Insurance Company Coalition Expresses Support for Anti-Tax-Haven Legislation

We are pleased to submit the attached supplemental comments to reiterate our strong support of the Senate Finance Committee staff discussion draft and H.R. 6969 (introduced by Rep. Neal). Our comments also respond to, and rebut, comments submitted by those opposed to the legislation.

Our Coalition represents 14 major U.S.-based insurance groups with over 150,000 employees in offices located throughout the United States. Collectively, we pay substantial U.S. taxes, invest significantly in the municipal bond market, and offer millions of U.S. individuals and businesses financial protection from unpredictable risks.

This legislation is essential to close a current law loophole that permits foreign-controlled insurers to strip their income into tax havens and avoid paying billions of dollars in U.S. taxes annually. This loophole provides foreign-controlled insurers a significant tax advantage over their domestic competitors in attracting capital to write U.S. business. There is no credible reason why the United States tax system should favor foreign-owned groups over domestic insurers in selling insurance in the United States.

Contrary to opponents' claims, this legislation will not harm insurance capacity, increase prices to consumers, contravene treaties, or interfere with free trade. The legislation merely prevents foreign insurers from stripping their U.S. income to tax havens and restores a level competitive playing field.

At a time of burgeoning deficits and possible tax increases on U.S. workers and businesses, it seems unfathomable that we would continue to allow foreign-based insurers to avoid U.S. tax on their U.S.-based business. It is time to close this loophole and prevent further erosion of our tax base.

Once again, our coalition wants to thank you and your staffs for this opportunity to comment, and for your efforts to address this loophole and eliminate the unfair competitive advantage. We are happy to help address any concerns that may arise and are available at any time to discuss this issue.

Sincerely,

The loophole. Domestic insurance companies with foreign-based parents can escape U.S. tax on much of their underwriting and investment income derived from policies covering U.S. risks merely by reinsuring their U.S.-written business with a foreign affiliate in a low-tax or no-tax jurisdiction. By contrast, domestically-controlled insurers must pay current U.S. tax on all profits from similar policies. This provides foreign groups a significant and unfair tax advantage over domestic groups in attracting capital for writing insurance to cover U.S.-based risks. The legislation is necessary to preserve U.S. tax on profits from U.S. business activity and to treat all insurance of U.S. risks comparably.

Different purposes served. Third party reinsurance and related party reinsurance serve different purposes and should not be confused. Third party reinsurance is an arm's length arrangement that shifts insurance risk (and a portion of potential profits) to the third party reinsurer. This helps to ensure that the financial liability for catastrophes or a series of large (or unexpected) losses does not overburden one company and adds overall capacity to the market.

Related party reinsurance, by contrast, does not shift risks outside the group and thus has little to do with adding capacity for catastrophe coverage. Rather, it is used as an alternative to a capital infusion to manage intra-group capital and to provide ratings support for subsidiaries. For foreign-based groups, it is also used to strip income overseas to avoid U.S. tax liability. There is no reason foreign-based groups should obtain a competitive tax advantage over U.S.-based companies for using this capital management technique.

Competitive tax advantage. Use of related party reinsurance provides a significant competitive tax advantage for foreign-based insurers. For example, over the past several years, Bermuda-based companies have enjoyed a significantly higher average after-tax return on equity (ROE) and generally lower effective tax rates than U.S.-based companies. Also, the difference between pre-tax ROE and after-tax ROE is significantly higher for domestic companies than for Bermuda-based companies.

The 1% reinsurance excise tax is insufficient to offset the income tax advantage provided by the loophole. Moreover, the excise tax is often waived by treaty. In fact, several companies have recently redomesticated to Switzerland to avoid imposition of the excise tax.

Dowling and Partners, a leading expert on the insurance industry, concludes, "The percentage of U.S. (re)insurance premium written by entities enjoying an offshore tax advantage will continue to grow. It's simple economics that the movement offshore will continue."

Migration of capital. The unfair competitive tax advantage for foreign-based insurers already has caused a significant portion of the capital base of the P&C industry to move offshore, resulting in significant erosion of U.S. tax revenues. This trend is alarming and must be addressed.

• Several U.S. insurance groups "inverted" into tax havens, moving their capital and tax base offshore, e.g., White Mountains Group, Everest Re Group, Arch Capital Group, and others. In addition, several new holding companies have been formed (and several U.K.-based companies have redomesticated) in tax havens.

• Offshore companies have used the competitive advantage to attract capital and to acquire U.S. companies or lines of business.

• Since Katrina, over $30 billion of new capital has been raised by the insurance industry, most of which has gone to offshore tax havens.

To date, consumer pricing has been unaffected by the competitive tax advantage. For example, when prices recently softened in the market, the head of the Association of Bermuda Insurers and Reinsurers (ABIR) stated that the profits Bermuda companies are making "is going back to shareholders. It's not contributing to the downward [pricing] spiral."¹ This is because pricing is extremely competitive, determined by market supply and demand.

Thus, requiring foreign groups to pay equitable U.S. taxes should not adversely affect capacity or pricing in the insurance market.

