FLORIDA TAX REVIEW FLORIDA TAX REVIEW Volume 14 2013 Number 8 319 STARTUP LTD.: TAX PLANNING AND INITIAL INCORPORATION LOCATION by Susan C. Morse* I. INTRODUCTION ............................................................................. 320 II. WHY STARTUPS’ PLACE OF INCORPORATION MATTERS FOR TAX ........................................................................................ 322 A. Being a US-Parented MNC ............................................... 322 B. Being a Non-US-Parented MNC ....................................... 326 III. US INCORPORATION: A DOMINANT STRUCTURE WITH EXCEPTIONS .................................................................................. 329 A. US-Based Startups Generally Incorporate in the US ....... 329 B. Exceptions ......................................................................... 334 IV. EXPLAINING THE DOMINANT STRUCTURE ....................................... 337 A. Limited Tax Benefits of a Non-US-Parent Structure ........ 337 B. Legal Benefits of a US-Parent Structure........................... 339 C. Liquidity and Other Resource Constraints ....................... 342 D. Considering Future Law Changes .................................... 349 V. THEORIZING THE EXCEPTIONS: WHAT FACILITATES A NON-US PARENT STRUCTURE? ................................................... 354 A. Tax Factors ....................................................................... 354 B. Nontax Legal Factors ....................................................... 356 * Assistant Professor, University of Texas School of Law. This project was largely completed while the author served as Associate Professor of Law at UC Hastings College of the Law, and the UC Hastings Summer Research Stipend Program provided funding. Many thanks for helpful comments to participants at the University of Florida Levin College of Law Eighth Annual International Tax Law Symposium and at the 2013 Critical Tax Conference; to Eric Allen, William Dodge, Victor Fleischer, and Stephen Shay; and to participants in workshops and colloquia at UC Hastings, Santa Clara University School of Law, Sturm College of Law at the University of Denver, Boston College School of Law, and the School of Taxation and Business Law at the University of New South Wales. 320 Florida Tax Review [Vol. 14:8 C. Liquidity and Other Resource Constraints ....................... 357 D. Investor Preferences ......................................................... 358 VI. CONCLUSION ................................................................................. 359 I. INTRODUCTION Do startup firms consider taxes when they decide where to organize? This Article analyzes the incorporation decisions of relatively new, US-based private business enterprises with global ambitions. Such startup firms generally organize as US corporations. This Article theorizes this dominant structure and its exceptions, drawing from prior literature and illustrating with informal interview results. It identifies explanatory factors including limited tax benefits of non-US incorporation, legal benefits of US incorporation, startups’ liquidity and other resource constraints, and investor preferences. Part II explains the different treatment of US- and non-US-parented multinational corporations (MNCs) under US federal income tax law. Some non-US-parented MNC structures have tax advantages compared to US- parented MNC structures. This prompts the hypothesis that US-based startup corporations may incorporate outside the United States,1 for example in a tax haven jurisdiction, if the non-tax costs of a non-US incorporation decision are sufficiently low.2 Part III reports that available empirical work shows few examples of startup corporations headquartered in the United States and incorporated in tax haven jurisdictions.3 Prior work4 and informal interviews reported here 1. See, e.g., Mihir A. Desai & Dhammika Dharmapala, Do Strong Fences Make Strong Neighbors?, 63 NAT’L TAX J. 723, 724 (2010) [hereinafter Desai & Dharmapala, Strong Fences] (noting incentive for new firms with global business plans to use a non-US parent corporation). 2. See Daniel Shaviro, The David R. Tillinghast Lecture The Rising Tax- Electivity of U.S. Corporate Residence, 64 TAX L. REV. 377, 383–84 (2011) [hereinafter Shaviro, Rising Tax-Electivity of U.S. Corporate Residence] (noting that the degree of electivity turns on “nontax consequences”). 3. See, e.g., Eric J. Allen & Susan C. Morse, Tax-Haven Incorporation for U.S.-Headquartered Firms: No Exodus Yet, 66 NAT’L TAX J. 395, 395 (2013) [hereinafter Allen & Morse, Tax-Haven Incorporation for U.S.-Headquartered Firms] (reporting twenty-seven instances of tax haven incorporation among 918 identified US-headquartered multinational IPO firms). 4. This analysis builds on work that has considered the puzzle of the typical startup’s decision to organize as a corporation and not as an entity taxed as a partnership. See, e.g., Joseph Bankman, The Structure of Silicon Valley Start-Ups, 41 UCLA L. REV. 1737 (1994) [hereinafter Bankman, The Structure of Silicon Valley 2013] Tax Planning and Initial Incorporation Location 321 support the conclusion that venture-backed Silicon Valley startups generally organize in the United States, form corporations rather than partnerships or LLCs, and prefer the state of Delaware as an incorporation jurisdiction. Available examples, drawn in large part from initial public offering (IPO) evidence, corroborate the conclusion that Delaware incorporation is the norm. The evidence also reveals exceptions to the general rule of US incorporation concentrated in particular industries, such as insurance and marine transportation. Part III also identifies several additional examples of US-based, non-US-incorporated firms, scattered among different industries. Part IV explores several possible reasons for the dominant structure of US incorporation for startup firms. One reason is the limited relative tax advantages of non-US incorporation. For example, US-parented MNCs can often obtain low tax rates on non-US income, sometimes access offshore cash without onerous US tax results, and erode their US taxable income base to some extent. Another reason is that a US-parented structure can provide corporate governance and other non-tax legal advantages. In addition, because US incorporation is a familiar “cookie cutter” structure for startups, it requires fewer monetary and other startup company resources. This is important to startups with liquidity constraints. Investor preference, or at least investor familiarity with the Delaware corporate form, also plays a role. Market participants share the perception that venture capital (VC) firms, an important source of financing for startup companies, favor investments in US corporations, although venture capitalists describe portfolios that include investments that take a range of organizational forms. Even if current US tax laws do not provide a significant incentive for a typical startup corporation to incorporate outside the United States, Congress could change these laws in the future so as to greatly increase the tax burden placed on US-parented multinationals. Yet the possibility of future change in US tax law does not appear to affect most startups’ incorporation location decisions. Startups have reason to discount the possibility of future changes in US tax law adverse to US-parented MNCs because of the strength and diversity of corporate lobbying efforts. Startups and their advisors may conversely worry more about the possibility of Start-Ups]; Victor Fleischer, The Rational Exuberance of Structuring Venture Capital Start-ups, 57 TAX L. REV. 137 (2003) [hereinafter Fleischer, The Rational Exuberance of Structuring Venture Capital]; Daniel S. Goldberg, Choice of Entity for a Venture Capital Start-up: The Myth of Incorporation, 55 TAX LAW. 923 (2002) [hereinafter Goldberg, Choice of Entity for a Venture Capital Start-up]; Calvin H. Johnson, Why Do Venture Capital Funds Burn Research and Development Deductions, 29 VA. TAX REV. 29 (2009) [hereinafter Johnson, Why Do Venture Capital Funds Burn Research and Development Deductions]; Eric J. Allen & Sharat Raghavan, The Impact of Non-Tax Costs on Tax-Efficiency of Venture Capital Investments (April 16, 2013) [hereinafter Allen & Raghavan, The Impact of Non-Tax Costs on Tax-Efficiency] (available at ssrn.com) (manuscript on file with the author). 322 Florida Tax Review [Vol. 14:8 changes adverse to non-US-parented MNCs because they perceive a higher likelihood of such changes inside or outside the United States. They also likely face a higher degree of uncertainty about the possibility of adverse changes applicable to non-US-parented firms. Part V theorizes the noted exceptions to the dominant trend of US incorporation for US-based startups. It suggests that tax factors may drive some non-US incorporation decisions. For example, a favorable set of tax rules supports a Bermuda incorporation decision for certain insurance companies that cover US risks. In addition, nontax legal factors play a role in encouraging some firms to incorporate outside the United States. A number of tax and nontax legal factors facilitate marine transportation firms’ non-US incorporation decisions, for example. Online gambling provides another example of an industry in which regulatory factors have supported a non-US incorporation decision. Additional anecdotal examples of US-based, tax-haven-incorporated firms also suggest that resource constraints and investor preferences influence non-US incorporation decisions. With respect to resource constraints, several US-based, non-US incorporated firms received funding from deep-pocketed sources other than venture capital funds. In each of these cases, one or more wealthy individuals, a large corporation or a private equity firm made an up-front investment in a custom, tax-haven-parented structure. With respect to investor preferences, founder or investor links with an incorporation jurisdiction might lead to a preference for non-US incorporation. An investor preference for non-US incorporation might proceed from current tax and non-tax legal considerations specific to the investor, or simply from greater familiarity with non-US law on the part of the investor or the investor’s advisors. Separately, the investor might believe that non-US incorporation would further the goal of positioning the startup firm as a more attractive acquisition candidate. II. WHY STARTUPS’ PLACE OF INCORPORATION MATTERS FOR TAX A. Being a US-Parented MNC US law uses a brittle place-of-incorporation rule to determine corporate tax residence.5 A firm incorporated in Delaware that lacks any corporate functions in the United States must still pay US corporate income tax. A firm incorporated in Bermuda and headquartered in and managed 5. See I.R.C. § 7701(a)(4). 2013] Tax Planning and Initial Incorporation Location 323 from the United States does not automatically pay US corporate income tax by reason of corporate residence.6 A typical US-parented MNC structure features a US corporate parent that owns non-US operating subsidiaries through intermediate low-tax holding corporations.7 Because the US rules treat separately incorporated affiliates as separate taxpayers,8 non-US corporate subsidiaries of a US parent are not automatically required to pay US federal income tax on all of their income. Under applicable anti-deferral and other rules, a US-parented MNC must currently pay tax on the income of its foreign subsidiaries to the extent such income falls into the definition of “subpart F income”9 or is distributed as dividend income, in each case subject to the reduction of US tax under applicable foreign tax credit provisions.10 US-parented MNCs use transfer pricing, deduction allocation, income source and subpart F planning to allocate profits to low-tax subsidiaries and use foreign tax credit planning, treaty planning, base erosion strategies, and borrowing to minimize non-US taxes and the residual US federal income tax due upon repatriation.11 As a result, the non-US income 6. See Shaviro, Rising Tax-Electivity of U.S. Corporate Residence, supra note 2, at 377–78 (2011) (noting the difference in US tax results that turns on place of incorporation). 7. Edward D. Kleinbard, Stateless Income, 11 FLA. TAX REV. 699, 706–09 (2011) [hereinafter Kleinbard, Stateless Income] (giving Google structure as an example). 8. See Commissioner v. Bollinger, 485 U.S. 340 (1988); Moline Properties Inc. v. Commissioner, 319 U.S. 436 (1943). 9. I.R.C. § 951(a). Subpart F is intended to impose current tax on mobile and passive income. See, e.g., Lawrence Lokken, Whatever Happened to Subpart F? U.S. CFC Legislation After the Check-the-Box Regulations, 7 FLA. TAX REV. 185, 192 (2005) (noting that subpart F does not target active business income); Stephen E. Shay, Exploring Alternatives to Subpart F, 82 TAXES 29, 29–30 (2004) (referring to subpart F’s targeting of passive and base company income). 10. See I.R.C. § 901 (granting foreign tax credit); § 902 (providing for deemed paid foreign tax credit when foreign corporations distribute dividends to certain US corporate shareholders); § 904 (providing foreign tax credit limitation rules). 11. See J. Clifton Fleming, Jr., Robert J. Peroni & Stephen E. Shay, Worse Than Exemption, 59 EMORY L. J. 79, 85 (2009) (noting “overly generous” tax benefits enjoyed by US corporations); Kleinbard, Stateless Income, supra note 7, at 715–26 (arguing that the US international corporate tax system is an “ersatz territorial” system); Lawrence Lokken, Territorial Taxation: Why Some U.S. Multinationals May Be Less Than Enthusiastic About the Idea (and Some Ideas They Really Dislike), 59 SMU L. REV. 751, 759 (2006) (explaining that “refined” tax planning permits many US-parented MNCs to pay less under current rules than they would under a territorial system). With respect to the important technology industry strategy of offshoring intangibles at advantageous valuations, see Yariv Brauner, 324 Florida Tax Review [Vol. 14:8 earned by US-parented MNCs often enjoys both a low foreign effective tax rate and a low US effective tax rate.12 For example, it is estimated that the US income tax burden on non-US business income earned by non-US subsidiaries in MNC groups is perhaps between 3 and 6 percent.13 US-parented MNCs may also may seek ways to allocate income related to US operations to low-taxed non-US affiliates, and conversely to allocate deductions away from low-taxed non-US affiliates and to US operations.14 The ability of firms to erode the US tax base through such planning has led some to conclude that the current US system raises less tax revenue than would a “territorial” system that refrained from taxing non-US Value in the Eye of the Beholder: The Valuation of Intangibles for Transfer Pricing Purposes, 28 VA. TAX REV. 79, 85–95 (2008). 12. See Harry Grubert, Foreign Taxes and the Growing Share of U.S. Multinational Company Income Abroad: Profits, Not Sales, Are Being Globalized, 65 NAT’L TAX J. 247, 281 (2012) (reporting a decline in the average effective foreign tax rate for non-US subsidiaries of US parent corporations from about 21 percent in 1996 to about 16 percent in 2004). See also ABA Tax’n Sec. Task Force, Report of the Task Force on International Tax Reform, 59 TAX LAW. 649, 655–56 (2006) (“The effective rate of US and foreign income taxation of foreign income is understood to be materially lower than the effective rate on domestic income.”). 