. | ||||||
Before using real data, the mechanics of the calculations can be illustrated with a simple example shown in Table 3. Suppose XYZ Corp. has 60 percent of its $100 of worldwide profits outside the United States before passage of the TCJA and pays no U.S. tax on those profits (because those profits aren’t distributed from a controlled foreign corporation and aren’t subject to tax under subpart F). And suppose that after passage of the TCJA, those profits continue to be free of U.S. tax under the new territorial system. (GILTI will be discussed later.)
The change in the dollar amount of U.S. tax resulting from the rate change is -14 percent times $40, that is, -$5.60. The effective tax rate declines by 5.6 percentage points. As shown in Table 3, this could happen, for example, when the foreign tax rate is 10 percent and the U.S.-foreign rate differential declines from 25 percent (equal to 35 minus 10 percent) to 11 percent (equal to 21 minus 10 percent) and that decline in the difference, that is, 14 percent is multiplied by 0.6. Note that the effect of the change in U.S. rates on these statistics (measuring change) is independent of the foreign rate.
What about the profit-shift effect? Suppose XYZ Corp. after passage of the TCJA reduces its foreign profits from $60 to $50. The benefit of the rate cut effect calculated in the previous paragraph will be offset by the increase in foreign profits. The profit-shifting effect does depend on the foreign tax rate; in Table 3, we assume a 10 percent rate. The profit-shifting effect equals the domestic-foreign rate difference of -11 percent multiplied by the change in the foreign share of -10 percent. That leaves us with an effective tax rate increase from profit shifting into the United States of 1.1 percent.
Making the grand assumption (on top of so many other implicit simplifying assumptions so far) that there are no other changes in law or company behavior, the change in the effective tax rate equals -4.5 percent, the sum of the rate cut effect (-5.6 percent) and the profit-shift effect (1.1 percent).
| Pre-TCJA | Rate Cut Effect | Profit-Shift Effect |
|---|---|---|---|
U.S. income | $40 | $40 | $50 |
Foreign income | $60 | $60 | $50 |
Worldwide income | $100 | $100 | $100 |
Foreign share of income | 60% | 60% | 50% |
U.S. tax rate | 35% | 21% | 21% |
Foreign tax rate | 10% | 10% | 10% |
U.S. tax | $140 | $8.40 | $10.50 |
Foreign tax | $60 | $60 | $50 |
Total tax | $200 | $14.40 | $15.50 |
Effective tax rate | 200% | 14.4% | 15.5% |
Change from initial ETR | 0% | -5.6% | -4.5% |
(1) Rate cut effect: | |||
U.S. rate change | -14% |
|
|
Pre-TCJA U.S. share of profit | 40% |
|
|
Rate cut effect | -5.6% |
|
|
(2) Profit-shift effect: | |||
Foreign minus U.S. rate | -11% |
|
|
Change in foreign share of profits | -10% |
|
|
Profit-shift effect | 1.1% |
|
|
Effects (1) and (2) combined | -4.5% |
|
|
Source: Author’s calculation. | |||
Real-World Example
Now let’s do for IBM Corp. what we just did for XYZ Corp. (Calculations used to generate all of IBM’s statistics presented in tables 1, 2, and 4 are shown in the table at the end of this article.) IBM has 33.8 percent of its worldwide profit inside the United States before passage of the TCJA. Because of the corporate rate reduction of 14 percentage points, the effective tax rate declines by 4.7 percentage points (equal to 33.8 percent multiplied by 14 percent).
To calculate the profit-shift effect, we need to know the foreign rate. There seems to be no good way of estimating a foreign profit rate from most companies’ 10-K reports, so in this article we calculate the profit-shift effect assuming both a 5 and 15 percent foreign corporate tax rate. For a 5 percent corporate rate, the foreign-domestic rate differential is -16 percent. IBM’s change in its foreign share (from Table 2) is 32.4 percent. The profit-shift effect is the product of these two figures, or -5.2 percent.
The estimated change in IBM’s effective tax rate is -9.66 percent, equal to the sum of the rate cut effect (-4.7 percent) and the profit-shift effect (-5.2 percent). That is 0.3 percent lower than the actual reduction in the adjusted effective tax rate. This difference can be thought of as the net sum total of all the other changes in law (aside from the corporate rate cut) and company circumstances (aside from the shifting profit).