Treaty compliant. The legislation is consistent with our treaty obligations because it does not "materially disadvantage" foreign groups relative to domestic insurers in writing coverage of U.S.-based risks and it directly relates to a "tax relevant difference" between foreign and U.S.-based insurance groups. Also, foreign groups affected by the proposal may elect to forgo the reinsurance or be treated as U.S. taxpayers with respect to such income. Thus, they are always able to obtain the same treatment as a U.S. group insuring U.S.-based risks. Finally, tax treaties do not require a treaty state to apply "the non-discrimination principle" to allow deductions with respect to certain payments between related parties.

Transfer pricing rules inadequate. Transfer pricing rules cannot reliably police these transactions because of the uniqueness of reinsurance transactions and the lack of meaningful comparables. Congress attempted to strengthen the transfer pricing rules in 2004 in an effort to close the loophole, but most foreign-based groups continue to have exceptionally low effective tax rates and the migration overseas caused by the loophole has not abated. This is because the core problem is not a transfer pricing issue. The reinsurance ceded to related parties is significantly greater than, and different in quality and purpose to, the reinsurance ceded to third parties. The flight of companies, capital, and tax revenue can only be staunched by restricting the ability to strip profits overseas.

No adverse impact on crop insurance market. Crop insurance is a federally subsidized "take-all-comers" program with prices set by the Government without individual underwriting. Insurers have no control over pricing. This market is attractive for both growth and diversification. Thus, enactment of the proposed legislation should not impact pricing or capacity in the crop insurance market.

Legislation necessary to restore parity. Contrary to opponents' claims, the legislation is not protectionist. It merely levels the playing field by treating all insurers of U.S. risks equitably from a tax perspective. Foreign-based competitors should not be advantaged over U.S. insurers in serving the U.S. market. Under the legislation, foreign insurers will have continued access to the U.S. market, but will be taxed in a manner similar to U.S. groups. Foreign insurers may continue to avail themselves of non-tax benefits that may result from reinsuring with a foreign affiliate, such as a relaxed regulatory environment.

Loophole diminishes market for tax-exempt bonds. Domestic insurers are major investors in tax-exempt bonds, while foreign groups, having little or no need for tax-exempt income, are not. Thus, as the use of related party reinsurance increases and more investment income is stripped overseas, the market for tax-exempt bonds will diminish, causing the cost of state and local government borrowing to increase.

When this loophole first appeared in the late 1990s, it was described as the foreign-controlled insurance companies' "own Bermuda Triangle... Instead of ships and planes vanishing without a trace, these companies have figured out how to make their federal tax burden disappear."²

Since that time, the magnitude of the problem has increased and Treasury has lost billions of dollars in tax revenues annually. A number of U.S. P&C companies (e.g., Arch Capital Group and Everest Re Group) have inverted to low-tax or no-tax countries, and many new holding companies have been formed (and other foreign companies have redomesticated) in tax havens, to take advantage of this loophole. In either case, these foreign-based companies have sought, and will continue to seek, to use this competitive advantage to attract capital and to acquire U.S. companies or U.S. lines of business. Since Katrina, over $30 billion in new capital has been raised by the insurance industry, most of which went to offshore tax havens.

As a result of this continued migration, premiums ceded to related-party foreign reinsurers have more than doubled over the last seven years, most of which is to affiliates in low-tax or no-tax jurisdictions. In 2007 alone, over $34 billion of U.S. P&C insurance premiums moved offshore to related foreign affiliates, stripping billions in associated tax revenues.

In 2004, Congress expressed concern that "foreign related party reinsurance arrangements may be a technique for eroding the U.S. tax base,"³ and attempted to fix the problem by tightening the transfer pricing rules. However, because the essential problem is not a transfer pricing issue, these changes have not been effective in stemming the migration abroad. Additional legislation is essential to eliminate this unfair tax advantage, which erodes our tax base and undermines the competitiveness of our domestic insurance industry.

The Senate Finance Committee discussion draft and H.R. 6969 both provide an appropriate and effective remedy to the problems caused by offshore related party reinsurance. Similar to the earnings stripping rules under section 163(j), the proposed legislation strikes a balance and only targets excessive related party reinsurance transactions that are being used to strip income out of the U.S. tax base and avoid U.S. tax.

The leading opponents to the legislation include several former U.S. companies that inverted into tax havens, and companies formed in tax havens, to take advantage of this loophole. In their efforts to preserve their tax break, the opponents make several misleading, and often contradictory, arguments. For example, as discussed more fully below, they have argued that they have no significant tax advantage relative to U.S. companies. Yet, at the same time, they argue that eliminating the competitive advantage under the legislation by taxing them similarly to U.S. companies would adversely affect capacity and pricing in the U.S. marketplace.

It is understandable that they wish to maintain this unfair advantage -- after all, over the past several years, they have used it to attract significant capital abroad, and to acquire numerous U.S. companies and lines of business. However, despite the opponents' rhetoric, there is no tenable policy reason why the U.S. tax system should favor foreign companies over domestic companies in writing direct insurance coverage of U.S. risks, attracting domestic capital, or acquiring U.S. companies or lines of business.

Background -- Present Law

Insurance companies generally are allowed a deduction for premiums paid for reinsurance coverage. Whenever the reinsurance is ceded to a foreign reinsurer, a 1% premium excise tax generally applies unless waived by treaty. In addition, a domestic insurance company that is part of a domestic group reinsuring risks with a foreign affiliate must continue to pay tax (under subpart F of the Code) on the income earned with respect to such reinsured risks. However, subpart F does not apply when a domestic insurance company that is part of a foreign group reinsures with a foreign affiliate. As a result, the foreign group is subject only to the 1% excise tax (if not waived by treaty) and avoids paying U.S. tax on the reinsurance profits. Thus, if the foreign parent is located in a low-tax jurisdiction, little or no taxes are paid by the foreign group on its profits, even on income that is directly related to the U.S. insurance activity.