13. See Melissa Costa & Jennifer Gravelle, Taxing Multinational Corporations: Average Tax Rates, 65 TAX L. REV. 391, 403–04 (2012) (reporting based on 2007 Treasury tax return data an annual US tax on adjusted foreign-source book income of about $18 billion, representing 3.3 percent of such income); see also HARRY GRUBERT & JOHN MUTTI, TAXING INTERNATIONAL BUSINESS INCOME: DIVIDEND EXEMPTION VERSUS THE CURRENT SYSTEM 31-32 (2001) (reporting a 3.3 percent estimate for the overall burden of a dividend repatriation tax assuming an excess limitation foreign tax credit position and a non-US effective tax rate below 10 percent); Rosanne Altshuler & Harry Grubert, Where Will They Go if We Go Territorial? Dividend Exemption and the Location Decisions of U.S. Multinational Corporations, 54 NAT’L TAX J. 787, 797 (2001) (calculating a rate of 5.4 percent based on certain assumptions including a 1.7 percent “excess burden” measure and the US taxation of royalties from intangible assets). 14. See, e.g., Kimberly A. Clausing, Multinational Firm Tax Avoidance and Tax Policy, 62 NAT’L TAX J. 703, 711, 717 (2009) (estimating “financial” income- shifting and “real” productive asset location-shifting responses to higher US tax rates and concluding that the financial effects, producing lost tax revenue of about $87 billion in 2002, were more than double the real effects). The OECD base erosion and profit shifting, or BEPS, project lists as one of its headline goals “limit[ing] base erosion via interest deductions and other financial payments.” ORGANISATION FOR ECONOMIC COOPERATION AND DEVELOPMENT, ACTION PLAN ON BASE EROSION AND PROFIT SHIFTING 17 (2013). 2013] Tax Planning and Initial Incorporation Location 325 business income,15 although the revenue estimate depends on the details of such a territorial system.16 Once a firm has chosen a US-parented structure, it is difficult to change to a non-US parented structure. Applicable rules require gain recognition (but prevent loss recognition) upon such an inversion.17 Moreover, under an anti-inversion statute enacted in 2004,18 a MNC is still treated as a US-parented firm even after acquisition by a foreign corporation if (i) at least 80 percent of the foreign corporation’s stock is owned by former owners of the US parent (by reason of their former ownership of the US parent) and (ii) the firm lacks substantial business activities in the country in which the new foreign parent is incorporated.19 The anti-inversion statute and related regulations leave room for US corporations to invert in connection with an acquisition transaction, but severely limit options for stand-alone inversion transactions.20 Policymakers may take an interest in 15. See supra note 11. 16. See, e.g., Harry Grubert, Enacting Dividend Exemption and Tax Revenue, 54 NAT’L TAX J. 811, 814 (2001) (providing a static revenue gain estimate of $9 billion based on 1996 Treasury data and evaluating possible behavioral responses to territoriality adoption including “adjustments to overhead expenses and royalty payments”). 17. See I.R.C. § 367; U.S. TREASURY OFFICE OF TAX POLICY, CORPORATE INVERSION TRANSACTIONS: TAX POLICY IMPLICATIONS (2002). 18. See I.R.C. § 7874. 19. See Jefferson P. VanderWolk, Inversions Under Section 7874 of the Internal Revenue Code: Flawed Legislation, Flawed Guidance, 30 NW. J. INT’L L. & BUS. 699, 699 (2010). Section 7874 also imposes other inversion restrictions, such as a gain recognition requirement triggered by a lower ownership overlap threshold. 20. After the initial passage of section 7874, some firms took advantage of a facts-and-circumstances “substantial business activities” test and expatriated to create structures with parents in Ireland, Switzerland, and the UK. See Stuart Webber, Inverted U.S. Firms Relocate Headquarters to Europe, 64 TAX NOTES INT’L 589, 591 (Nov. 21, 2011) [hereinafter Webber, Inverted U.S. Firms Relocate] (providing summary chart); Bret Wells, Cant and the Inconvenient Truth About Corporate Inversions, 136 TAX NOTES 429, 430–36 (July 23, 2012) [hereinafter Wells, Cant and the Inconvenient Truth About Corporate Inversions] (describing different forms of acquisition and stand-alone inversion transactions). Recent temporary and proposed Treasury regulations substantially curtail stand-alone transactions by imposing a 25 percent threshold for sales, property, and employees with respect to the substantial business activities requirement. See Temp. Reg. § 1.7874-3T; Kevin M. Cunningham, The New Section 7874 Substantial Business Activity Exception Regulations: Closing the Door, 67 TAX NOTES INT’L 961, 962 (Sept. 3, 2012) (explaining the change from a facts-and-circumstances rule); Eric Solomon, Corporate Inversions: A Symptom of Larger Tax System Problems, 136 TAX NOTES 1449, 1454–55 (Sept. 17, 2012) [hereinafter Solomon, Corporate Inversions: A Symptom of Larger Tax System Problems] (stating that inversion can 326 Florida Tax Review [Vol. 14:8 whether startups’ habit of incorporating in the United States will change under reform alternatives that increase the tax burden placed on US-parented multinational corporations (MNCs). In particular, they may worry that such reforms might drive US-based startups to incorporate outside the United States instead, taking economic activity with them.21 B. Being a Non-US-Parented MNC An alternative structure features an MNC headquartered in the United States, but whose parent is incorporated in a non-US country. The non-US parent might be incorporated in a tax haven that imposes a very low, often zero, rate of corporate income tax,22 and treats foreign subsidiaries and income favorably, for example through a territorial system that only taxes domestic income exempts dividends distributed from foreign subsidiaries to domestic parents.23 The non-US parent would typically own a US subsidiary that houses the US management and US business operations, as well as other subsidiaries incorporated in non-US jurisdictions.24 Tax treaty planning constitutes an important element of a non-US- parented, US-headquartered MNC structure. A treaty relationship between the non-US parent and the US subsidiary would substantially reduce the continue via merger and acquisition transactions after the revised § 7874 regulations). 21. See, e.g. SENATE FINANCE COMMITTEE STAFF, TAX REFORM OPTIONS FOR DISCUSSION: INTERNATIONAL COMPETITIVENESS 4 (May 9, 2013) (expressing concern about US tax incentives for multinationals “to be foreign-based”); see also Desai & Dharmapala, Strong Fences, supra note 1, at 724 (“In a global setting in which formal corporate residence is increasingly elective, new firms that anticipate generating significant amounts of non-US income will have an incentive to incorporate their parent firm outside the United States.”); Roger H. Gordon, Discussion, in FUNDAMENTAL TAX REFORM: ISSUES, CHOICES, AND IMPLICATIONS 365-67 (John W. Diamond & George R. Zodrow eds., 2008) (expressing the concern that worldwide consolidation would encourage MNCs owned by US investors to incorporate outside the United States and would encourage US investors to invest in non-US corporations instead of US corporations). 22. See, e.g., Mihir A. Desai, The Decentering of the Global Firm, 32 WORLD ECON. 1271, 1276–77 (2009) (giving example of an MNC parented by a tax haven-incorporated firm and with “decentered” financial, legal and managerial homes). 23. See, e.g., Webber, Inverted U.S. Firms Relocate, supra note 20, at 591 (providing recent examples of expatriation to European parent structures); Bret Wells, Cant and the Inconvenient Truth About Corporate Inversions, supra note 20, at 430–36 (same). 24. OFFICE OF TAX POLICY, DEPARTMENT OF THE TREASURY, CORPORATE INVERSION TRANSACTIONS: TAX POLICY IMPLICATIONS 12-13 (2002) (explaining post-inversion tax treatment). 2013] Tax Planning and Initial Incorporation Location 327 parent’s chance of exposure to US tax because a treaty relationship increases the threshold for business taxation from the US statutory “effectively connected income” standard to the treaty-based “permanent establishment” standard.25 A treaty relationship also reduces withholding taxes that would apply, for example with respect to interest, dividend, and royalty payments from the US subsidiary to the parent.26 Because direct tax treaties between tax havens and the United States generally do not exist, any treaty planning for a US-parented structure with a tax-haven-incorporated parent would typically rely upon intermediate affiliates with excellent treaty networks.27 25. See, e.g., NEW YORK STATE BAR ASSOCIATION TAX SECTION, REPORT ON OUTBOUND INVERSION TRANSACTIONS 42-43 (2002), http://www.nysba.org/ Content/ContentFolders20/TaxLawSection/TaxReports/1014report.pdf [hereinafter NYSBA, OUTBOUND INVERSION TRANSACTIONS REPORT] (outlining analysis in support of Bermuda parent treatment as engaged in a US trade or business and/or having a US permanent establishment). See generally Lawrence Lokken, Income Connected with U.S. Trade or Business: A Survey and Appraisal, 86 TAXES, Mar. 2008, at 61. In one case, the Tax Court held that a foreign parent provided brokerage services through its US subsidiary and therefore was engaged in a US trade or business. See InverWorld, Inc., et al v. Commissioner, 71 T.C.M. (CCH) 3231 (1996), T.C.M. (RIA) ¶ 1996-301, at 2104–12, reconsideration denied, 73 T.C.M. (CCH) 2777, T.C.M. (RIA) ¶ 1997-226. 26. See, e.g., OFFICE OF TAX POLICY, DEPARTMENT OF THE TREASURY, CORPORATE INVERSION TRANSACTIONS: TAX POLICY IMPLICATIONS 24 (2002) (noting that a 30 percent withholding tax applies to “related party” interest payments from US subsidiary to non-US parent in the absence of a treaty relationship). 27. Tax-haven parents with indirect US subsidiaries have used Barbados and Luxembourg as intermediate treaty countries, for example. See NYSBA, OUTBOUND INVERSION TRANSACTIONS REPORT, supra note 25, at 6, 19-20 (reporting intermediate Barbados affiliate in 1994 Bermuda-parented Helen of Troy structure and intermediate Luxembourg affiliate in 2001 Bermuda-parented Accenture structure). Public disclosures typically do not make such intermediate structures clear. For example, Michael Kors did not disclose an effectively connected income, branch profits, or permanent establishment risk. Its list of subsidiaries includes a Swiss “Holdings” affiliate that might serve an intermediate treaty planning role between the British Virgin Islands parent and the US indirect subsidiary. See Michael Kors Holdings Ltd., Registration Statement (Form F-1) Ex. 21.1 (Dec. 2, 2011), http://www.sec.gov/Archives/edgar/data/1530721/000119312511328487/ d232021dex211.htm (listing subsidiaries). Freescale Semiconductor also does not disclose such risks, although it does disclose that a shareholder’s tax results might be different if that shareholder received dividends or gains related to a US trade or business of the shareholder. See Freescale Semiconductor Holdings I, Ltd., Registration Statement (Form S-1A) 180 (May 20, 2011), http://www.sec.gov/ Archives/edgar/data/1392522/000119312511146916/ds1a.htm. Freescale, like Helen of Troy and Accenture, has a Bermuda parent. However, its list of subsidiaries does not include an intermediate Barbados or Luxembourg holding company or another company whose name suggests that it is used as a treaty intermediate between 328 Florida Tax Review [Vol. 14:8 Because the US rule for corporate tax residence turns on incorporation location, not on management and control, a non-US-parented MNC avoids exposure to US federal income tax on subpart F income earned by non-US subsidiaries, so long as 10 percent-or-more-shareholders that are US persons do not own more than 50 percent of the non-US parent.28 In addition, a non-US-parented MNC can use base erosion or earnings-stripping strategies, under which a US subsidiary makes deductible interest or other payments to its non-US parent, to reduce the amount of income subject to US tax.29 Research on non-US-parented structures resulting from inversion transactions undertaken prior to the enactment of the 2004 anti-inversion rules suggests that the tax savings provided by such structures depends in significant part on the success of earnings-stripping strategies. One study shows no systematic increase in company valuation following the announcement of an inversion.30 But another concludes that markets exhibit more positive reactions to inversions in the presence of greater leverage.31 This is consistent with evidence that earnings stripping results in lower post- Bermuda and the US. A Swiss company appears to serve as an intermediate for indirect European subsidiaries and a British Virgin Islands company as an intermediate for several Asian subsidiaries. See Freescale Semiconductor Holdings I, Ltd., Registration Statement (Form S-1A) Exhibit 21.1 (May 20, 2011), http://www.sec.gov/Archives/edgar/data/1392522/000119312511032381/dex211.ht m (listing subsidiaries). InterWAVE provides a counterexample of an IPO firm that discloses an effectively connected income risk. See interWAVE Communications Int’l Ltd., Registration Statement (Form F-1A) 81 (Jan. 28, 2000), http://www.sec.gov/Archives/edgar/data/1095478/000091205700002759/000091205 7-00-002759.txt. 28. See Mihir A. Desai & James R. Hines Jr., Expectations and Expatriations: Tracing the Causes and Consequences of Corporate Inversions, 55 NAT’L TAX J. 409, 421 (2002) [hereinafter Desai & Hines, Expectations and Expatriations] (citing avoidance of subpart F as the frequently articulated reason for corporate inversions). 29. See, e.g., OFFICE OF TAX POLICY, DEPARTMENT OF THE TREASURY, CORPORATE INVERSION TRANSACTIONS: TAX POLICY IMPLICATIONS 13 (2002) (discussing deductible payments resulting from intercompany debt and reinsurance). 30. See C. Bryan Cloyd, Lillian F. Mills & Connie D. Weaver, Firm Valuation Effects of the Expatriation of U.S. Corporations to Tax-Haven Countries, 25 J. AM. TAX’N ASS’N 87 (2002). 31. See Desai & Hines, Expectations and Expatriations, supra note 28, at 435 (reporting correlation between higher leverage and favorable stock price reactions upon inversion transaction announcement). 