All Together Now
Table 4 reports the result for all our tech companies when we perform the same calculations as were performed for IBM above. And the profit-shift effects are calculated assuming both a 5 and 15 percent foreign rate.
| (1) Actual Change in Adjusted ETR | (2) Estimated Rate Cut Effect | (3) Estimated Profit-Shift Effect (5%) | (4) Estimated Profit-Shift Effect (15%) | (5) = (2) + (3) Estimates Combined (5%) | (6) = (2) + (4) Estimates Combined (15%) |
|---|---|---|---|---|---|---|
Amazon.com Inc. | -17.5% | -17.3% | -4.2% | -1.6% | -21.5% | -18.9% |
Adobe Inc. | -16.3% | -8.1% | -7.3% | -2.9% | -15.4% | -11% |
Texas Instruments | -14.6% | -11% | 1.1% | 0.4% | -9.9% | -10.6% |
Facebook Inc. | -12.8% | -6.9% | -3.3% | -1.3% | -10.2% | -8.1% |
HP Inc. | -12.6% | -1.2% | -4% | -1.7% | -5.2% | -2.9% |
Apple Inc. | -9.9% | -4.1% | 0% | 0% | -4.1% | -4.1% |
IBM Corp. | -9.7% | -4.7% | -5.2% | -1.9% | -9.9% | -6.7% |
Nvidia Corp. | -9.6% | -3.9% | 1.3% | 0.5% | -2.5% | -3.4% |
Intel Corp. | -8.9% | -8.1% | 0.3% | 0.1% | -7.8% | -8% |
Oracle Corp. | -8.2% | -4.6% | -0.2% | -0.1% | -4.8% | -4.7% |
Netflix Inc. | -6.1% | -9.9% | 1.2% | 0.4% | -8.7% | -9.4% |
Alphabet Inc. | -5.6% | -6.5% | -0.5% | -0.2% | -7% | -6.7% |
Cognizant Technology Solutions Corp. | -5.2% | -4.6% | 0.4% | 0.1% | -4.3% | -4.5% |
Applied Materials Inc. | -4.3% | -2.9% | -2.3% | -1% | -5.2% | -3.9% |
Cisco Systems Inc. | -3.9% | -2.9% | 5.1% | 1.9% | 2.2% | -1% |
Microsoft Corp. | -3.6% | -3% | 3.2% | 1.2% | 0.1% | -1.8% |
Qualcomm Inc. | -0.4% | -7.7% | 6.2% | 2.3% | -1.5% | -5.4% |
Thermo Fisher Scientific Inc. | 3.9% | -4.8% | 2.4% | 0.9% | -2.3% | -3.8% |
Micron Technology Inc. | 6% | -10.1% | -0.6% | -0.2% | -10.7% | -10.4% |
Source: Author’s calculation using data from company 10-K reports. | ||||||
To be sure, few of the estimates of adjusted effective tax rates in Table 4 come as close to the actual estimated effective tax rates as occurred in our IMB calculation that assumed a 5 percent foreign tax rate. But for all the omitted intricacies of the tax law changes, and all the assumed constancy of company circumstances and foreign government policy, our crude estimates have fared much better than most (including your author) could have expected.
Figure 2 provides an illustration of the estimated and actual values of the change in adjusted effective tax rates. Except for the discrepancy between the estimated and assumed change in effective tax rates for Micron Technology, the estimates approximate actual change reasonably well.
In Figure 2, when estimates of adjusted effective tax rates are above actual effective tax rates, that suggests other factors than the rate cut and profit shifting provided an additional benefit — that is, they lowered the effective tax rate by more than the expected amount.
There is some evidence that FDII tax benefits and GILTI tax burdens that are reported in the tax rate reconciliation statements can explain a significant part of the difference between actual and estimated changes in effective tax rates. Effective rate changes due to FDII (for the minority of companies reporting them) — for Microsoft (-1.1 percent in the latest year), for Intel ( -3.2 percent), for Cisco (-2.6 percent), and for Texas Instruments (4.9 percent) — coincide with these companies' larger-than-estimated rate cuts shown in Figure 2. The reported drag of GILTI (6.3 percent) on the effective tax rate of Thermo Fisher Scientific is consistent with that company's smaller-than-expected rate reduction in Figure 2.
Trying to grasp a company’s tax situation by examining its financial accounting as reported in 10-Ks is like doing a curbside appraisal of real estate. Even accounting professors liken it to looking through a glass darkly. Summary statistics like those presented here have cavalierly ignored the complexity of the tax situations and tax planning of a multinational corporation. So it is a pleasant surprise to discover how much we can learn even with these limitations and without delving into the dark complexity of the international TCJA rules.