Discussion of Issues

1. Related party reinsurance serves different purposes than third party reinsurance

Opponents to the legislation conflate third party reinsurance and related party reinsurance in an attempt to associate the latter with the benefits of the former. For example, they state "the U.S. insurance market could not provide adequate coverage to Americans without the availability of foreign reinsurance capital."⁴ They also assert that a majority of damage payments for destruction caused by the recent hurricanes came from foreign companies.

These statements are red herrings that have nothing to do with the target of the proposed legislation. These statements -- and others like them -- reflect the benefits of third party reinsurance, which is expressly unaffected by the legislation. Any implication that the legislation will adversely affect the availability or pricing for third party foreign reinsurance -- which adds overall insurance capacity to the U.S. market -- is highly misleading.

It is important that policymakers not confuse these two types of reinsurance and understand the different purposes they serve.

Third party reinsurance is an arm's length arrangement between unrelated parties that shifts a portion of the risks an insurer assumes to the third party reinsurer. In exchange, the primary carrier effectively forgoes and transfers a portion of any prospective profit to the third party reinsurer. Such reinsurance spreads and diversifies risk so that the financial liability for catastrophes or a series of large or unexpected losses does not overburden one company. This spreading of risk adds overall capacity to the market.

By contrast, a reinsurance transaction among affiliates transfers neither the risk nor the potential profit outside the group. Because each company within an affiliated group is attempting to maximize profits of the group as a whole, their interests are aligned. Thus, in contrast to third party reinsurance, related party reinsurance transactions -- which are commonly used to strip profits overseas -- are generally effectuated without additional underwriting and often as a mere bookkeeping entry.⁵

The recent Brattle Group report commissioned by the opposition to the bill acknowledges this essential difference. According to the report, in the case of third party reinsurance, there is "an incentive for the insurer to transfer the worst risks [to the third party reinsurer] and/or to be lax in its underwriting. If the insurer and reinsurer are part of the same corporate group, their incentives are better aligned."⁶

In fact, in the case of related party reinsurance, the incentives are the opposite from third party reinsurance. A recent study cited in the Brattle Group report did an empirical analysis and found that affiliated insurers with catastrophe exposure tend to retain risks of catastrophic losses, rather than spread them through related party reinsurance to affiliates. The empirical analysis showed that third party reinsurance and affiliate reinsurance serve "different purposes." The study summarized its findings as follows:

While U.S. companies commonly use affiliate reinsurance for capital management and to offer ratings support, there is no federal tax benefit for doing so.⁸ By contrast, foreign-based groups obtain a significant tax advantage for using related party reinsurance. First, by effectively shifting their reserves on domestically-written business overseas, they may avoid U.S. tax on the bulk of their underwriting and investment income. Also, use of related party reinsurance allows them to avoid U.S. rules requiring discounting of loss reserves, which accelerate the payment of taxes by domestic groups.

It is indefensible that foreign-based groups should obtain a competitive tax advantage from using this capital management technique and shifting business from one pocket to another. The proposed legislation would eliminate the tax advantage, while still allowing foreign-based groups to use this method of capital management on the same terms as U.S. companies.

2. Contrary to opponents' claims, there is a significant competitive tax advantage derived from use of related party reinsurance for foreign-based insurance groups relative to domestic insurers.

Opponents argue that a competitive tax disadvantage for U.S. insurers is "flatly contradicted by the facts" and that the bill is unnecessary as the U.S. P&C industry "continues to thrive." However, the opponents offer no support for their claim. The comparative statistics actually show that offshore companies indeed do have a competitive tax advantage.

While the entire P&C insurance industry has reported generally favorable results in the years 2002 to 2008, a composite of publicly traded Bermuda domiciled companies (dubbed "Bermuda, Inc." by Dowling & Partners Securities, LLC) has outperformed the U.S. P&C Industry on a return-on-equity (ROE)⁹ basis over the period, despite the fact that two out of the seven years have included above-average catastrophe losses. The vast majority of catastrophe losses in those years borne by "Bermuda, Inc." arose from third party catastrophe reinsurance since, as A.M. Best market share data show, foreign insurers have a very small share of the market for direct homeowners and commercial multiperil coverage in the coastal states.

The analysis below separates the impacts of catastrophes, as third-party reinsurance will be unaffected by the legislation. The average ROE for "Bermuda, Inc." exceeded that of the U.S. P&C Industry both before and after adjusting for catastrophe losses. After adjusting for the catastrophe losses in 2004 and 2005 for both groups, the "Bermuda, Inc." ROE exceeded that of the U.S. P&C Industry by 6.3 percentage points. Clearly, the comparative data points to an advantage.

 

ROE: P/C vs. "Bermuda, Inc."

 

 

 

"Bermuda, Inc." is a composite of publicly traded Bermuda (re)insurers defined by Dowling & Partners

Further analysis shows that a large portion of the advantage is derived from the absence of taxes. As shown in the table below, the differential between pre-tax ROE and after-tax ROE for the Bermuda Inc. group is a negligible average of 1.4 percentage points. By contrast, the differential between the pre-tax ROE and the after-tax ROE for the U.S. P&C Industry averages 4 percentage points.