2013] Tax Planning and Initial Incorporation Location 329 inversion effective tax rates for inverted firms compared to a control sample.32 III. US INCORPORATION: A DOMINANT STRUCTURE WITH EXCEPTIONS A. US-Based Startups Generally Incorporate in the US A “startup,” for purposes of this Article, is a relatively new private business enterprise with global ambitions. This Article focuses on initial incorporation decisions of “relatively new” firms and not on later changes in a firm’s place of incorporation, for example as a result of a standalone inversion transaction, strategic acquisition, or private equity or other financial acquisition. A “US-based” firm has a preponderance of ties to the United States. In the IPO study cited in this Article, my co-author Eric Allen and I used a definition of more than 50 percent US revenue, employees, or identified real property to identify US-headquartered firms.33 Some of the material presented in this Article, including the informal interviews, draws heavily from venture capital experience, although the definition of “startup” used here is not limited to venture-backed firms. Other sources of initial outside capital could include individuals, corporations, governments, or investment groups other than venture capital firms. The availability of data and prior work constitutes one reason for the focus on venture capital experience. In addition, many venture-backed portfolio companies happen to fall within the startup definition adopted for purposes of this Article, since such companies often have or expect to develop global components in their respective business plans. Prior work that investigates the choice of organizational form by US- based startups backed by US venture firms observes that US incorporation, particularly Delaware incorporation, is the market norm.34 Venture capital 32. See Jim A. Seida & William F. Wempe, Effective Tax Rate Changes and Earnings Stripping Following Corporate Inversions, 57 NAT’L TAX J. 805, 825 (2004) (finding an 11.6 percentage point post-inversion tax rate reduction). 33. See Allen & Morse, Tax-Haven Incorporation for U.S.-Headquartered Firms, supra note 3, at 403. This Article does not consider inversion transactions undertaken by mature firms, including transactions prior to or after the 2004 passage of anti-inversion legislation and transactions in connection with a cross-border acquisition. Compare, e.g., Wells, Cant and the Inconvenient Truth About Corporate Inversions, supra note 20. 34. See, e.g., Bankman, The Structure of Silicon Valley Start-Ups, supra note 4, at 1739–40 (“In almost all cases, the [portfolio firm] will be structured as a corporation.”); Johnson, Why Do Venture Capital Funds Burn Research and Development Deductions, supra note 4, at 50–51 (noting that corporate investors, which could use losses if a startup firm had a pass-through structure, generally do 330 Florida Tax Review [Vol. 14:8 investments generally use standard contract structures, including a C corporation organizational form and a capital structure featuring preferred and common stock.35 Lawyers maintain these contract structures and influence them, just as lawyers have been shown to significantly affect a new corporation’s choice of domicile36 and client firms’ choice of takeover defenses or other structures over time.37 Several interviews reported here support and add color to the observation that that US incorporation is the norm for US-based startups.38 One law firm partner said that “9.5 out of 10” startups used a Delaware corporation;39 the others used limited liability company (LLC) or California corporation structures. Another said that “most VCs don’t want to invest except into a Delaware corporation.”40 Entrepreneurs agreed. One characterized the VC expectation of Delaware corporate organization as “cookie-cutter.”41 Another entrepreneur reported that incorporation was an not invest in venture capital funds); Allen & Raghavan, The Impact of Non-Tax Costs on Tax-Efficiency, supra note 4, at 39 (finding that only seventeen out of 995 firms in a sample of venture-backed IPO firms from 1996-2008 had LLC status at the time of IPO). 35. Convertible preferred stock provides an attractive tradeoff between venture capitalists’ goal of claiming a control stake in a portfolio company and their goal of accessing maximum returns on their investment, for example in the event of an IPO. The use of preferred stock also permits the use of tax-advantaged equity compensation strategies for portfolio company employees. See Ronald J. Gilson, Engineering a Venture Capital Market: Lessons from the American Experience, 55 STAN. L. REV. 1067 (2003) [hereinafter Gilson, Engineering a Venture Capital Market]; Ronald J. Gilson & David M. Schizer, Understanding Venture Capital Structure: A Tax Explanation for Convertible Preferred Stock, 116 HARV. L. REV. 874 (2003) (presenting tax explanation). 36. See, e.g., Robert Daines, The Incorporation Choices of IPO Firms, 77 N.Y.U. L. REV. 1559 (2002) [hereinafter Daines, The Incorporation Choices of IPO Firms]. 37. See John C. Coates IV, Explaining Variation in Takeover Defenses: Blame the Lawyers, 89 CAL. L. REV. 1301 (2001) [hereinafter Coates, Explaining Variation in Takeover Defenses]. See also Victor Fleischer & Nancy C. Staudt, The Supercharged IPO, 67 VAND. L. REV. ____ [hereinafter Fleischer & Staudt, The Supercharged IPO] (forthcoming 2014) (presenting evidence of the impact of professional networks on firms’ use of a specific planning strategy). 38. Consistent with commitments made to interviewees, interviews cited in this Article omit identifying details. No person was interviewed twice, and the citations to different interviews therefore reference conversations with unique individuals. 39. Telephone Interview with San Francisco law firm partner (Jan. 28, 2013). 40. Telephone Interview with San Francisco law firm partner (Feb. 16, 2013). 41. Telephone Interview with San Francisco entrepreneur (Feb. 16, 2013). 2013] Tax Planning and Initial Incorporation Location 331 “absolute requirement” at the time of his company’s first venture funding and that his company reorganized from an LLC to a Delaware corporation at that time.42 And an entrepreneur who had organized his firm as an LLC and who was fully aware of the tax ramifications of different organizational decisions reported that one reason he had avoided venture funding was to avoid “dogma” including the heuristic of Delaware incorporation.43 As discussed further below, VCs do invest in portfolio firms not organized as Delaware corporations. Nevertheless, these comments reflect perceptions of an industry norm. The choice of Delaware incorporation involves three subsidiary decisions. First, Delaware incorporation rejects the alternative choice of home-state, for example California, organization. Second, it rejects the option of organizing as an LLC taxed as a flow-through for US tax purposes. Third, and most important for purposes of this Article, it rejects the option of incorporating outside the United States. With respect to the question of Delaware or home-state incorporation, literature in the corporate governance area demonstrates that US-based firms (or their lawyers) frame the incorporation location decision largely as a binary choice between home-state and Delaware incorporation.44 Cited reasons for Delaware incorporation include the quality of courts, the substance of corporate governance law, and the habits of advisors. With respect to the question of organization as an LLC or corporation, an LLC would permit investors to benefit from one layer of tax rather than two in the event of a profitable startup, or to use the passed- through losses more often generated by a startup.45 The losses “burned” as a result of venture capitalists’ choice of a corporate form46 are estimated to be worth “billions”47 and wasting them may seem “hard to reconcile with any 42. Telephone Interview with Silicon Valley entrepreneur (Jan. 30, 2013). A California corporation, like a Delaware corporation, forms the foundation for a future structure as a US-parented MNC. 43. Telephone Interview with Silicon Valley entrepreneur (Feb. 11, 2013). 44. See Lucian Arye Bebchuk & Alma Cohen, Firms’ Decisions Where to Incorporate, 46 J.L. & ECON. 383 (2003); Daines, The Incorporation Choices of IPO Firms, supra note 36; Jens Dammann & Matthias Schundein, The Incorporation Choices of Privately Held Corporations, 27 J.L. ECON. & ORG. 79 (2011). 45. See, e.g., Goldberg, Choice of Entity for a Venture Capital Start-up, supra note 4. 46. Johnson, Why do Venture Capital Funds Burn Research and Development Deductions, supra note 4, at 53. 47. Allen & Raghavan, The Impact of Non-Tax Costs on Tax-Efficiency, supra note 4, at 3 (estimating lower bounds for the value tax benefit value of foregone losses at $1.4 billion – $4.4 billion based on IPO evidence from 1996- 2011). 332 Florida Tax Review [Vol. 14:8 strong form of efficient market hypothesis.”48 Yet contemporaneous work shows that out of 995 venture-backed firms that conducted IPOs on US markets between 1996 and 2008, only forty-eight firms initially organized as LLCs, and only seventeen retained LLC status until the IPO. The few firms that organized as LLCs were more profitable than those that did not.49 Prior literature also considers the possible impact of factors that may favor incorporation despite the tax advantages of LLC organization, including limited ability to use tax losses among existing investors in venture capital firms and reduced transaction costs that result from the relative simplicity of C corporation organization.50 The question of US or non-US incorporation, which is the focus of this Article, adds to the Delaware-versus-home state and LLC-versus- corporation elements of a startup’s organization decision. Available empirical evidence shows that US incorporation is the dominant organizational decision for US-based startup firms with global ambitions. In particular, initial public offering data from 1997 through 2010 reveals that fewer than 3 percent of identified US-based multinational corporations in the data set incorporated in tax havens. In addition, fewer than 2 percent of such firms incorporated in non-tax-haven, non-US jurisdictions.51 Informal interview evidence corroborates the view that US incorporation is the norm for US-based startups with global ambitions. One entrepreneur, for example, characterized US organization as the undisputed default rule, even, for example, for a startup that anticipated that the most significant market for its product would be in Europe.52 One lawyer 48. Bankman, The Structure of Silicon Valley Start-Ups, supra note 4, at 1767 (raising the possibility of a gambler’s mentality). 49. Allen & Raghavan, The Impact of Non-Tax Costs on Tax-Efficiency, supra note 4, at 2. 50. See Bankman, The Structure of Silicon Valley Start-Ups, supra note 4, at 1767–68 (considering reduced transaction costs and venture capitalists’ collective action problem); Fleischer, The Rational Exuberance of Structuring Venture Capital, supra note 4, at 139–40 (adding that venture capitalists generally would not benefit from startup losses and that loss limitations prevent typical investors in venture funds from taking full current advantage of losses produced by pass-through portfolio companies). 51. See Allen & Morse, Tax-Haven Incorporation for U.S.-Headquartered Firms, supra note 3, at 407. The IPO data set excludes firms that stay private and those that only list on non-US exchanges. It includes firms that were not “relatively new” at the time of IPO, for example because a firm may have gone public for a second time after a taking-private transaction conducted by a financial investor. In addition, data as of the IPO date typically lags incorporation. See id. at 401–02. Nevertheless, the study provides some support for the conclusion that Delaware incorporation is the market norm for US-based startup firms. 52. Telephone Interview with Silicon Valley entrepreneur (Jan. 30, 2013). 2013] Tax Planning and Initial Incorporation Location 333 responded to the question of whether the global elements of a client’s plan influenced advice on the place of incorporation for a firm. “No,” he said. “They have no money.”53 Startups that are not US-based, on the other hand, may well incorporate outside the United States, even if they receive funding from US- based venture capital firms. Some firms that originate outside the United States and initially organize as non-US corporations may keep their non-US status after they move to the United States. Israeli-parented firms, for example, comprise about 1 percent of the firms that conducted US-based IPOs between 1997 and 2010.54 One lawyer recalled instances of such firms retaining an Israeli-parented structure and establishing US subsidiaries, even though the lawyer thought such an approach tended to produce inefficiencies since two sets of lawyers were required rather than one.55 Another mentioned Israeli and UK firms as examples of non-US-incorporated structures that tended to survive the migration to the US venture capital market.56 This contrasts with another reported approach, in which a non-US firm that wants to access the US market reorganizes into a US-parented structure.57 Similarly, some venture capital firms maintain non-US offices or assign certain partners to oversee non-US-based portfolio companies. These portfolio companies may be organized under non-US law rather than US law.58 Examples of foreign office locations include Israel, India, and China. 53. Telephone Interview with San Francisco entrepreneur (Feb. 16, 2013). 54. See Allen & Morse, Tax-Haven Incorporation for U.S.-Headquartered Firms, supra note 3, at 407. 55. Telephone Interview with San Francisco law firm partner (Jan. 28, 2013). 56. Telephone Interview with Silicon Valley law firm partner (Feb. 27, 2013). 57. E.g., Telephone Interview with San Francisco law firm partner (Feb. 16, 2013). 58. For example, Sequoia Capital manages portfolios of companies based in India, see Sequoia Capital, http://www.sequoiacap.com/india (last visited June 3, 2013), and China, see Sequoia Capital, http://www.sequoiacap.cn/en/ (last visited June 3, 2013). The website of Bessemer Venture Partners lists dozens of portfolio companies based in India, Israel and Europe, some of which can be identified as non-US companies by their “Ltd.” suffix. See Bessemer Venture Partners, http://www.bvp.com/portfolio (last visited June 3, 2013) (sort by Europe, India and Israel geographies). 334 Florida Tax Review [Vol. 14:8 B. Exceptions The dominant structure of US incorporation for US-based startups is subject to several important exceptions. This small set of examples of US- based startups that incorporate outside the United States, and in particular in tax havens, adds context to the default rule of US incorporation. The below discussion covers exceptions in the insurance, marine transportation, and online gambling industries. It also includes several other examples of firms that do not follow industry lines. Of nearly 3,000 firms in a dataset of IPOs on US markets between 1997 and 2010, only forty-seven were US-based, tax-haven-incorporated firms. Of these, thirteen were insurance carriers, which generally insure US risks.59 The structure of such an insurance company typically features a Bermuda parent and a US subsidiary. The US subsidiary sources and services the insurance policies covering US risks, while premiums paid to the US subsidiary are substantially eroded by means of deductible reinsurance payments made to a Bermuda affiliate.60 As described in Part IV, a collection of rules, primarily tax rules, facilitates insurance firms’ decision to incorporate in Bermuda. Marine transportation is the second industrial category identified as a typical candidate for a US-based, tax-haven-incorporated firm structure in the study of US IPO data. Four firms out of the forty-seven identified in the IPO study were marine transportation firms.61 Each of the four firms identified in the data set engages in commercial shipping, but other 59. See Allen & Morse, Tax-Haven Incorporation for U.S.