David R. Evanson
Privately Published, Summer, 2006
An Indecent Proposal
Deck: Proposed cost sharing regulations are likely to result in less revenue, not more, and undermine the competitiveness of American MNCs in the process.
Suddenly, it seems like legacy is a bad word. Though the term conjures up a connection with the past, and all of the positive connotations associated with longevity, continuity and tradition, in modern business, that’s not always a good thing anymore.
For instance, the malaise which confronts auto manufacturers as well as parts suppliers is often attributed to legacy issues with their healthcare and pension plans. And in IT, the omnipresent challenge, sometimes the raison d’ etre for many Chief Technology Officers, is helping the business remain competitive by overcoming the many difficulties posed by its legacy systems. Securities regulation too, has legacy issues as watchdogs grapple with bending depression era laws around an industry in the full bloom of innovation.
Therefore it should come as little surprise, that legacy issues would ultimately exert their influence on tax policy. For surely the architects of American tax policy did their work in a world that is very different than today. Many tax laws and regulations were crafted when international trade was little more than an afterthought. In fact, many of the tax laws that businesses are regulated by today pre-date the notion of a global enterprise.
And in this context, it’s not surprising that Cost Sharing Arrangements (CSAs) between U.S. companies and their foreign counterparts have become a source of concern. Of course, it didn’t used to be this way. After all, an arm’s length transaction which occurred between controlled entities within the same tax jurisdiction – the United States – though not without twists and turns, nonetheless did not result in the wholesale loss of revenue.
But cost sharing arrangements between a domestic parent and a foreign subsidiary, which results in income, hence tax revenue being channeled offshore, presumably at a lower tax rate, is another matter altogether. And though not a revenue issue per se, a cost sharing arrangement between a domestic parent and a foreign subsidiary which results in the migration of intellectual property into foreign markets that could perhaps one day be used against us competitively or militarily, provokes resistance too.
Were it that the tax revenue implications of CSAs were incidental, a mere by-product, perhaps concerns regarding compliance may not have reached current levels. But from the perspective of the Service, it’s the degree of intention [italicize intention] that appears to have stirred real concern. Speaking before the Senate Committee on Finance during June, 2006 IRS Commissioner Mark Everson said, ” . . . large businesses increasingly engage in sophisticated transactions for both non-tax purposes and tax purposes, resulting in complex relationships with multiple filing requirements. Tax administration continues to be challenged by the increasing number of high value, sometimes cross-border, mergers, acquisitions and other multifaceted international and domestic tiered transactions . . .”
Thus, like other industries, we must now confront the legacy of our tax policy in a world gone global. And admittedly, revisions to the existing regulations governing Cost Sharing Arrangements, may be necessary. However, the current proposals embodied in Treasury Regulation §1.482-7, particularly as it relates to the application of the Investor Model for an arm’s length transaction, falls short of what should or could rightfully be called reform. This is because compensation for transactions which utilize the investor model must be stipulated ex ante [italicize ex ante], i.e. before all the facts are known, at a level that is consistent with what an investor would pay and what a participant would be willing to allow an investor to earn.
There are some very cogent macroeconomic arguments to bolster contention that the use of the investor model is an inappropriate tool to bring about greater compliance: the migration of employment, the migration of a national research and development enterprise and reduced competitiveness of U.S. businesses in increasingly global capital markets, are just some of the results which could flow from changes, as proposed. How the application of the investor model to CSAs could result in this phenomena will be discussed later on.
However, at a broad brush level, there appears one other argument worth taking into account. And this argument rests with the sheer precedent setting notion which lies beneath the application of the investor model in cost sharing arrangements between a domestic parent company and its foreign subsidiaries. Specifically, never has the return which an investor could earn on a risky investment been specified by a tax or regulatory authority.
To see the importance of this, one need look no further than the growth in the GDP since 1929. By almost any yardstick, the American economy has enjoyed one of recorded history’s most spectacular bull runs. In this expansion, the role of tax policy has been largely incidental. That is, at times tax policy has promoted growth (mortgage interest deduction), while at other intervals, tax policy has stifled growth (luxury excise tax). Without question however, one of the primary determinants of this economic success has been investment. While regulation has been decried – rightly or wrongly, for better or for worse – investors have continued to commit capital to the U.S. economy because they have enjoyed unfettered access to whatever returns the organization of their resources would allow for.
Source: Bureau of Economic Analysis
And because of this precedent, we should be wont to consider policies that disrupt the vital linkage between the commitment of capital and the returns that can be earned from this commitment. After all, there is ample evidence that economies are unsustainable when the outcomes in commercial transactions are centrally planned.