 

Comparative Pre-Tax and After-Tax Operating

 

ROE Differentials

 

 

 

"Bermuda, Inc." is a composite of publicly traded Bermuda (re)insurers defined by Dowling & Partners

The ABIR comment letter includes a chart demonstrating the aggregate net premiums written for the U.S. P&C industry and the offshore affiliate reinsurance premiums from 1997-2007, but it fails to show the relative growth. During that time frame, net premiums written by the U.S. P&C Industry grew 62%, or at a compound average annual rate of 4.9%. By contrast, offshore affiliate reinsurance premiums grew by over 700%, or at a compound average annual rate of 23%. As the Class of 2005/2006 startups expands from principally third party reinsurers into direct insurance (combined with affiliate reinsurance) through the acquisition or formation of U.S. subsidiaries, significant growth in offshore affiliate reinsurance is expected to continue.

3. The reinsurance excise tax is insufficient to eliminate the competitive tax advantage.

As described above, use of related party reinsurance allows foreign-controlled insurance groups to avoid virtually all U.S. income tax on their U.S. written business. The only tax imposed (which is often waived by treaty) is a one-percent excise tax on the reinsurance premiums paid from the U.S. member to its offshore affiliate. By contrast, a U.S.-controlled insurer must pay income tax on its underwriting and investment income. This difference in treatment gives foreign-based insurers a competitive advantage in raising capital from investors over domestic-based insurers because (as shown above) it provides higher ROEs for foreign-based groups.

Opponents have argued that the reinsurance excise tax is sufficient to eliminate any competitive income tax advantage. In testimony before the Senate Finance Committee and in the press, Donald Kramer, the CEO of Ariel Reinsurance Company, has argued that there is no significant tax advantage for foreign-based companies because they are subject to the 1% reinsurance excise tax and get no tax benefit for their losses.¹⁰ However, Mr. Kramer himself previously explained his reasons for establishing a reinsurance company in Bermuda to serve the U.S. market:

Dowling and Partners, a leading expert on the insurance industry, has written for several years about the "Bermuda 'Quota Share' Tax Option/Advantage" which it often refers to as "the better mousetrap."

The economics are indeed "simple." An excise tax of 1% on $100X of premium is equivalent to a 35% income tax on only $2.85X of profit. Therefore, if a foreign-owned insurer believes it will earn more than merely $2.85X of profit from underwriting and investment, it will elect to bear the excise tax and reinsure with an affiliate in a low-tax or no-tax jurisdiction.¹⁴

As a further illustration, assume that a foreign-controlled U.S. insurer reinsures $100X of premium with an affiliate in a no-tax jurisdiction. After upfront expenses of $20X, the affiliate earns 6% per year on its investment of the remaining $80X. Assume further that the remaining $80X is paid out ratably in claims over five years (i.e., a five-year "tail"). The foreign affiliate would earn nearly $17X of investment income over the five years. The U.S. tax savings would be $5X (approximately $6X of avoided U.S. income tax -- the $1X excise tax). And this example assumed no underwriting profit! The benefit is even larger for longer-tailed lines.

Moreover, it is important to note that this small excise tax is not even imposed in many cases because it is waived by treaty. For example, ACE recently announced it was moving to Switzerland to take advantage of a treaty exemption, saying, "Switzerland affords us the security of a network of tax treaties," and a change of jurisdiction may "help reduce reputational, political, regulatory and financial risks" to the company.

A leading Bermuda insurance executive, speaking on the condition of anonymity, explained the motives for ACE's move more fully as follows:

In summary, the cost of the reinsurance excise tax is generally significantly less than the benefit of earning investment income that is free of any U.S. federal and state tax. Moreover, the reinsurance excise tax is often waived by treaty. The existence of the excise tax is therefore insufficient to prevent foreign-based insurance groups from effectively electing out of the U.S. income tax system through the device of related party reinsurance.

4. Contrary to opponents' claims, the bill will not adversely affect insurance capacity for catastrophe coverage.

Various comment letters from opponents to the bill allege that the bill will adversely affect insurance capacity in the United States, particularly in States that have high catastrophic risk, such as the coastal states.

This is untrue for several reasons. First, the opponents try to obfuscate the issue by using data regarding third party reinsurance. The bill specifically does not affect third party reinsurance (i.e., reinsurance that spreads risks among unrelated parties). Thus, while it is true that third party reinsurance written by foreign insurers provides needed capacity for catastrophic coverage, this capacity is not affected by the legislation.

By contrast, the target of the bill -- excessive related party reinsurance -- adds no additional capacity to the market, but rather requires a mere bookkeeping entry to move premium from the U.S. company's books to the foreign parent's books. The risk remains in the same corporate group, but underwriting and investment income is shifted to avoid U.S. tax.

Moreover, A.M. Best market share data show that foreign insurers currently have a very small share of the market for direct homeowners and commercial multiperil coverage in the coastal states. Thus, the target of the bill -- affiliate reinsurance -- plays little, if any role, in providing catastrophe coverage in these markets. For example, in Florida, the Bermuda insurers collectively have less than half a percent of the homeowners' insurance market and foreign insurers collectively have less than 3 percent. As the table below shows, the figures are similar for other coastal states and the overall United States as well.