-Headquartered Firms, supra note 3, at 413–15. Some of the Bermuda-parented insurance firms in the IPO dataset fit the definition of startup used in this Article. As an example, Validus Holdings Re, which went public in July 2007, disclosed that it was “formed in October 2005” with the sponsorship of a set of private equity funds. See Validus Holdings Ltd., Registration Statement (Form S-1A) 2 (July 19, 2007), http://www.sec.gov/Archives/edgar/data/1348259/000095012307010068/e28184a6s v1za.htm. 60. See NYSBA OUTBOUND INVERSION TRANSACTIONS REPORT, supra note 25, at 27–29 (describing Bermuda-parented insurance company structure). 61. See Allen & Morse, Tax-Haven Incorporation for U.S.-Headquartered Firms, supra note 3, at 413–15. Some of the tax-haven-parented marine transportation firms in the IPO dataset fit the definition of startup used in this Article. As an example, General Maritime Corporation, a Marshall Islands corporation which went public in June 2001, disclosed that it was “newly formed” and would assemble vessels and support services assets from three different sources. See General Maritime Corporation, Registration Statement (Form S-1A) 3–4 (June 12, 2001), http://www.sec.gov/Archives/edgar/data/1127269/00009120570151949/ a2051255zs-1a.txt. 2013] Tax Planning and Initial Incorporation Location 335 international marine transportation businesses, such as passenger cruise lines, also use tax haven parents.62 Online gambling provides another example of an industry in which firms that target the US market have used tax-haven-parented structures. For example, one leading company incorporated in Gibraltar conducted an IPO on the London Stock Exchange in 2005 using a proxy statement that reportedly disclosed that 90 percent of its customers were US.63 As further discussed in Part IV, industry-specific reasons encourage the use of tax- haven-parented structures for US-based firms in the insurance, marine transportation and online gambling examples. Other examples of tax-haven-parented, US-based startup firms resist categorization along industry lines. Because of the small sample size and inconsistency of data in publicly available disclosures, these examples also resist statistical analysis. They include firms with less restrictive resource constraints and/or non-US investors that own a significant percentage of stock. Firms with less restrictive resource constraints, including an absence of dependence on deferred law firm fee arrangements or post-incorporation venture capital financing, include several funded by established public corporations such as Motorola,64 Sun65 and Tyco International.66 In one other 62. Carnival, for example, incorporated in Panama in 1972 and operates as a dual listed company with Carnival plc, a UK company organized in 2000. See Carnival Corporation Annual Report (Form 10-K) 3 (Jan. 30, 2011), http://www.sec.gov/Archives/edgar/data/815097/000119312511018320/d10k.htm. Royal Caribbean reports that it “was founded in 1968 as a partnership. Its corporate structure evolved over the years and the current parent corporation, Royal Caribbean Cruises Ltd., was incorporated on July 23, 1985 in the Republic of Liberia . . . .” See Royal Caribbean Cruises Ltd., Annual Report (Form 10-K) 1 (Feb. 25, 2013), http://www.sec.gov/Archives/edgar/data/884887/000104746913001567/a2213132z1 0-k.htm. 63. See Christine Hurt, Regulating Public Morals and Private Markets: Online Securities Trading, Internet Gambling, and the Speculation Paradox, 86 B.U. L. REV. 371, 415 (2006) [hereinafter Hurt, Regulating Public Morals] (reporting PartyGaming IPO). 64. See Iridium World Communications Ltd., Registration Statement, (Form S-1A) 99 (June 10, 1997), http://www.sec.gov/Archives/edgar/data/ 948421/0000950133-97-002150.txt (disclosing approximately 28 percent ownership by Motorola, Inc.). 65. See OpenTV Corp., Registration Statement (Form F-1A) 3 (Nov. 19, 1999), http://www.sec.gov/Archives/edgar/data/1096958/0000950013099006648/00 00950130-99-006648.txt (disclosing that OpenTV began as a joint venture between Thomson Multimedia S.A. and Sun Microsystems). 66. See TyCom Ltd., Registration Statement (Form S-1A) 2, F-6 (July 24, 2000), http://www.sec.gov/Archives/edgar/data/1108511/0000950013000004016/ 0000950130-00-004016.txt (explaining that the registrant formed as a wholly-owned 336 Florida Tax Review [Vol. 14:8 case, a private equity investment fund sponsored a capital-intensive startup.67 A broad group of equity partners together with current profits apparently funded Accenture’s startup phase.68 Wealthy individuals sponsored firms including RSL Communications.69 In addition, some US-based, tax-haven-parented startup firms have major founders or shareholders with non-US connections. Some firms appear to have started doing business outside the United States.70 Others had major non-US shareholders at the time of IPO, including corporate investors71 and subsidiary of Tyco International Ltd., which would continue to own 89 percent of the stock post-IPO). 67. See Aircastle Ltd., Registration Statement (Form S-1A) 50 (Aug. 2, 2006), http://www.sec.gov/Archives/edgar/data/1362988/000095013606006258/ file1.htm (“We were formed in October 2004 with a capital commitment of $400 million from funds managed by Fortress for the purpose of investing in aviation assets.”). 68. See Accenture Ltd., Registration Statement (Form S-1A) 62–63, 73 (July 18, 2001). http://www.sec.gov/Archives/edgar/data/1134538/0000950130 01503127/ds1a.htm (showing that no shareholder owned more than 5 percent of any class of Accenture Ltd. shares, that its partners owned 82 percent of the voting equity, and annualized after-tax profits exceeding $1.5 billion for each year of operation following separation from Arthur Andersen). 69. Ralph Lauder funded RSL Communications. See RSL Communications Ltd, Registration Statement (Form S-1A) 101 (Mar. 20, 1998), http://www.sec. gov/Archives/edgar/data/1036297/0000889812-98-000682.txt (stating that Lauder owned about 43 percent of the registrant’s common stock prior to IPO). See also Kerry A. Dolan, Srpska Calling, FORBES (Nov. 2, 1998, 12:00AM), http://www.forbes.com/forbes/1998/1102/6210141a.html (describing RSL’s “deep pockets”). 70. See Garmin Ltd., Registration Statement (Form S-1A) 36 (Dec. 6, 2000), http://www.sec.gov/Archives/edgar/data/1121788/000095013100006701/000 0950131-00-006701.txt (disclosing that Garmin “formed in Taiwan”); Vistaprint Ltd., Registration Statement (Form S-1A) 35 (Sept. 26, 2005), http://www.sec.gov/Archives/edgar/data/1262976/000119312505190875/ds1a.htm (noting that business initially commenced in France, then moved to the US, where it was conducted by a US corporation, which later amalgamated with a new Bermuda company). 71. See interWAVE Communications International Ltd., Registration Statement (Form F-1A) 75 (Jan. 28, 2000), http://www.sec.gov/Archives/edgar/ data/1095478/000091205700002759/0000912057-00-002759.txt (listing Nortel Networks Corp. as 22 percent shareholder pre-IPO); Iridium World Communications Ltd., (Form S-1A) 99 (June 6, 1997), http://www.sec.gov/Archives/edgar/data/ 948421/0000950133-97-002129.txt (disclosing approximately 11 percent ownership by a Japanese-based consortium and 9 percent ownership by a German-based investor pre-IPO); OpenTV Corp., Registration Statement (Form F-1A) 3 (Nov. 19, 1999), http://www.sec.gov/Archives/edgar/data/1096958000095013099006648/000 2013] Tax Planning and Initial Incorporation Location 337 also including Michael Kors, which received most of its funding prior to its 2011 IPO from two non-US individual investors through their jointly held investment vehicle.72 IV. EXPLAINING THE DOMINANT STRUCTURE A. Limited Tax Benefits of a Non-US-Parent Structure The tax benefits of a non-US-parented structure derive from three possible factors. First, a non-US-parented structure may reduce the current tax paid by a MNC with respect to non-US income. Second, it may reduce the tax paid upon the repatriation of income from subsidiaries to the MNC parent. Third, it may reduce the current tax paid by an MNC with respect to US income. The tax-efficient structures of many US-parented MNCs initially suggest that a startup has little to gain from a non-US-parented structure. The effective rate of US income tax collected on non-US income earned under US-parented structures is already very low.73 Even if a non-US-parented structure brings the rate of current US taxation on non-US income closer to zero, it may not result in significant cost savings. A non-US-parented structure might also permit the tax-efficient repatriation of profits to the parent company, for example for the purpose of dividend distributions to ultimate shareholders. Under the currently dominant US-parented MNC structure, a firm pays US income tax less foreign tax credits, in addition to withholding taxes, when the MNC repatriates profits from a foreign subsidiary to the US parent. Some firms, in particular technology firms, can achieve very low rates of non-US tax rate on non-US income. As a result, they may have less ability to use foreign tax credits to shelter US income tax upon the repatriation of profits.74 0950130-99-006648.txt (disclosing that OpenTV began as a joint venture between Thomson Multimedia S.A. and Sun Microsystems). 72. See Michael Kors Holdings Ltd., Registration Statement (Form F-1) 87– 88 (Dec. 2, 2011), http://www.sec.gov/Archives/edgar/data/1530721/0001193 12511328487/d232021df1.htm (disclosing approximately 52 percent pre-IPO beneficial ownership by Silas K.F. Chou, a Hong Kong individual, and Lawrence S. Stroll, a Canadian individual). 73. See supra note 13 and accompanying text (reviewing data on US taxation of non-US income of US-parented MNCs). 74. See Rosanne Altshuler & Harry Grubert, Repatriation Taxes, Repatriation Strategies and Multinational Financial Policy, 87 J. PUB. ECON. 73, 74–75 (2002) (noting the connection between lower foreign tax rates and higher incentives to avoid repatriation). It is possible for a US parent to benefit from offshore cash without a taxable repatriation in some circumstances, as high-profile cases illustrate. See, e.g., Peter Burrows, Apple Avoids $9.2 Billion in Taxes With 338 Florida Tax Review [Vol. 14:8 Many US-parented firms have substantial profits that remain offshore, in part because of the residual tax expense of repatriation.75 A non- US-parented structure would remove this residual US tax obstacle to repatriation from a US subsidiary to a foreign parent. Such a non-US- parented structure presents other planning challenges, however, including the challenge of how to avoid withholding taxes on such a repatriation payment. Withholding taxes are generally charged at rates as high as 30 percent on related party interest or dividends paid by a US firm and on royalties paid for the use of US intellectual property,76 and treaty planning is presumably necessary to reduce these taxes.77 The possible reduction in tax on US income as a result of a non-US- parented structure also deserves consideration. Evidence from inversion transactions suggests that the main benefit of transforming from a US- parented to a tax-haven-parented corporation lies in the reduction of tax on US income, not in the reduction of tax on non-US income. Inverted companies erode the tax base of their US operating subsidiaries through strategies such as intercompany leverage, which results in deductible payments from US subsidiaries to non-US parents. Empirical evidence indicates that the benefit of an inverted transaction to a firm correlates with the availability of such base erosion strategies, although base erosion can also feature in US-parented structures.78 Importantly, all three of these advantages — reduction of tax on non- US income, reduced tax upon repatriation, and reduction of tax on US income — will benefit a mature and profitable firm more than a loss-making, newly-formed startup corporation. For example, in order to take advantage of a non-US-parented structure to reduce tax on US income, a firm needs taxable income and the capacity to establish intercompany agreements that support deductible payments from the US to the non-US parent. Many startups have neither. Finally, a startup incorporated outside the United States that has US shareholders faces the risk of categorization as a passive foreign investment company, or PFIC. A foreign corporation is a PFIC if passive income makes Debt Deal, Bloomberg, May 3, 2013, http://www.bloomberg.com/news/2013-05- 02/apple-avoids-9-2-billion-in-taxes-with-debt-deal.html. 75. See Susan C. Morse, A Corporate Offshore Profits Transition Tax, 91 N.C. L. REV. 549, 550 (2013) (referencing estimate of $1 trillion to $2 trillion of untaxed offshore earnings). 76. See, e.g., I.R.C. §§ 861 (providing source rules), 881 (taxing non-US corporations on the receipt of certain US-source income). 77. See supra notes 25–27 (discussing tax treaty planning in the context of non-US-parented firms’ access to permanent establishment rules). 78. See supra notes 14–15 and accompanying text and 29–32 and accompanying text (reviewing base erosion strategies of US-parented and inverted firms). 2013] Tax Planning and Initial Incorporation Location 339 up at least 75 percent of its gross income or assets held to produce passive income make up at least 50 percent of its total asset value.79 Concern about a startup’s PFIC status can result from current operating losses together with investment income produced by working capital, including working capital attributable to IPO proceeds. PFIC status imposes unattractive tax results on US shareholders.80 Registration statements for non-US-parented MNCs that go public typically disclose the possibility of PFIC status,81 and PFIC status presents a concern for shareholders of private companies as well, despite a limited start-up company exception.82 B. Legal Benefits of a US-Parent Structure A US-parent structure, meanwhile, provides non-tax legal benefits. US incorporation provides the benefit of access to Delaware corporate governance law, for example, and non-US incorporation does not.83 Startup advisors also perceive an advantage to US incorporation for purposes of the protection of intellectual property and other property rights. The corporate governance advantages of US incorporation include access to relatively investor-friendly and highly reliable Delaware or other state corporate governance law.84 The 2011 registration statement for Michael Kors, incorporated in the British Virgin Islands, lists risk factors related to the difficulty of initiating shareholder derivative actions, enforcing 79. See I.R.C. § 1297. 80. These results include the possible imposition of the maximum ordinary income tax rate on “excess distributions” including gain on sale of stock, together with an interest charge. I.R.C. § 1291. 81. See, e.g., Freescale Semiconductor Holdings I, Ltd., Registration Statement (Form S-1A) 179 (May 20, 2011), http://www.sec.gov/Archives/ edgar/data/1392522/000119312511146916/ds1a.htm; Michael Kors Holdings Ltd., Registration Statement (Form F-1) 114 (Dec. 2, 2011), http://www.sec.gov/ Archives/edgar/data/1530721/000119312511328487/d232021df1.htm. See also Aircastle Ltd., Registration Statement (Form S-1A) 36 (Aug. 2, 2006), http://www.sec.gov/Archives/edgar/data/1362988/000095013606006258/file1.htm (disclosing that the firm expected to be categorized as a PFIC and as a CFC). 