FYs Ending Dec. 31 | 2019 | 2018 | 2017 | 2016 | 2015 |
IBM Corp. | |||||
Total before-tax income (BTI) | $10,166 | $11,342 | $11,400 | $12,330 | $15,945 |
BTI from U.S. operations | -$315 | $627 | $560 | $3,650 | $5,915 |
BTI from non-U.S. operations | $10,481 | $10,715 | $10,840 | $8,680 | $10,030 |
Cash payments for income tax | $2,091 | $1,745 | $1,597 | $1,078 | $2,657 |
Income tax provision | $731 | $2,619 | $5,642 | $449 | $2,581 |
Statutory rate | 21 | 21 | 35 | 35 | 35 |
Enactment of U.S. tax reform* | 1 | 18 | 48 | — | — |
Tax differential on foreign income | -11 | -9 | -26 | -21 | -17 |
Intra-entity transfers* | 0 | 0 | -5 | — | — |
Domestic incentives | -2 | -3 | -2 | -1 | -2 |
State and local | -1 | -1 | 1 | 1 | 1 |
Japan resolution* |
|
|
| -10 | — |
Other | -1 | -3 | -2 | 0 | -1 |
Effective rate | 7 | 23 | 49 | 4 | 16 |
Adjusted ETR (excludes items with *) | 6 | 5 | 6 | 14 | 16 |
U.S. statutory rate | 21% | 21% | 35% | 35% | 35% |
Effective tax rate | 7% | 23% | 49% | 4% | 16% |
Cash ETR | 20.6% | 15.4% | 14% | 8.7% | 16.7% |
Adjusted ETR | 6% | 5% | 6% | 14% | 16% |
Foreign share of profit | 103.1% | 94.5% | 95.1% | 70.4% | 62.9% |
Foreign effect on ETR | -11% | -9% | -26% | -21% | -17% |
Calculate weights (for Table 1): | 2019 | 2018 | 2017 | 2016 | 2015 |
BTI | $10,166 | $11,342 | $11,400 | $12,330 | $15,945 |
Five-year sum | $61,183 | $61,183 | $61,183 | $61,183 | $61,183 |
Weights | 16.6% | 18.5% | 18.6% | 20.2% | 26.1% |
Weighted average — five years (Table 1): |
|
| 2015-2019 |
|
|
Reported ETR |
|
| 19.5% |
|
|
Cash ETR |
|
| 15% |
|
|
Adjusted ETR |
|
| 10% |
|
|
Foreign share of profit |
|
| 82.9% |
|
|
Foreign effect on ETR |
|
| -17% |
|
|
Calculate weights (Table 2): | 2019 | 2018 |
| 2016 | 2015 |
BTI | $10,166 | $11,342 |
| $12,330 | $15,945 |
Two-year sums | $21,508 | $21,508 |
| $28,275 | $28,275 |
Weights | 47.3% | 52.7% |
| 43.6% | 56.4% |
Weighted averages (Table 2): | 2018-2019 |
| 2015-2016 |
| Difference |
U.S. statutory rate | 21% |
| 35% |
| -14% |
Reported ETR | 15.4% |
| 10.8% |
| 4.7% |
Cash ETR | 17.8% |
| 13.2% |
| 4.6% |
Adjusted ETR | 5.5% |
| 15.1% |
| -9.7% |
Foreign share of profit | 98.5% |
| 66.2% |
| 32.4% |
Foreign effect on ETR | -9.9% |
| -18.7% |
| 8.8% |
Table 3 | 1st Effect |
| 2nd Effect |
|
|
New U.S. rate | 21% |
| Foreign rate |
| 5% |
Old U.S. rate | 35% |
| New U.S. rate |
| 21% |
Rate change | -14% |
| Difference |
| -16% |
times |
|
| times |
|
|
Original U.S. share | 33.8% |
| Change in foreign share |
| 32.4% |
equals |
|
| equals |
|
|
Estimated rate change (1) | -4.7% |
| Estimated rate change (2) |
| -5.2% |
Actual change in adjusted ETR |
|
| -9.66% |
|
|
Estimated change in ETR (effects combined, (1) + (2)) |
|
| -9.92% |
|
|
Difference attributable to other factors besides rate change and profit shifts |
|
| -0.3% |
|
|
Shaded cells are original data from 10-K reports. Dollar amounts in millions. Source: Author’s calculations using data from company 10-K reports. | |||||