                                                          Coastal

In fact, we believe the proposal will help to sustain a healthy domestic insurance market and preserve the real reinsurance market that truly spreads insurance risk among unrelated companies as foreign reinsurers will refocus on their core third party reinsurance business.

5. The bill will not adversely affect the pricing of insurance to the U.S. market.

The market for writing U.S. property and casualty business is highly competitive and a marginal cost increase to a small number of market players should not affect prices. In practice, the price of a policy is determined by how low a particular competitor is willing to go. When prices are rising (a "hard" market),¹⁷ additional capital flows into the market, attracted by the prospects of good returns. The combination of higher prices and additional capacity eventually turns the market soft and prices decline to a level where losses cause a reduction in capital, and then prices rise again.

Historically, the competitive tax advantage has not resulted in lower prices to consumers. For example, the last time significant additional capital was being raised in the U.S for foreign-domiciled companies to write U.S. business (in the years following 9-11 in 2001 and following the hurricanes in 2005) there was a dramatic hardening of pricing in the U.S. P&C insurance market. Had the tax advantage not existed, domestic companies still would have raised additional capital because of the higher expected returns. Instead, the capital flowed offshore to garner even higher prospective returns as a result of the tax advantage.

Recently, when pricing softened in the market, the head of the Association of Bermuda Insurers and Reinsurers was quoted in an article in National Underwriters as saying that the profits Bermuda companies are making "is going back to shareholders. It's not contributing to the downward spiral [in commercial insurance pricing]."¹⁸

In other words, pricing in the U.S. property and casualty insurance market -- whether in a soft market or a hard market -- has not been impacted by the tax advantage or where the capital is raised.

It is possible that, if foreign companies continue to be allowed to avoid U.S. taxes on their U.S. business, they might eventually seek to pass some of this benefit on to consumers. However, as historically has been the case, it is more likely that the benefits of this unfair tax advantage will be used to expand operations at the expense of U.S. competitors or be distributed to investors, rather than be passed on to consumers. In any event, the simple fact is that any benefit arises from an unfair tax advantage for foreign-based companies at the expense of their U.S. competitors and other U.S. taxpayers. The U.S. tax system should not favor foreign-based insurers over U.S. companies in writing insurance covering U.S. risks.

6. The proposal does not violate the terms of existing treaties or the General Agreement on Trade in Services (GATS).

As explained below, the bill is consistent with our tax treaty obligations and does not violate the non-discrimination clauses of existing tax treaties.

There are two non-discrimination provisions under Article 24 of the U.S. Model Treaty that are relevant with respect to the proposal. First, in Article 24(4), disbursements to residents of a treaty partner must be "deductible under the same conditions as if they had been paid" to a U.S. resident.¹⁹ Second, under Article 24(5), an enterprise controlled by residents of a treaty partner cannot be subjected to taxation that is "more burdensome" than taxation of U.S. controlled enterprises.²⁰

The Model Treaty technical explanation explains the purposes of these provisions. They are intended to ensure that "two persons who are comparably situated must be treated similarly." Also, while the operative language is different in the various articles, the Model Treaty technical explanation clarifies that, in both cases, only differences that "materially disadvantage" a foreign person relative to a domestic person are to be considered.

i. Not "comparably situated" -- According to the Model treaty technical explanation, the non-discrimination rules "apply only if the nationals or residents of the two states are comparably situated" and do "not prohibit differing treatment of entities in differing circumstances." The technical explanation further elaborates, "If the difference in treatment is directly related to a tax-relevant difference in the situations of the domestic and foreign persons being compared, that difference is not to be treated as discriminatory." Thus, a foreign-controlled entity is only protected if it is "in substantially similar circumstances both in law and in fact" to domestically-controlled entities.

The Model Treaty technical explanation provides several examples of tax relevant differences. First, in a situation analogous to the related party reinsurance problem addressed by the bill, the technical explanation states that taxing a distributing corporation under section 367(e) on a distribution to a foreign shareholder is permissible discrimination because a liquidating distribution by a foreign-owned corporation removes U.S. corporate assets from the U.S. taxing jurisdiction, while a liquidation of a U.S.-owned corporation would not.²¹ Another example provided by the Model Treaty technical explanation is the fact that nonresident aliens cannot be S corporation shareholders because they are not subject to U.S. tax and thus are not similarly situated to U.S. taxpayers.

Similar to the above examples, foreign-controlled groups should not be treated as "comparably situated" to U.S.-controlled groups because the ceding of reinsurance by the U.S. member of a foreign-controlled group to a foreign affiliate effectively removes underwriting and investment income from U.S. taxing jurisdiction.

ii. No material disadvantage -- According to the Model Treaty technical explanation, "only differences in tax treatment that materially disadvantage the foreign person relative to a domestic person" are discriminatory and subject to the provisions of Article 24. The proposal does not materially disadvantage foreign-controlled entities relative to U.S.-controlled entities because it is merely intended to place reasonable limits on a tax advantage that foreign-controlled insurers enjoy over U.S.-based groups in writing policies to insure U.S. risks. Moreover, merely by electing to forgo the offshore related party reinsurance (or reinsuring to a domestic affiliate), the foreign-based group can ensure that it is taxed similarly to a U.S.-controlled entity and thus is not materially disadvantaged.