Home Court Advantage
The application of the investor model to CSAs provokes one of the central issues of tax policy: is it an instrument of social policy, or simply an instrument to generate revenue? Social policy and tax policy are often uneasy bedfellows. Often times it’s difficult to measure success, or success when achieved can be controversial.
Yet quite the opposite is true regarding a frame of reference focused solely on raising revenue. That is, the central merit of any particular tax policy or regulation is whether or not it increases or decreases tax revenues, as designed. With respect to the application of the investor model to CSAs, the basic premise of increasing tax revenue must be challenged. Specifically, will more tax revenue be generated if the R&D functions of U.S. corporations migrate overseas?
The notion of this question presumes economically rational behavior on the part of American corporations if they are forced to apply the investor model in their CSAs. Specifically, by stipulating the return foreign subsidiaries can earn, and as a consequence shifting income from intellectual property developed overseas in CSAs back to the United States, U.S. MNCs are strongly encouraged to conduct their research and development in other countries. By doing so, MNCs realize the opportunity to enjoy the benefits of lower tax rates in markets where such rates exist, and pay marginal rates in the U.S. of approximately 35%, only on income generated within our borders.
This contention is not simply armchair speculation. Rigorous academic research has confirmed the positive correlation between tax changes and the level of research and development. In their landmark 2000 study, Do R&D Tax Credits Work, British economists Bloom, Griffith and Van Reenen study tax changes and R&D investments over a 19 year period in Australia, Canada, France, Germany, Italy, Japan, Spain, United Kingdom and the United States. The results were unequivocal: ” . . . fiscal provisions matter. The econometric analysis suggests that tax changes significantly effect the level of R&D . . .” The economists estimated that a 10% fall in the cost of R&D stimulates a short increase of 1% in total R&D and a 10% increase long term.
But it’s not all academic. In the real world the stakes could not be higher. While a full scale econometric analysis is beyond the scope of this article, a “scenario check” using available facts is not, and what it shows is illuminating. Take for instance the pharmaceutical industry, which relies heavily on cost sharing arrangements with foreign subsidiaries to develop and distribute products for discrete markets. According to the Pharmaceutical Research and Manufacturers of America (PHRMA), the net R&D investment of its members during 2004 was $36.6 billion, which was part of an industry wide investment of $47.6 billion. Of the PHRMA member’s commitment to R&D, 79% or $29.2 billion was spent here in the United States.
So what could happen if $5 billion of this investment, or about 17% migrates overseas? Applying economic multipliers to this figure does not provide a precise answer, but nonetheless offers insight into the magnitude of the loss. Specifically, the Bureau of Economic Analysis’s healthcare services multipliers for output, earnings and employment are 2.33, 0.8969 and 26.6310 respectively. This means that the reduction of research and development expenditures by $5 billion will reduce overall output by $11.65 billion (2.33 x $5 billion), reduce overall household incomes by $4.48 billion (0.8969 x $5 billion) and reduce employment by 133,155 jobs (26.6310 x [$5 billion/$1 million]).
It’s important to note that output and income figures would be annual [italicize annual] losses. Moreover, since PHRMA member R&D expenditures have grown at a compound annual growth rate of 8.65% since 2000, there is not just the present losses on tax revenues, but also the opportunity costs associated with losses on future revenues. It’s also important to note that pharmaceuticals are just one industry. Energy, electronics, information technology and aerospace industries, to name a few, rely heavily on innovation and have, in aggregate, equally massive research and development budgets.
How likely is this scenario? Given economically rational behavior on the part of American MNCs, it’s likely. For surely if are forced to pay U.S. marginal tax rates even on income earned in foreign markets, with some of these markets offering low corporate tax rates, it’s reasonable to speculate that the American portion of these research and development budgets will atrophy.
As the foregoing scenario check suggests, the tax revenue and economic output losses may be quite substantial under an investor model for CSAs. However, it may also be that tax revenues will not be as large as anticipated either. In the words of commissioner Everson, “LSMB [Large and Mid-Size Business Division] taxpayers are sophisticated, well-capitalized, well-organized, and adept at planning . . . . [with] the resources and willingness to aggressively defend and contest tax positions . . .”
Were it perhaps just a question of revenue, the new regulations, as proposed might be worth the gamble. That is, perhaps MNCs would relent in the drive to optimize their tax footprint, substantial new revenue would materialize, and the loss of R&D employment with associated benefits would be de minimus. Under this scenario, concern about the proposed regulations is little more than distaste for change.