82. See I.R.C. § 1298(b)(2). 83. See Mitchell A. Kane & Edward B. Rock, Corporate Taxation and International Charter Competition, 106 MICH. L. REV. 1229, 1239–40 (2008) (contrasting “corporate surplus” and “tax surplus”). 84. Id. at 1255–58 (arguing that the first-best solution to the problem of corporate tax consequences influencing corporate governance choices is to segregate them so that tax results follow from a “real seat” rule while corporate governance results follow from a “place of incorporation” rule). 340 Florida Tax Review [Vol. 14:8 judgments against officers and directors, and pursuing minority or other shareholder rights.85 Related research that investigates companies that cross-list on different securities markets indicates that cross-listed firms trade at a premium because their willingness to comply with stricter accounting, disclosure, and other rules serves as a “bonding” signal that encourages investors to invest.86 Structures subject to less regulatory oversight may permit more rent extraction by corporate managers, particularly in widely held corporations;87 or make transparent reporting of earnings more elusive.88 Even if non-US incorporation saves corporate governance costs, for example by reducing compliance costs produced by Sarbanes-Oxley and other US regulatory requirements,89 investors may experience the absence of tighter corporate governance regulation as a net disadvantage. Informal interview results corroborate some of these corporate governance concerns with non-US incorporation. One lawyer explained that using a Delaware corporation ensured that an investor would be protected under US law, for example because board and/or shareholder votes would be required for certain corporate actions and because investors could limit the amount of capital exposed to non-US law, for example through contract limitations on how much of the investors’ cash could be transferred to a foreign subsidiary of a US parent.90 One venture capitalist explained that it was generally not advisable for VCs to invest in a firm formed outside the US, because of the risk that non-US governments might take over a company 85. See Michael Kors Holdings Limited, Registration Statement (Form F-1) 25–26 (Dec. 2, 2011), http://www.sec.gov/Archives/edgar/data/1530721/00011 9312511328487/d232021df1.htm (listing risk factors related to corporate governance). 86. See, e.g., John C. Coffee Jr., Racing Toward the Top?: The Impact of Cross-Listings and Stock Market Competition on International Corporate Governance, 102 COLUM. L. REV. 1757 (2002). 87. See Mihir A. Desai & Dhammika Dharmapala, Earnings Management, Corporate Tax Shelters, and Book-Tax Alignment, 62 NAT’L TAX J. 169 (2009); Mihir A. Desai & Dhammika Dharmapala, Corporate Tax Avoidance and Firm Value, 91 REV. ECON. STAT. 537 (2009) (finding a correlation between institutional ownership and tax avoidance); Michelle Hanlon & Joel Slemrod, What Does Tax Aggressiveness Signal? Evidence from Stock Price Reactions to News About Tax Shelter Involvement, 93 J. PUB. ECON. 126 (2009) (finding smaller stock price declines for firms with good governance). 88. See Victor Fleischer, Regulatory Arbitrage, 89 TEX. L. REV. 227, 263– 64 (2010) (discussing “opacity costs”). 89. See Kate Litvak, The Effect of the Sarbanes-Oxley Act on Non-U.S. Companies Cross-Listed in the U.S., 13 J. CORP. FIN. 195 (2007). 90. Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). 2013] Tax Planning and Initial Incorporation Location 341 or otherwise unilaterally reduce the value of an investment.91 Another venture capitalist whose firm’s portfolio included non-US-parented companies said that such non-US investments were more common in a transaction in which the investment firms owned a controlling interest in the portfolio company, in which case corporate governance concerns might have less importance.92 The importance of corporate governance is also suggested by the exceptions to the rule of US incorporation for US-based startups. One lawyer explained that a startup with an Israeli or UK parent would often keep its offshore parent instead of migrating to a US-parented structure when entering the US market.93 One venture capitalist cited Europe as an exception to the rule of the strong preference for Delaware incorporation because of a higher level of comfort with European law.94 Property protection may provide another reason to choose a US- parented structure for a startup.95 One lawyer said flatly with respect to IP development, “I want them to do it in the US, to be honest.” But this preference had more nuance. The lawyer had discomfort with the idea of a development team or IP ownership in India or China, but accepted the idea that IP might be developed or owned in Switzerland.96 Another lawyer emphasized a remedies advantage of US court jurisdiction over Delaware corporations, which the lawyer said ensured the ability to sue and recover damages if successful in the event of a controversy.97 This focus on local 91. Telephone Interview with Silicon Valley venture fund partner (Feb. 8, 2013). 92. Telephone Interview with Silicon Valley venture/private equity fund partner (Feb. 22, 2013). 93. Telephone Interview with Silicon Valley law firm partner (Feb. 27, 2013). Other lawyers also mentioned an exception for Israeli firms. See, e.g., Telephone Interview with San Francisco law firm partner (Jan. 28, 2013). 94. Telephone Interview with Silicon Valley venture fund partner (Feb. 8, 2013). 95. Formal legal rules generally do not sanction different IP protection by one country if the IP is held by a firm incorporated in a different country. International intellectual property conventions include anti-discrimination requirements. See ROBERT P. MERGES, PETE S. MENELL & MARK A. LEMLEY, INTELLECTUAL PROPERTY IN THE NEW TECHNOLOGICAL AGE 55 (6th ed. 2012). Nevertheless, the prospect of defending intellectual property rights in a jurisdiction other than the jurisdiction of incorporation may increase uncertainty for reasons including concern about de facto differences in the application of the law. See, e.g., Rama Lakshmi, India Rejects Novartis Drug Patent, WASH. POST, Apr. 1, 2013, http://www.washingtonpost.com/world/asia-pacific (reporting on Indian Supreme Court case denying protection for an improved form of a pre-existing compound and suggesting that the case might discourage foreign pharma investment in India). 96. Telephone Interview, San Francisco law firm partner (Feb. 16, 2013). 97. Telephone Interview, Silicon Valley law firm partner (Feb. 4, 2013). 342 Florida Tax Review [Vol. 14:8 presence appears to derive in part from legal realism concerns. An investor with an office in India, for example, may have higher confidence as a practical matter about its ability to get a hearing and a remedy in Indian court.98 C. Liquidity and Other Resource Constraints As prior literature observes, business frictions can interfere with tax- motivated planning.99 Even if a structure with a parent in a low-tax jurisdiction offered tax advantages that clearly outweighed corporate governance and other legal disadvantages, certain business frictions, primarily liquidity and other resource constraints, would likely prompt many startups to continue to choose Delaware incorporation. Liquidity constraints partly explain the premium placed on simplicity in the default startup structure. Declining to search for an optimal organizational structure and instead accepting the “satisfactory solutio[n]” provided by the dominant heuristic of Delaware incorporation also saves non-cash resources.100 Other work has noted the importance of cash conservation and the up-front and ongoing expense of nonstandard structures for startup firms.101 Interview evidence also supports the conclusion that many startups have severe liquidity constraints. One entrepreneur admitted that he and his co- founder were “very cheap” and simply “never brought in outside legal counsel” until financing.102 Another explained that “you want to defer expense as far as possible.”103 One lawyer reported that startups often had 98. Telephone Interview, Silicon Valley law firm partner (Feb. 4, 2013). 99. See generally David M. Schizer, Frictions as a Constraint on Tax Planning, 101 COLUM. L. REV. 1312, 1323–35 (2001) (evaluating impact of behavioral distortions on tax planning choices); David A. Weisbach, Line Drawing, Doctrine, and Efficiency in the Tax Law, 84 CORNELL L. REV. 1627, 1665–68 (1999) (same). 100. Herbert A. Simon, Rational Decision Making in Business Organizations, 69 AM. ECON. REV. 493, 498 (1979). See generally DANIEL KAHNEMAN, THINKING FAST AND SLOW (2011) (distinguishing between automatic and heuristics-based “System 1” decisions and energy-demanding “System 2” decisions). 101. See, e.g., Bankman, The Structure of Silicon Valley Start-Ups, supra note 4, at 1749–50 (noting the often-mentioned reason for incorporation of minimizing legal and organizational costs). Planning related to non-US income may account for a substantial part of the total cost of US-parented firms’ tax compliance. See Marsha Blumenthal & Joel B. Slemrod, The Compliance Cost of Taxing Foreign-Source Income: Its Magnitude, Determinants, and Policy Implications, 2 INT’L TAX & PUB. FIN. 37 (1995). 102. Telephone Interview with Silicon Valley entrepreneur (Jan. 30, 2013). 103. Telephone Interview with Silicon Valley entrepreneur (Feb. 11, 2013). 2013] Tax Planning and Initial Incorporation Location 343 difficulty meeting costs in the thousand-dollar range prior to venture investment. A good startup attorney, he said, “understands the need to conserve cash.”104 One venture capitalist could not think of an example where an early globalization strategy requiring significant up-front capital investment had improved a startup’s valuation. Such plans “take a lot of money,” and when “the music stops,” for example prior to the company achieving self-sufficiency through a revenue stream, the capital is gone.105 Reluctance to slow down business plan implementation may limit startups’ focus on legal issues generally. One entrepreneur reported that “legal is the last thing on my mind;” when a prospective strategic acquirer asked him “how many compliance people” he had, he was struck by the fact that the acquirer had “the luxury of asking that question.”106 Another entrepreneur explained that even though unfavorable customer agreements had required renegotiation at the time of acquisition, costing time and money, the company would not necessarily have invested in legal review of the agreements even with the benefit of hindsight, since such review would have slowed down sales.107 Startup lawyers triage the legal issues they recommend that their clients address. Issues like intellectual property ownership and a clean capitalization table take precedence over organization decisions.108 Startup lawyers readily give examples of startups that have stumbled over IP ownership or “founder in the woodwork”109 problems. In contrast, they generally do not see evidence that organizing as anything other than a Delaware C corporation changes the startup firm’s chances of a successful exit or the likely valuation of that exit.110 Resource constraints encourage startups to prefer the simplest organizational structure: the Delaware C corporation. This organizational choice is simple in part because of path dependence; it is the “cookie-cutter” structure,111 the “pre-approved” package,112 the “gold standard,”113 104. Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). 105. Telephone Interview with Silicon Valley venture fund partner (Feb. 8, 2013). 106. Telephone Interview with San Francisco entrepreneur (Feb. 16, 2013). 107. Telephone Interview with Silicon Valley entrepreneur (Feb. 30, 2013). 108. E.g., Telephone Interview with San Francisco law firm partner (Jan. 28, 2013) (explaining the importance of “buddy” issues relating to co-developed IP and emphasizing the importance of focusing on “the most high level issues”). 109. Telephone Interview with Silicon Valley venture fund partner (Feb. 8, 2013). 110. E.g. Telephone Interview with San Francisco law firm partner (Feb. 16, 2013) (“I don’t think there’s any example of people getting extra money for [an offshore structure].”). 111. Telephone Interview with San Francisco entrepreneur (Feb. 16, 2013). 344 Florida Tax Review [Vol. 14:8 “dogma,”114 the “generally accepted norm.”115 Under current practice, Delaware corporations are simpler than LLCs or partnerships and also simpler than US-based non-US corporations. Important examples of Delaware corporation simplicity identified in the corporation-versus-LLC literature include employee options, corporate governance, and exit strategy,116 although it would surely be possible to optimize and simplify a different structure if it were widely used.117 Similarly, Delaware organization is simpler than non-US organization for US-based startups. The addition of non-US entities to the mix adds more complexity because the relevance of different laws requires more than one set of legal and other advisors.118 Managing multiple corporate entities and related agreements between them complicates the project of running the company.119 Available resources for efficient corporation formation services for startups lean toward a Delaware corporate structure, especially the resources provided by law firms. A founder looking for online advice about how to organize a startup will soon find LegalZoom, which provides form documents for LLCs as well as S and C corporations.120 However, the online 112. Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). 113. Telephone Interview with Silicon Valley venture fund partner (Feb. 22, 2013). 114. Telephone Interview with Silicon Valley entrepreneur (Feb. 11, 2013). 115. Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). 116. See Fleischer, The Rational Exuberance of Structuring Venture Capital, supra note 4, at 167–84 (explaining reasons why venture capitalists may stick with the “devil they know”). 117. See Bankman, The Structure of Silicon Valley Start-Ups, supra note 4, at 1767–68 (considering venture capitalists’ collective action problem). 118. Telephone Interview with San Francisco law firm partner (Jan. 28, 2013). 119. Telephone Interview with San Francisco law firm partner (Feb. 16, 2013) (characterizing the use of an offshore intellectual property holding company subsidiary as “very expensive” tax planning that “completely breaks the idea of simplicity”); Telephone Interview with Silicon Valley entrepreneur (Jan. 30, 2013) (explaining that setting up places of business in different global locations was “much more difficult than we anticipated,” due to issues like local bank account and office requirements). 120. Legal Zoom provides a choice among the US states for jurisdiction of organization but does not mention the possibility of offshore incorporation. See LEGAL ZOOM, http://www.legalzoom.com/ (last visited July 11, 2013). 