Finally, the bill provides an election by the foreign affiliate to be taxed on the reinsurance as a U.S. taxpayer. In passing the Foreign Investment in Real Property Tax Act of 1980 (FIRPTA), Congress provided a similar election to be treated as a U.S. taxpayer.²² According to the legislative history, the FIRPTA election was intended to avoid any treaty discrimination claims by ensuring that foreign taxpayers were not "treated less favorably than a domestic corporation carrying on the same activities." Thus, for reasons similar to FIRPTA, the bill's inclusion of an election to be taxed as a U.S. taxpayer should put to rest any possible argument that foreign-controlled taxpayers are "materially disadvantaged" relative to U.S.-based groups.

iii. Treated the same as U.S. persons -- The U.S. has consistently taken the position that the non-discrimination rules do not apply to cases where any difference in tax treatment is based on a distinguishing characteristic other than nationality (e.g., is the payee subject to later U.S. tax?).

For example, the technical explanations of the Swiss and German treaties state that the different treatment under section 367(e) "is not based upon the nationality of the owners of the distributing corporation, but is based upon whether such owners would be subject to corporate tax if they subsequently sold or distributed the same property," equating a foreign corporation to a tax-exempt organization for purposes of the test. Similarly, section 163(j) is viewed as non-discriminatory in part because it applies, not only to foreign related parties, but also to stripping payments to U.S. tax-exempt entities.²³

The proposal applies to any reinsurance premium paid to an affiliate if no U.S. income tax is imposed on the premium income. Thus, similar to section 163(j), the proposal should be treated as non-discriminatory, because it applies to U.S. persons, as well as foreign persons, based upon whether the affiliate recipient of the premium would be subject to U.S. tax on the associated earnings.

iv. Payments to Related Persons -- Under the Model Treaty technical explanation, neither treaty state is required to apply "the non-discrimination principle" to allow deductions in certain cases involving payments between related parties. For example, the denial of interest deductions to related persons under section 163(j) is allowed.

The proposal is modeled on section 163(j) and only denies deductions for excessive reinsurance premiums paid to related persons. Thus, given that the amount of reinsurance is excessive and would never be ceded in an arm's length relationship, the United States is free to make appropriate adjustments under Article 9 (similar to the denial of interest deductions under section 163(j) and Article 11).

However, Article XIV of GATS specifically provides that "nothing in this agreement shall be construed to prevent the adoption or enforcement by a Member of measures . . . inconsistent with Article XVII, provided that the difference in treatment is aimed at ensuring the equitable or effective imposition or collection of direct taxes in respect of services or services suppliers of other Members." The proposal is permissible because it is aimed precisely at equitable taxation of insurance underwriting and investment income attributable to U.S. risks that, under current law, can be stripped from U.S. taxing jurisdiction through excessive offshore related party reinsurance.

Moreover, as described above, the proposal provides an election to be taxed similarly to U.S. companies on the affiliate reinsurance premiums. The election ensures that foreign insurers are not treated "less favourable" than domestic insurers. Thus, the election provides "national treatment" and should put to rest any possible argument that the proposed legislation violates Article XVII of GATS.

7. Transfer pricing rules cannot adequately address the competitive tax advantage caused by offshore related party reinsurance.

Recognizing the problem caused by the tax treatment of offshore related-party reinsurance, Congress attempted to close the loophole by strengthening the applicable transfer pricing rules in the American Jobs Creation Act of 2004.

Unfortunately, the adjustment to the transfer pricing rules adopted in 2004 failed to stem the growth of stripping through the use of foreign related party reinsurance. Since 2004, there has been a spate of offshore companies (including many newly-established ones) acquiring U.S. subsidiaries and then having them reinsure the bulk of their business to offshore affiliates.

Moreover, most Bermuda insurance groups writing insurance business through U.S. subsidiaries continue to have exceptionally low effective tax rates. For example, the effective tax rate of the "Bermuda, Inc." public companies that have used affiliate reinsurance has averaged 10.8% over the past three years, while the average for the domestic companies that are members of our coalition has been 27%. This indicates they continue to be able to shift the bulk of their profits offshore to avoid tax under existing mechanisms. If the transfer pricing rules worked adequately, most of the profits would have remained in the U.S., the competitive tax advantage would have disappeared, and the bulk of this activity would have dried up. Thus, the 2004 modifications to the transfer pricing rules have not effectively prevented the stripping of income overseas.

Strengthening the transfer pricing rules did not work effectively because the core problem is not a transfer pricing issue. The amount of U.S.-written business being ceded by U.S. members of foreign-based groups to such related offshore parties is significantly greater than would ever be ceded to an unrelated party. The cession of larger than normal amounts is not meaningfully policed by transfer pricing restrictions because it is virtually impossible to find meaningful third party comparables.

Moreover, it is extremely difficult to measure arm's length compensation for risk transfer in this business. Reinsurance transactions are as unique as their reinsured risks, and the pricing relies on complex actuarial analyses. Profitability is determined by the timing of claim payments and the investment income earned until such payments are made. Assessing risk transfer in the insurance business requires knowledge not just of expected loss ratios, but also of payment patterns, which can be changed by mere words in a reinsurance contract. Because of the numerous variations in the terms and conditions of reinsurance contracts, and because the true costs of insurance products are not known until long after prices are established, determining a true arm's length price in this context becomes exceptionally difficult.