But the case is not that simple. Specious revenue gains must also be considered in light of competitive issues. There are a number of emerging economies with well educated workforces that have shown their mettle for conducting highly technical research and development. These countries include: India, China, Indonesia, Brazil, Russia and the Croatia, to name a few. As the chart below indicates, the wage disparities between the United states and these countries, as measured by per capita gross domestic product (GDP) is stark.
Chart: 2005 Per Capita GDP – United States Vs. Selected Emerging Economies
United States $41,800
Brazil $ 8,400
China $ 6,800
Philippines $ 5,100
Indonesia $ 3,600
India $ 3,300
Source: Central Intelligence Agency World Fact Book
Thus the question we must ask is do we want to provide MNCs tax incentives on top of already compelling wage differentials to move some or all of their research and development offshore? The per capita GDP figures are jarring, but in real life it means that an engineer in India might command a salary of $19,000 a year, versus $25,000 in eastern Europe, versus $75,000, or more [italicize more] here in the United States. Support personal that would cost $3.00 an hour in the Philippines costs more than $12 per hour here.
And it’s not like the United States hasn’t been here before. That is, we witnessed a dramatic decline in manufacturing brought about almost exclusively by wage differentials alone [italicize alone] Therefore, how quickly might the U.S. research and development enterprise wither under the twin burdens of highly skilled, inexpensive foreign workers and [italicize and] tax incentives to conduct research outside of the United States?
Source: Bureau of Economic Analysis
But the concept of competitiveness is multifaceted. That is, it occurs within industries, and between workforces and countries. But it also occurs in the capital markets. Corporations are competing for the capital of investors. Those which are capable of generating the highest return on invested capital are the ones most likely to be attractive to investors.
Once again, it’s important to understand the implications of applying the investor model and the economic consequences which flow from it. Specifically, the stipulation that foreign subsidiaries are relegated to earning a routine return on their investments in CSAs has the effect of shifting income that results from the exploitation of intellectual property developed through cost sharing arrangements back to the United States. (Parenthetically, it’s worth noting the absurdity of such a position by analogy. Specifically, were the investor model applied to the aerospace industry, then Boeing would still be making payments to the estate of the Wright Brothers for Boeing use of basic aviation technology.)
This shifting of income back to the United States has the potential to put American MNCs at a disadvantage to their competitors around the globe because the United States maintains one of the highest corporate income tax rates in the world. That is, ceteris paribus, the application of the investor model means that an American corporation will likely deliver a lower after tax return on its R&D investments than almost every one of its foreign competitors will.
Sources: Congressional Budget Office, Corporate Income Tax Rates: International Comparisons; PricewaterhouseCoopers.
With respect to competitiveness in the capital markets, it’s important to understand what this really means in the context of truly global markets. It’s not as if the emerging companies inside of emerging economies cannot be accessed by institutional investors any longer. Today, investors in Dubai can access companies in Dubuque as easily as those in Durban. Thus, because capital can flow freely, capital will flow freely. To date, there’s been no evidence to the contrary that, as in what was once the microcosm of the U.S. capital markets, companies which delivered the highest return on invested capital enjoyed the lowest weighted average cost of capital, while those less efficient endured a higher cost of capital.
An Indecent Proposal
To all of the foregoing a contrarian might reasonably ask, “So what?” The U.S. economy has been through several incarnations, and will continue to evolve. After all what was once an agrarian economy now derives just 1% of its Gross Domestic Product from agriculture. And in the latter half of the 20th century, the United States witnessed the decline of its manufacturing sector as services rose to the fore. And still, despite these changes, gut wrenching as they may have been for large segments of the workforce, the U.S. economy remains the envy of the world.
And while this is true, there is a fundamental difference in shifts that cause changes in the composition of the workforce versus those which cause a shift in where knowledge is developed and stored. The American economy was able to weather, and in fact prosper from its transition into a service economy precisely because it was able to develop and maintain technological leadership in the industries which constituted its foundations.
Yet the adoption of an investor model in an increasingly global economy, provides, as noted above, strong incentives for the very pearl of American technological leadership – our basic research and development enterprise – to migrate into foreign lands. And what then? We become an economy and a nation based on stewardship. There are any number of nations rich beyond all measure in natural resources that must rely exclusively on foreign technology to develop and exploit these resources. The experience of these nations provides ample evidence that an economy based on stewardship is not in the interest of its businesses or its citizens. And for this reason, the proposed cost sharing regulations, while well intentioned, represent an untenable and ultimately indecent proposal.