2013] Tax Planning and Initial Incorporation Location 345 term sheet generators supported by large Silicon Valley law firms assume a C corporation structure.121 Law firms have standardized and made cost-effective the formation of a startup as a Delaware corporation. More than one lawyer put the current cost of the “thirty-plus” documents needed to form a company, ranging from articles of incorporation to employee option plans, at $2000-$3000.122 Uniform questionnaires facilitate the process, as do, at least in some cases, law firm outposts located in low-cost locations.123 In contrast, documents for a firm that organizes as an LLC might cost $10,000 and forming a Bermuda corporation, perhaps $30,000.124 The dominance of the Delaware corporate form permits not only upfront cost savings, but also lower diligence costs in the event of later transactions. One lawyer explained the approach of organizing each startup firm with the same number of shares of authorized common stock and a similarly sized option pool. The uniform approach permitted the easy conversion of financing terms to a preferred stock price, and the expression of most preferred stock pricing as a figure in the range of twenty-five cents to one dollar for each share of preferred stock. Since this approach is similar to other firms’ approach, it makes negotiating more straightforward and diligence less expensive.125 A cost difference in the thousands of dollars may seem an insufficient reason to choose one organizational approach over another for a corporation that might be worth billions of dollars someday. Yet this cost differential is important for a startup firm that makes its organizational decision before it obtains outside financing. For example, it may make the decision in connection with a preliminary round of financing in which it seeks smaller investments from friends and family or from angel investors. Startups may also begin work on their business plan without significant 121. For example, the “Series Seed” documents developed by a Fenwick & West lawyer assume a Delaware corporation. See SERIES SEED FINANCING DOCUMENTS, http://www.seriesseed.com/posts/documents.html (last visited July 11, 2013). See also Wilson Sonsini, Term Sheet Generator, http://www.wsgr.com/ WSGR/Display.aspx?SectionName=practice/termsheet.htm (last visited July 11, 2013) (giving jurisdiction options of Delaware, California or “Other” and assuming as the next step the issuance of preferred stock). 122. Telephone Interview with San Francisco law firm partner (Feb. 16, 2013); see also Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). 123. Telephone Interview with San Francisco law firm partner (Feb. 16, 2013). 124. Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). 125. Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). 346 Florida Tax Review [Vol. 14:8 venture capital financing in particular in areas that require little capital investment, such as the development of mobile or cloud software applications.126 In some cases, law firms’ willingness to defer startups’ obligation to pay legal fees contributes to lawyers’ reluctance to recommend “exotic” structures.127 The fee deferral limit might range from $15,000128 to $25,000129 and accordingly will not cover the expense of offshore incorporation. The limit puts the attorneys at risk for the amount of the deferred fees in the event the startup fails to achieve financing, and lawyers may not consider the investment worthwhile. One lawyer said that he would be happy to implement a complex structure if he were paid “full freight,” but not “on a deferred fee basis.”130 The reluctance to spend resources on a nonstandard corporate structure is consistent with the view that venture capital firms do not place any value on such a structure. Prior research suggests that venture capitalists insist on a Delaware corporate structure for their portfolio companies.131 However, three venture capitalists informally interviewed did not go so far and instead denied that they filtered out corporations that were organized as other than Delaware corporations. The investors generally acknowledged 126. Telephone Interview with Silicon Valley law firm partner (Jan. 28, 2013) (noting the emergence of low-capital business models). 127. Telephone Interview with Silicon Valley law firm partner (Feb. 27, 2013). 128. Telephone Interview with Silicon Valley law firm partner (Feb. 27, 2013). 129. Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). 130. Telephone Interview with San Francisco law firm partner (Feb. 16, 2013). 131. See Bankman, The Structure of Silicon Valley Start-Ups, supra note 4, at 1767–68 (attributing corporate structure decisions to venture capitalists); Gilson, Engineering a Venture Capital Market, supra note 35 (same); Johnson, Why Do Venture Capital Funds Burn Research and Development Deductions, supra note 4, at 89 (same). Researchers have similarly reported that private equity ownership of portfolio firms correlates with tax planning. See, e.g., Brad A. Badertscher, Sharon P. Katz & Sonja O. Rego, The Separation of Ownership and Control and Tax Avoidance, 56 J. ACCT’G & ECON. 228, 242 (2013) (reporting that portfolio companies controlled by private equity firms had higher tax avoidance measures than management-owned companies); Steven N. Kaplan & Per Strömberg, Leveraged Buyouts and Private Equity, 23 J. ECON. PERSP. 121, 134-35 (2009) (noting tax benefits of additional leverage in private equity-owned portfolio companies). 2013] Tax Planning and Initial Incorporation Location 347 their preference for Delaware incorporation, but also expressed the view that “the lawyers” overstated this preference.132 One venture capitalist cited portfolio companies that were incorporated in various US states as well as outside the United States and in LLC form.133 Another said that organizational form was not something the firm sought to “optimize.”134 Firms may rely on the judgment of earlier investors in a portfolio company or on the judgment of co-investors and to accept their planning with respect to organizational form.135 Some venture firms have investment fund segments that specifically target non-US portfolio companies, which are typically organized as non-US firms.136 The norm of US incorporation may derive less from venture capital preference and more from the involvement of lawyers that advise startup firms.137 Lawyers have developed an out-of-the-box structure for startup firms that is easy and cheap to implement because it is so frequently replicated. The lawyer’s interests and the startup’s issues are generally closely aligned with respect to the goal of conserving cash and other resources, particularly if the lawyer has agreed to defer fees. As one lawyer explained, if startups have “no revenue,” and “no product,” they have “no future,”138 and cannot pay their legal bills. Other work demonstrates that lawyers can significantly influence client legal decisions, including state of incorporation,139 use of takeover defenses,140 and aggressive structures to permit founder liquidity.141 132. Telephone Interview with Silicon Valley venture fund partner (Feb. 8, 2013) (stating that the California code was “stable” and “good enough” and that LLC formation “can be fixed later,” when “institutional investors” are involved). 133. Telephone Interview with San Francisco entrepreneur (Feb. 16, 2013). Contemporaneous examination of venture-backed IPO evidence also finds some evidence of a small number of venture-backed LLCs. Allen & Raghavan, The Impact of Non-Tax Costs on Tax-Efficiency, supra note 4. 134. Telephone Interview with Silicon Valley venture fund partner (Mar. 8, 2013). 135. Telephone Interview with San Francisco entrepreneur (Feb. 16, 2013); Telephone Interview with Silicon Valley venture fund partner (Mar. 8, 2013). 136. See supra note 58 (giving examples drawn from Sequoia Capital and Bessemer Partners portfolios). 137. Cf. Reinier H. Kraakman, Gatekeepers: The Anatomy of a Third-Party Enforcement Strategy, 2 J.L. ECON & ORG. 53 (1986). 138. Telephone Interview with Silicon Valley law firm partner (Feb. 16, 2013). 139. See, e.g., Daines, The Incorporation Choices of IPO Firms, supra note 36. 140. See Coates, Explaining Variation in Takeover Defenses, supra note 37. 141. See Fleischer & Staudt, The Supercharged IPO, supra note 37 (presenting evidence of the impact of professional networks on firms’ use of a specific planning strategy). 348 Florida Tax Review [Vol. 14:8 In the corporation-versus-LLC context, the overwhelming adherence of startups to the standard Delaware corporation organization advice offered by lawyers and other gatekeepers has raised the question of whether the default corporation approach is a rational or irrational habit.142 The corporation-versus-LLC question raises this issue because there are known net tax costs presented by the corporate organizational form, whether it is the loss of the benefit of tax losses or the cost of double taxation in the event of profit.143 In the US-versus-non-US incorporation case, however, the rationality question is not broadly presented, because the choice of US incorporation instead of non-US incorporation does not clearly result in net tax costs for a typical startup and its investors.144 Nevertheless, the influence of advisors over startups’ US-versus- non-US organization decision is important. For example, it influences the analysis of what might happen in the event the US did (contrary to predictions discussed below based on legislative process and interest group constraints) increase the tax burden of US-parented MNCs relative to non- US-parented MNCs. In the absence of a mediating group of advisors and in the absence of a standard market norm for a startup’s organizational structure, an increase in the tax burden of US-parented MNCs should marginally increase the frequency of US-based startups that incorporate outside the US based on each individual startup’s cost-benefit analysis. However, if mediated by gatekeepers such as venture capitalists and lawyers, the possible adoption of non-US-parented MNC startup structures in response to a change in law is more complicated. One issue is agency costs. The existing default Delaware incorporation structure optimizes various 142. Compare Bankman, The Structure of Silicon Valley Start-Ups, supra note 4, at 1767 (acknowledging transaction costs and loss limitations but also suggesting various “irrational” possible reasons for corporate startup structure including the possibility that “individual investors are irrationally attracted by the remote possibility of enormous return”) and Johnson, Why Do Venture Capital Funds Burn Research and Development Deductions, supra note 4, at 89 (“The explanations offered on why the funds accept such high taxes do not justify or explain the destruction of the tax benefits. The results cannot be justified by drafting habits in a billion dollar fund because the stakes are too high to be justified by inertia.”) with Fleischer, The Rational Exuberance of Structuring Venture Capital, supra note 4, at 139–140 (placing greater weight on transaction costs and loss limitations including the fact that venture capitalists generally would not benefit from startup losses). 143. Allen & Raghavan, The Impact of Non-Tax Costs on Tax-Efficiency, supra note 4. 144. See, e.g., Shaviro, Rising Tax-Electivity of U.S. Corporate Residence, supra note 2, at 383–84 (2011) (noting that the degree of electivity turns on a comparison between tax advantages and “nontax consequences” such as corporate governance). 2013] Tax Planning and Initial Incorporation Location 349 venture capital goals. Its widespread use may also reflect lawyers’ aversion to taking on the professional risk of recommending an untested structure.145 Another issue is collective action, since developing a new startup structure norm might be worthwhile only if the new structure could be used for a large number of clients.146 Both the agency cost issue and the collective action issue suggest that significant path-dependent obstacles would block the development of a new norm of non-US incorporation of US-based startups in reaction to a tightening of US tax law applicable to US-parented MNCs. This applies in particular to venture-backed startups because the advisor community that serves these firms shows a strong commitment to a US-incorporation norm for US-based firms. Gatekeeper network effects would influence any change to the prevailing norm, including a change from a default of US incorporation to a default of non-US incorporation.147 D. Considering Future Law Changes In the future, Congress could impose onerous tax rules on US- parented MNCs.148 It also could impose onerous non-tax rules on US- 145. See Gilson, Engineering a Venture Capital Market, supra note 35. Cf. Ruth Mason, Delegating Up: State Conformity With the Federal Tax Base, 62 DUKE L.J. 1267, 1323–24 (2013) (referencing other literature on the “stickiness” of contract default terms). 146. See Bankman, The Structure of Silicon Valley Start-Ups, supra note 4, at 1767–68 (considering reduced transaction costs and venture capitalists’ collective action problem). See generally MANCUR OLSON, THE LOGIC OF COLLECTIVE ACTION (1965). 147. See Marcel Kahan & Michael Klausner, Standardization and Innovation in Corporate Contracting (or “The Economics of Boilerplate”), 83 VA. L. REV. 713, 729 (1997) (contending that “learning and network benefits” encourage boilerplate and “path dependence”). Cf. Robert Cooter, Do Good Laws Make Good Citizens? An Economic Analysis of Internalized Norms, 86 VA. L. REV. 1577, 1585– 87 (2000) (noting that social norms present the possibility of multiple equilibria in part because of “fixed costs and network effects”). 