The Brattle Group report commissioned by the opposition essentially admits the lack of comparability between affiliate reinsurance and third party reinsurance in making its adverse selection/moral hazard argument. It states that the insurer has an incentive, with respect to third party reinsurance, to transfer its worst risks and/or to be "lax in its underwriting." In addition, according to the report, "negotiations over non-affiliate reinsurance are complex and time consuming because a third party reinsurer must scrutinize potential risks for evidence of these problems."²⁵

The incentives are the opposite with respect to affiliate reinsurance. As discussed above, a recent economic report cited by the Brattle Group found that insurers generally want "to compartmentalize risk of catastrophic losses, such as hurricane damage, within one subsidiary, insulating the rest of the group from adverse loss experience."²⁶ Thus, the nature of the risks being transferred is very different in the case of affiliate reinsurance relative to third party reinsurance. Moreover, if the affiliate reinsurance transaction were truly "arm's length," one would expect that "lengthy negotiations" over the price and terms of the contract and separate risk assessment would be needed. However, as the Brattle Group report states, this is avoided because the interests of the parties are "better aligned."

Finally, use of related party reinsurance allows foreign-based groups to avoid U.S. tax on much of their investment income earned on U.S. risk reserves. It also allows them to avoid U.S. rules requiring discounting of loss reserves, which accelerate the payment of taxes by domestic groups. For example, a large Bermuda-based group engaged in a loss portfolio transfer in the first quarter of 2007, where the U.S. members ceded $1 billion of reserves to the Bermuda affiliate. Through this bookkeeping entry, the transfer of reserves offshore allowed the company effectively to recapture its reserve discount for U.S. tax purposes. Additionally, the investment income on the $1 billion of reserves was transferred offshore to Bermuda and avoided U.S. tax. These disparities are not adequately considered in, and thus captured by, the transfer pricing rules.

When Congress became aware of a similar problem with respect to debt, it addressed the problem (under section 163(j)) by curtailing the amount of tax-favored borrowing by U.S. firms from related foreign parties. As the staff of the Joint Committee on Taxation states in its hearing pamphlet, "a set of definitive rules similar to the earnings stripping rules would probably have a more systematic effect on taxpayers than relying on transfer pricing principles."²⁷

As with section 163(j), disallowing a deduction for premiums paid to affiliates is not contrary to the arm's length standard (as some opponents have claimed). The arm's length standard is not absolute. The Internal Revenue Code is replete with examples of special rules applicable to transactions entered into with affiliates to prevent tax avoidance, even if the arm's length standard is met. In addition to section 163(j), these include sections 45,108(e)(4), 163(e)(3), and 267. These rules recognize that transactions with affiliates can provide opportunities for tax avoidance and are more prone to abuse.

8. Contrary to opponents' claims, the proposal should have little or no impact on the crop insurance market.

The Federal Crop Insurance Program is a "take-all comers" market with prices set by the Federal Government and without individual underwriting. The U.S. Department of Agriculture provides insurance coverage to U.S. farmers through the Federal Crop Insurance Corporation (FCIC). The Risk Management Agency (RMA) administers the federal program on behalf of the FCIC. The RMA develops and approves premium rates and subsidies, expense reimbursements and approves participation of private insurers in the program through a Standard Reinsurance Agreement ("SRA") that is constant among all contract holders. Insurers who process the business have no control whatsoever over pricing. Thus, the exit or entry of firms into the market should have no bearing on price.

Multi Peril Crop Insurance (MPCI) is heavily reinsured by the FCIC through mechanisms that limit both the potential risk and the potential profit to the SRA holder. This government involvement combined with little correlation between this line and other property risks has made this market attractive for both growth and diversification by P&C insurers. It is highly unlikely, given the attractiveness of this market, that the proposal would have any impact on capacity for crop insurance.

Because the SRA holders typically utilize managing general agents (MGAs) to process the business, service to the farmers in processing and paying claims are not likely to suffer in the unlikely event that any firm exits the market. The MGAs collect the premiums and deposit them in an FCIC escrow account. Likewise, losses are paid out of the escrow account all year long. In effect, the program is similar to the Federal Flood program wherein losses are funded primarily by the U.S. Government. Moreover, because money is deposited with RMA within 30 days, there is little or no investment income earned on crop insurance. Thus, any tax benefit from related party reinsurance, and concomitantly any impact of the proposed legislation, should be limited.

9. Contrary to opponents' claims, the legislation is not protectionist and does not favor U.S.-based groups over foreign-controlled groups.

Fixing the unfair tax advantage is not protectionist because it does not favor domestic companies over foreign competitors. The fix merely would level the playing field by similarly taxing U.S.-owned insurers and their foreign-based competitors in writing U.S. business.

We do not seek special treatment in accessing foreign markets relative to our foreign competitors. Conversely, foreign-based competitors should not be advantaged in the U.S. market relative to U.S. insurers in writing U.S. business under the tax code. Thus, legislation that removes any such advantage and is designed to achieve parity in treatment should not be viewed as protectionist.

Also, it is important to note that the proposed legislation does not distinguish between insurance companies controlled by U.S. groups and insurance companies controlled by foreign groups. The bill would apply to any reinsurance premium paid by a "covered insurance company" if the premium is neither subpart F income nor subject to tax in the U.S. Thus, a U.S. controlled group could be subject to the bill if the reinsurance premium will not be treated as subpart F income and will otherwise avoid U.S. tax.