148. Courts have held that the Constitution does not prevent Congress from imposing a full current tax on all of the income earned by non-US subsidiaries of US parents. See, e.g., Garlock Inc. v. Commissioner, 489 F.2d 197, 202-03 (2d Cir. 1973) (holding subpart F’s current taxation of US shareholders Constitutional) (citing Heiner v. Mellon, 304 U.S. 271, 281 (1938) (holding Constitutional the taxation of a partner on partnership distributive share regardless of “the fact that it may not be currently distributable as a matter of state law”)), cert. denied, 417 U.S. 911 (1974). See also Cook v. Tait, 265 U.S. 47 (1924) (rejecting the contention that substantive due process concerns blocked the United State’s ability to tax noncitizen individuals); Flint v Stone Tracy Co., 220 U.S. 107, 151-52 (1911) (holding a 350 Florida Tax Review [Vol. 14:8 parented MNCs. But available data does not suggest that startups place importance on the possibility of future changes in US tax law that would be adverse to US-parented MNC structures when they consider how to organize. Several possible reasons may support this lack of concern. Legislative process obstacles and interest group lobbying may block US legal changes adverse to US-parented MNCs. Similar considerations may not produce similarly formidable obstacles to legal change adverse to non-US- parented MNCs, whether under US or non-US law. In addition, the experience of US-based startups and their advisors with US law and legal change may provide a greater degree of certainty about the low likelihood of change in the United States, particularly with respect to change applicable to US-parented structures. US-based advisors have good reasons based on a significant body of data to believe that legal changes adverse to a US-parented MNC structure are unlikely. Legislators may hotly criticize MNC tax planning in public hearings.149 But Congressional policy goals also explicitly include corporate income tax constitutional because of its nature as an excise tax, not a “direct” tax subject to apportionment). Several proposals for international corporate tax reform could, depending on the details, increase the tax burden of US-parented multinationals. One option is worldwide consolidation. See, e.g., Kleinbard, Stateless Income, supra note 7, at 152–55 (listing advantages of worldwide consolidation, including satisfaction of capital export neutrality, solution to the problem of “stateless income,” and finessing of the “otherwise intractable” problems of transfer pricing and expense allocation). Another is the imposition of a minimum tax on non-US subsidiaries’ income. See WHITE HOUSE & DEP’T OF TREASURY, THE PRESIDENT’S FRAMEWORK FOR BUSINESS TAX REFORM 14 (2012). A third option is territoriality, or dividend exemption. See JANE G. GRAVELLE, CONG. RESEARCH. SERV., R42624, MOVING TO A TERRITORIAL INCOME TAX: OPTIONS AND CHALLENGES (2012). This approach would permanently exempt an MNC’s non-US business income from US income tax. Territoriality is generally presented as a business-friendly reform. However, its adoption could produce increased taxes on US-parented multinationals depending on the details, including choices about the disallowance of deductions such as overhead expense allocable to non-tax business income and the inclusion of royalty income paid by non-US subsidiaries to US parents. See CONG BUDGET OFFICE, OPTIONS FOR TAXING U.S. MULTINATIONAL CORPORATIONS 22 (2013) (citing expense allocation and royalty taxation as possible sources of increased revenue). The current taxation of low-taxed foreign income under territoriality or dividend exemption presents another open question. See J. Clifton Fleming, Jr., Robert J. Peroni & Stephen E. Shay, Designing a U.S. Exemption System for Foreign Income When the Treasury is Empty, 13 FLA. TAX REV. 397 (2012). 149. See, e.g., Nelson D. Schwartz & Charles Duhigg, Apple’s Web of Tax Shelters Saved It Billions, Panel Finds, N.Y. TIMES, May 20, 2013, at A1 (reporting on Senate Permanent Subcommittee on Investigations hearing featuring Apple CEO Tim Cook and Senator Carl Levin). 2013] Tax Planning and Initial Incorporation Location 351 “increas[ing] US competitiveness” and “reduc[ing] tax incentives for multinationals to be foreign-based.”150 In addition, the US legislative process presents sequential hurdles to enactment and therefore favors the status quo.151 In the area of corporate tax law reform, agency costs further hamper change. For example, managers face an incentive to favor policies like accelerated depreciation that provide targeted incentives for new corporate investment, even though shareholders prefer policies that also enrich existing investment.152 Moreover, the heterogeneity of interests among different corporations may strengthen the importance of interest group influence in the area of corporate tax policy. This is because there is an incentive for corporations that disproportionately benefit from a certain tax break to lobby energetically to keep that tax break rather than supporting more general reform proposals.153 While overall tax reform adverse to corporate interests is possible, it is unusual.154 And when broad reform does occur, it faces the prospect of later erosion.155 Startup lawyers, venture capitalists and entrepreneurs may be more concerned about future law changes applicable to non-US parented MNCs. Some future law changes that could affect non-US-parented structures would arise under non-US law. In the informal interviews conducted, some concern about the uncertain application of non-US law was expressed. It is possible that startups and their advisors place a higher likelihood on the possibility of such adverse non-US law change. They might do so because they perceive 150. SENATE FINANCE COMMITTEE STAFF, TAX REFORM OPTIONS FOR DISCUSSION: INTERNATIONAL COMPETITIVENESS 4 (May 9, 2013). 151. See, e.g., WILLIAM N. ESKRIDGE, JR., PHILIP P. FRICKEY & ELIZABETH GARRETT, LEGISLATION AND STATUTORY INTERPRETATION 70 (2d ed. 2006) (“The most salient aspect of the modern legislative process is that it is filled with a complex set of hurdles that proponents of a new policy must overcome before their bill becomes law.”). 152. See Jennifer Arlen & Deborah M. Weiss, A Political Theory of Corporate Taxation, 105 YALE L.J. 325, 336–38 (1995) (arguing that managers have incentives to favor policies that encourage additional investment or otherwise make possible increases to individual returns such as salaries). 153. See Michael Doran, Managers, Shareholders and the Corporate Double Tax, 95 VA. L. REV. 517, 536–42 (2009) (citing “unevenness resulting from the different use of corporate tax preferences, interest deductions, and tax shelters”). See also MARTIN SULLIVAN, CORPORATE TAX REFORM: TAXING PROFITS IN THE 21ST CENTURY (2011) (explaining some interest group and other concerns that make reform unlikely). 154. See, e.g., Daniel Shaviro, Beyond Public Choice and Public Interest: A Study of the Legislative Process as Illustrated by Tax Legislation in the 1980s, 139 U. PA. L. REV. 1 (1990). 155. See, e.g., Edward J. McCaffery & Linda R. Cohen, Shakedown at Gucci Gulch: The New Logic of Collective Action, 84 N.C. L. REV. 1159 (2006). 352 Florida Tax Review [Vol. 14:8 relatively lower legislative process constraints outside the United States, where fewer legislative process obstacles to change may exist, for example in parliamentary systems. Other future law changes that could affect non-US-parented MNCs would arise under US law. In the tax area, the anti-inversion rule of section 7874 provides a prominent example of a US statute that targets a non-US- parented structure. Under this statute, a standalone firm with a parent incorporated in the United States generally cannot invert into a non-US- parented structure respected as such for US tax purposes absent substantial business activities in the country where the new parent is incorporated.156 Other proposals could tighten earnings-stripping rules and limit the ability of US subsidiaries of non-US parents to erode US tax bases via deductible payments such as interest and royalties;157 impose a mind-and-management residence rule to replace the current place-of-incorporation rule;158 or impose a higher rate on dividends paid to US shareholders from non-US corporations compared to dividends from US corporations.159 If non-US-parented MNCs are less effective than US-parented MNCs at lobbying Congress, anti- foreign-corporation proposals like these may have a greater chance of enactment relative to proposals that impose more onerous requirements on US-parented firms. 160 Another possibility is that uncertainty aversion causes a higher level of concern about the possibility of change adverse to a non-US-parented 156. See supra notes 18–20 and accompanying text (analyzing I.R.C. § 7874). 157. A series of administration proposals made by both the Bush and Obama administrations would significantly limit the deductibility of interest expense paid by US subsidiaries to foreign parents established in an inversion transaction. See Solomon, Corporate Inversions: A Symptom of Larger Tax System Problems, supra note 20, at 1451–52. 158. One example is legislation that would change the corporate tax residence rule from place-of-incorporation to place of management and control or place of listing. See Omri Y. Marian, Jurisdiction to Tax Corporations, 54 B.C. L. REV. __ (forthcoming 2013), manuscript at 45 (identifying proposals to change US corporate residence rule), 51–53 (advocating US residence status for corporations managed and controlled in the United States or listed on a US exchange). 159. Currently, dividends from “qualified foreign corporations,” meaning dividends on publicly traded stock or dividends from corporations eligible for the benefits of a satisfactory comprehensive income tax treaty, are taxed at the preferential rate under section 1(h) that also applies to dividends from US corporations. I.R.C. § 1(h)(11)(C). 160. See, e.g., David Rogers, Capital Climate Discomfits Multinationals – Business Frauds, Patriotic Fever Dominate Debates on Offshore Havens, Tax Breaks, WALL ST. J., July 25, 2002, at A4 (reporting the issue of anti-inversion legislation in the wake of Stanley Works’ effort to reincorporate in Bermuda as an potent congressional campaign issue). 2013] Tax Planning and Initial Incorporation Location 353 structure.161 There are numerous data points related to the possibility of Congressional change adverse to US-parented MNCs. US-based advisors have access to fewer data points related to the possibility of either US or non-US change adverse to non-US parented MNCs. For example, US advisors may be able to predict likely Congressional action related to the repeated renewal of specific tax breaks. “Look-through” rules for payments between related controlled foreign corporations and the subpart F active financing exception provide two current examples of provisions that US-parented MNCs regularly lobby to preserve to ensure that their treatment under the existing set of rules does not worsen.162 The repeated renewal of such tax breaks provides ample opportunity to observe corporate interest groups’ ability to lobby and influence legislation and to support a conclusion on the part of advisors and 161. Uncertainty aversion describes a preference for avoiding situations in which the chances of different possible outcomes are unknown. See David Schmeidler, Subjective Probability and Expected Utility Without Additivity, 57 ECONOMETRICA 571 (1989); see also Larry G. Epstein, A Definition of Uncertainty Aversion, 66 REV. ECON. STUD. 579 (1999). Uncertainty aversion can be analyzed separately from risk aversion, which refers to a preference for avoiding situations in which the outcome is not known, but the chances of different possible outcomes are known. See Frank H. KNIGHT, RISK, UNCERTAINTY & PROFIT Part III Ch VIII (1921) (using “risk” to mean a measurable or mathematical uncertainty like that faced in a game of chance and “uncertainty” to mean an unmeasurable uncertainty). See also KENNETH J. ARROW, ASPECTS OF THE THEORY OF RISK-BEARING (1965) (providing risk aversion model); Sarah B. Lawsky, Modeling Uncertainty in Tax Law, 65 STAN. L. REV. 241, 259–61 (2013) (citing Daniel Ellsberg, Risk, Ambiguity, and the Savage Axioms, 75 Q.J. ECON. 643 (1961)). The impact of uncertainty about future law changes has been considered broadly. See, e.g., Guy Halfteck, Legislative Threats, 61 STAN. L. REV. 629 (2008). Some tax research has focused on uncertainty under steady-state policies. For example, Sarah Lawsky has pointed out that a taxpayer’s uncertainty aversion may function as a built-in penalty. Sarah B. Lawsky, Probably? Understanding Tax Law’s Uncertainty, 157 U. PA. L. REV. 1017, 1073 (2009) (“[T]heoretical models of tax compliance may benefit from taking into account an additional aspect of deterrence: the built-in penalty that is uncertainty.”). Others have considered the interaction between penalties and uncertainty aversion. See Mark P. Gergen, Uncertainty and Tax Enforcement: A Case for Moderate Fault-Based Penalties, 64 TAX L. REV. 453, 472 (2011) (considering solutions to the problem that a penalty may overdeter particularly uncertainty-averse taxpayers and underdeter others); Kyle D. Logue, Optimal Tax Compliance and Penalties When the Law is Uncertain, 27 VA. TAX REV. 241, 293–96 (2007) (analyzing strict liability and fault- based tax penalty structures assuming legal uncertainty). 162. See, e.g., Press Release, Nat’l Foreign Trade Council, NFTC Urges Congress to Pass Tax Extenders Legislation, (March 15, 2012) (urging renewal of both provisions on competitiveness grounds). 354 Florida Tax Review [Vol. 14:8 investors that a change adverse to the interests of US-parented MNCs is certainly unlikely. Advisors likely face a higher degree of uncertainty related to the possibility of laws penalizing non-US-parented MNCs. In the United States, there are few data points related to proposed laws targeting non-US-parented firms. In addition, US-based advisors may know less about likely future changes in law in other jurisdictions. If so, higher uncertainty about the likely changes to laws affecting non-US MNCs could discourage use of the non-US MNC structure even if such a structure were advantageous from a tax perspective. V. THEORIZING THE EXCEPTIONS: WHAT FACILITATES A NON-US PARENT STRUCTURE? A. Tax Factors Because more mature and profitable firms are likely to benefit more from a non-US-parented structure, tax factors in general will present more advantages for some firms as opposed to others. The structure used by Bermuda-parented corporations that primarily insure US risks provides an example of the potential power of tax factors to encourage non-US incorporation.163 The Bermuda parent of a US insurance subsidiary must skirt several sets of rules to ensure that it will not be subject to US income tax on insurance or re-insurance premiums it receives.164 First, it must avoid the US rules that tax non-US persons on net income “effectively connected with a US business.”165 Second, it must avoid the US rules that tax non-US persons by imposing a 30 percent withholding tax on gross “fixed or determinable, annual or periodic,” or FDAP, income.166 Bermuda-parented MNCs that insure US risks achieve the first goal, related to avoiding effectively connected income treatment, with the help of a US-Bermuda tax treaty that provides a taxpayer favorable “permanent 163. Nontax regulatory factors, such as the opportunity to take advantage of less stringent investment standards, may also encourage tax-haven incorporation. See Thomas St.G. Bissell, A Comparison of the U.S. Tax Rules for U.S. and Offshore Insurance Products, 32 TAX MGMT. INT’L J. 14 (2003). 164. It must also avoid treatment as a passive foreign investment company, or “PFIC.” This is accomplished by application of the active insurance exception to the PFIC rules. See I.R.C. § 1297(b)(2)(A) and (B); David S. Miller, How U.S. Tax Law Encourages Investment Through Tax Havens, TAX NOTES 167, 173–75 (Apr. 11, 2011) (explaining application of PFIC exception to offshore insurance companies and listing twenty-one publicly traded offshore reinsurance companies). 165. See I.R.C. § 882. 166. See I.R.C. §§ 881(a) (imposing 30 percent tax on FDAP), 1441 (imposing withholding obligation). 2013] Tax Planning and Initial Incorporation Location 355 establishment” provision.167 Permanent establishment rules in treaties allow taxpayers from one treaty jurisdiction (here, Bermuda) to establish more of a presence in the other treaty jurisdiction (here, the US) before the other treaty jurisdiction is permitted to impose income tax.168 The Bermuda-US tax treaty provides among other things that the maintenance of a regular place of business “solely for the purpose of . . . collecting information, for the enterprise of insurance; or . . . advertising [or] for the supply of information, . . . for the enterprise” will not constitute a permanent establishment, even if carried on by a wholly-owned subsidiary, so long as the subsidiary is compensated on an arm’s length basis.169 This does not diverge substantially from the usual permanent establishment definitions in bilateral tax treaties, but it fails to acknowledge that the elements of the insurance business other than the giving and receiving of information can be carried on more easily from afar compared to many other businesses. In addition, it is unusual for the US and a tax haven jurisdiction like Bermuda to conclude a tax treaty that includes a permanent establishment or “business income” provision. Bermuda-parented MNCs avoid FDAP taxation despite the fact that the applicable statute includes US-source “premiums” in the list of items to be taxed;170 and premiums paid to insure US risks are US source income. A revenue ruling states that an excise tax applicable to premiums paid to a foreign insurer supersedes the FDAP tax, and that the FDAP tax substitutes for the collection of income tax.171 The excise tax charges four percent of property and casualty premiums and one percent of reinsurance, life insurance and other policy type premiums.172 The prevalence of the Bermuda-parented insurance structure suggests that the premium excise tax produces a lighter tax burden than the imposition of US corporate income tax on net income would. A 1990 Treasury study used a model to confirm this result.173 A key element of the 167. See William P. Elliott, A Guide to Captive Insurance Companies, 16 J. INT’L TAX’N 22 (2005). 168. See JOSEPH ISENBERGH, INTERNATIONAL TAXATION (3d ed. 2010). 169. U.S.A-Bermuda Insurance Income Tax Convention Act art. 3, Aug. 29, 1986. 170. See I.R.C. §§ 881(a) (imposing 30 percent tax on FDAP), 1441 (imposing withholding obligation). 171. See Rev. Rul. 89-91, 1989-2 C.B. 129. 172. See I.R.C. § 4371. 173. See U.S. TREASURY, REPORT TO CONGRESS ON THE EFFECT ON U.S. REINSURANCE CORPORATIONS OF THE WAIVER BY TREATY OF THE EXCISE TAX ON CERTAIN REINSURANCE PREMIUMS (Apr. 2, 1990) Tables 3, 4, 5 (showing top profitability in no-tax jurisdiction under almost all sets of assumptions under any of a 0 percent, 1 percent or 4 percent premium excise tax). 356 Florida Tax Review [Vol. 14:8 advantaged Bermuda-parented insurance structure is that investment income resulting from invested premiums is not subject to tax.174 B. Nontax Legal Factors Nontax legal factors also help explain why some non-US incorporation locations may be more favored than others. Marine transportation firms that are headquartered in the United States and incorporated outside the United States provide an example of an industry whose non-US organization decisions appear to be driven by various tax and nontax legal factors. Online gaming provides another example of an industry influenced by nontax legal factors. US law exempts income derived from the international operation of a ship if it is earned by a foreign corporation resident in a country that declines to tax similar income earned by US corporations.175 In addition, although some commerce, such as “coastwise” shipping between two US ports, is limited to US-flagged vessels,176 the use of non-US flagships in international commerce including calls at US ports is permitted and provides several nontax regulatory advantages. These include the ability to use a non- US shipyard for vessel construction as well as the possible avoidance of applicable labor regulations, union contracts,177 and a choice of law doctrine that may require a US forum in the event of worker injury.178 Online gambling firms may have had even stronger reasons to incorporate offshore. There is not yet a regulatory framework for online gambling in the United States and for some time its legality was in question in the United States, for example because of potential liability under the Wire 174. See NYSBA, OUTBOUND INVERSION TRANSACTIONS REPORT, supra note 25, at 28. 175. See I.R.C. § 883(a)(1); Peter A. Glicklich & Michael J. Miller, U.S. TAXATION OF INTERNATIONAL SHIPPING AND AIR TRANSPORT ACTIVITIES, 954 TAX MGMT. (BLOOMBERG/BNA) PORTFOLIO 945 (2012). 176. See Timothy Semenoro, The State of Our Seafaring Nation: What Course Has Congress Laid for the U.S. Maritime Industry?, 25 TUL. MAR. L.J. 355, 358 (2000). 177. Id. at 368–69. 178. See GRANT GILMORE & CHARLES LUND BLACK, JR., THE LAW OF ADMIRALTY (2d ed. 1975). Leading flag jurisdictions such as Liberia and Panama undertake to both provide satisfactory vessel safety and inspection requirements and also facilitate more cost-effective construction and operation. See, e.g., Brad Berman, Does the UNCTAD Convention on the Registration of Ships Need Amending?, http://www.itfglobal.org/seafarers/icons-site/images/120_BERMAN.pdf (emphasizing Liberian commitment to safety standards). 2013] Tax Planning and Initial Incorporation Location 357 Act and the Unlawful Internet Gaming Enforcement Act of 2006.179 One leading company incorporated in Gibraltar conducted an IPO on the London Stock Exchange in 2005 using a proxy statement that reportedly disclosed that 90 percent of its customers were US.180 In 2011, the Justice Department provided guidance that permits states to legalize online gaming.181 This raises the question of whether online gaming companies will continue to organize outside the United States.182 C. Liquidity and Other Resource Constraints The description in Part IV of startups’ reasons for incorporating in Delaware under the default approach emphasized liquidity and other resource constraints. This suggests that if a startup had plentiful cash and other resources and did not need to depend upon venture capital financing, it would be more likely to depart from the default Delaware corporation structure. The presence of corporate- and individual-funded startups in the IPO data set, as discussed above, is consistent with this suggestion.183 Similarly, one startup lawyer said that non-US incorporation might follow if a founder was a serial entrepreneur who had had the prior experience of a significant tax hit on a previous investment and who was confident about the availability of financing.184 Another lawyer explained that he would not set such a structure up on a deferred fee basis, but would do so if the client paid “full freight.”185 A desire to access capital markets in the future might prompt organization as a non-US firm in order to make a future acquisition transaction attractive to a strategic acquirer who preferred to use offshore 179. See Nelson Rose & Rebecca Bolin, Game on for Internet Gambling: With Federal Approval, States Line Up to Place Their Bets, 45 CONN. L. REV. 653, 657–69 (2012) [hereinafter Rose & Bolin, Game on for Internet Gambling] (explaining historic legal landscape). 180. See Hurt, Regulating Public Morals, supra note 63, at 415 (reporting on PartyGaming IPO). 181. See Rose & Bolin, Game on for Internet Gambling, supra note 179, at 674–84 (outlining likely state action regarding licensing of online gambling). 182. Cf. Matthew Garrahan, US States Make Play for Global Gaming, FIN. TIMES, Mar. 25, 2013, at 15 (reporting that states may enter into international regulatory compacts with respect to the regulation of online gambling). 183. See supra notes 64-69 and accompanying text. 184. Telephone Interview with Silicon Valley law firm partner (Feb. 4, 2013). One founder reported refusing venture capital funding for a second startup, organizing it as an LLC, and considering a non-US parent structure. Telephone Interview with Silicon Valley entrepreneur (Feb. 11, 2013). 185. Telephone Interview with San Francisco law firm partner (Feb. 16, 2013). 358 Florida Tax Review [Vol. 14:8 cash. The results of informal interviews did not suggest that this factor is thought to affect startup company valuation.186 However, it arguably has affected valuation in some acquisitions. Skype provides a good example, although it is not a pure startup example.187 Microsoft bought Skype, an MNC with a corporate parent resident in Luxembourg, where the corporate tax is 0.4 percent, from a private equity consortium including Silver Lake Partners and from eBay in 2011.188 The price was $8.5 billion. Microsoft’s ability to forecast a low tax rate for Skype profits and its ability to use offshore cash for the acquisition without paying any residual repatriation tax may have enabled it to pay much more than it would have been able to pay otherwise.189 D. Investor Preferences A number of tax-haven-incorporated firms in the IPO data set had investors or founders with links to a non-US jurisdiction. Some tax factors correlate with this result. For example, a non-US investor does not face any PFIC risk as a result of holding non-US company stock. In addition, the laws applicable in the non-US investor’s country and the country of incorporation will determine outcomes including treaty benefits such as withholding tax relief, and tax reporting requirements. These may be more favorable than those prescribed by US law. 186. E.g., Telephone Interview with Silicon Valley entrepreneur (Feb. 11, 2013) (remarking that pharmaceutical companies really care about the technology); Telephone Interview with San Francisco law firm partner (Feb. 16, 2013) (expressing doubt that an offshore structure would affect valuation). 187. A similar example is provided by the 1994 acquisition of Syntex, a Silicon Valley firm, by Roche Holdings Ltd. See Milt Freudenheim, Roche Set to Acquire Syntex, N.Y. TIMES, May 3, 1994, at D1, D6 (reporting that Syntex was based in Palo Alto and incorporated in Panama). Roche Holdings Ltd. was a Swiss corporation; it owned a non-resident Canadian corporation, which owned the acquiring company, a Panama corporation. See Syntex Corp., Tender Offer Statement (Schedule 14D-1A) (Sept. 9, 1994), http://www.sec.gov/Archives/edgar/data/96000/0000950103-94-003492.txt. 188. See Richard Waters, Tim Bradshaw & Maija Palmer, Microsoft in $8.5bn Skype Deal, FIN. TIMES, May 10, 2011, at 1 (reporting profit of $5 billion for the investors who had purchased 70 percent of Skype eighteen months before the announcement of the Microsoft deal). 189. See Ronald Barusch, Microsoft’s Brilliant, Legal Tax Dodge, WALL ST. J. DEALPOLITIK (May 11, 2011, 2:59 PM), http://blogs.wsj.com/deals/2011/05/11/dealpolitik-lesson-from-microsoftskype- congress-must-fix-corporate-tax-law/ (estimating deal price of $5.5 billion if Skype “had been a Delaware corporation run out of Silicon Valley” and Microsoft had used offshore cash). 2013] Tax Planning and Initial Incorporation Location 359 Non-US connections also mitigate some of the risks presented by a non-US incorporation structure. Venture firms that focus on particular non- US sectors often use non-US incorporation jurisdictions in those sectors, for example. Their local presence may reduce legal risks, for example those related to the ability to pursue court remedies in the event of controversy. In addition, non-US connections support easier access to local, non- US attorneys, which can reverse the gatekeeper effects that cause US lawyers to recommend Delaware incorporation. A local Indian or Israeli lawyer, for example, is likely to recommend Indian or Israeli incorporation. A local Chinese lawyer may recommend that a firm follow a familiar structure that uses a parent company incorporated in the British Virgin Islands.190 VI. CONCLUSION This Article, consistent with previous literature and with the support of additional informal interview results, presents the default norm of US incorporation, in particular Delaware incorporation, for US-based startups. The US incorporation structure dominates. However, there are exceptions to the general rule, for example in the insurance, marine transportation and online gambling industries and in isolated cases where resources permit and investors prefer a non-US incorporation structure. This Article theorizes the dominant structure. It explains that US- parented MNCs can often achieve tax-advantaged structures and may obtain other valued corporate governance and other legal advantages. In addition, liquidity and other resource constraints support US incorporation for many startups. The dominant structure is particularly entrenched because advisors to startup firms, in particular lawyers, firmly embrace it. This Article also theorizes the exceptions to the dominant structure. Some US-based, non-US incorporated firms, including insurance firms, may make their organizational choice because of the tax advantages of non-US incorporation. Other US-based, non-US incorporated firms, including marine 190. Historically, legislative restrictions relating to foreign ownership of Chinese firms, shareholder and creditor rights and listing approval made non-US ownership relatively attractive. See Nicholas Calcina Howson & Vikramaditya S. Khanna, The Development of Modern Corporate Governance in China and India, in CHINA, INDIA AND THE INTERNATIONAL ECONOMIC ORDER 513, 542-45 (Muthucumaraswamy Sornarajah & Jiangyu Wang eds. 2010). Tax havens may have provided corporate governance advantages as well as tax advantages. See Dhammika Dharmapala & James R. Hines, Jr., Which Countries Become Tax Havens, 93 J. PUB. ECON. 1058 (2009). Finally, substantial foreign direct investment tax incentives existed until 2007 and were available if investment was made through a non-Chinese corporation. See Jinyan Li, The Rise and Fall of Chinese Tax Incentives and Implications for International Tax Debates, 8 FLA. TAX REV. 669 (2007). 360 Florida Tax Review [Vol. 14:8 transportation and online gambling firms, may make their organizational choice because of non-tax legal advantages of non-US incorporation, as well as any tax advantages. Lower resource constraints and investor preference for non-US incorporation, for example because of independent sources of capital and/or business links to the jurisdiction of incorporation, also facilitate the incorporation of a US-based firm in a non-US jurisdiction. II. Why Startups’ Place of Incorporation Matters for Tax B. Being a Non-US-Parented MNC III. US Incorporation: A Dominant Structure with Exceptions A. US-Based Startups Generally Incorporate in the US A “startup,” for purposes of this Article, is a relatively new private business enterprise with global ambitions. This Article focuses on initial incorporation decisions of “relatively new” firms and not on later changes in a firm’s place of incorpora... Some of the material presented in this Article, including the informal interviews, draws heavily from venture capital experience, although the definition of “startup” used here is not limited to venture-backed firms. Other sources of initial outside c... Prior work that investigates the choice of organizational form by US-based startups backed by US venture firms observes that US incorporation, particularly Delaware incorporation, is the market norm.34F Venture capital investments generally use stand... B. Exceptions IV. Explaining the Dominant Structure A. Limited Tax Benefits of a Non-US-Parent Structure C. Liquidity and Other Resource Constraints D. Considering Future Law Changes A. Tax Factors B. Nontax Legal Factors C. Liquidity and Other Resource Constraints D. Investor Preferences