Similar to section 163(j), the legislation is intended to address stripping of capital and earnings out of the U.S. tax base to avoid U.S. tax. The exception provided under the bill recognizes that reinsurance is not tax-motivated if the premiums are taxed under subpart F or are otherwise subject to tax in the United States. Thus, as with section 163(j), targeted legislation to attack excessive stripping of income outside the United States should not be viewed as protectionist.

10. Contrary to opponents' claims, the bill properly accounts for the benefits of taxing in-bound reinsurance transactions.

The proposed legislation recognizes that the United States benefits by taxing inbound reinsurance transactions. The proposed legislation thus permits the netting of in-bound premiums against out-bound premiums prior to applying the limitation.

11. Contrary to opponents' claims, the proposed legislation addresses similar policy concerns and adopts a similar approach to the earnings stripping rules of section 163(j).

In a 2007 hearing pamphlet, the staff of the Joint Committee on Taxation stated:

The issuance of debt and the reinsurance of risks effectively transfer both capital and risk in exchange for a right to a return for the lender or reinsurer.

A foreign parent may capitalize its U.S. subsidiary with debt or equity. Debt capital provides a tax benefit that equity capital does not provide. Therefore, Congress enacted section 163(j) to limit the stripping of earnings through debt capital.

The proposed legislation addresses the same concern in the context of the insurance industry. A foreign insurer may capitalize its U.S. subsidiary with equity or, alternatively, it may reinsure the policies written by its U.S. subsidiary.

Reinsurance presents significant tax advantages over equity capital: (i) a deduction for the reinsurance premiums, (ii) the accumulation of investment income free from U.S. tax, (iii) avoidance of the discounting of loss reserve rules, and (iv) the avoidance of U.S. tax on the repatriation of the stripped earnings. Therefore, similar to section 163(j), the proposed legislation is necessary to limit the stripping of earnings.

The legislative history of section 163(j) refers to "the difficulties encountered in distinguishing debt from equity." The proposed legislation similarly endorses the difficulties encountered in distinguishing bona fide reinsurance from tax-motivated reinsurance or from transfers of equity. Moreover, section 163(j) limits "the ability to 'strip' earnings out of this country through interest payments in lieu of dividend distributions." The proposed legislation similarly limits the ability to strip earnings reinsurance premiums in lieu of dividend distributions.

12. The market for municipal bonds could be significantly hampered if the insurance business continues to migrate offshore to take advantage of this unintended loophole.

Closing the affiliate reinsurance loophole is important to help maintain the current market for state and local bonds.

Domestic property and casualty insurers are major investors in tax-exempt bonds, providing financing for vital state and local needs. According to the Built by Bonds coalition, "P&Cs are the single largest corporate investors in municipal bonds, holding 12% of bonds outstanding." This is nearly half of all corporate investment. Much of that investment is in long-term debt that funds investment in critical projects, such as schools, transportation, water and sewer projects, and other basic infrastructure.

Yet, not surprisingly, there is a significant disparity between the levels of investment in tax-exempt bonds by domestic insurers and foreign-controlled groups that take advantage of the affiliate reinsurance loophole. After all, there is little need for tax-exempt interest when a company is able to strip investment income overseas to a low-tax or no-tax jurisdiction and avoid paying U.S. tax.

Thus, unless the loophole is closed to prevent investment income from being stripped offshore, the market for tax-exempt bonds is likely to diminish as the use of affiliate reinsurance grows. Reducing such a large source of demand is likely to cause the cost of borrowing for state and local governments to increase significantly. Conversely, closing the loophole could significantly improve the credit market and reduce borrowing costs for state and local governments.

We commend you and your staff for your efforts to close this loophole and eliminate the unfair competitive advantage for foreign-controlled insurers. Passage of this bill will help restore competitive balance to the marketplace and prevent the costly migration of the domestic P&C insurance industry, as well as the attendant U.S. tax base. We are hopeful that legislation to close this loophole will be adopted this year.

² Editorial, The Baltimore Sun, May 15, 2000.

³ See, e.g., General Explanation of Tax Legislation Enacted in the 108^th Congress, Staff of the Joint Comm. On Taxation, p. 351 (May 2005).

⁴ See comment letter submitted by Association of Bermuda Insurers and Reinsurers, p. 15 (February 25, 2009).

⁵ The primary insurance transaction requires a lot of people and work, whereas the reinsurance transaction between affiliated parties requires at most a few people to accept the terms and reflect it on the books. With a common parent and shared interests, there is little if any need to review the work of the underlying primary underwriters. Since they are relying on the work of the underlying primary underwriters, the business moves with no consequential work involved.

⁶ The Brattle Group, "The Impact on the U.S. Insurance Market of a Tax on Offshore Affiliate Reinsurance: An Economic Analysis." May 1, 2009.

⁷ Lawrence S. Powell and David W. Sommer, "Internal versus External Capital Markets in the Insurance Industry: The Role of Reinsurance," Journal of Financial Service Review, 2007, Vol. 31, pp. 180, 187 (According to the study's empirical analysis, "the coefficient estimate for Catastrophe Exposure is significant and positive in the external [i.e., third party] reinsurance equation. It is not significantly different from zero in the internal [i.e., related party] reinsurance equation. This evidence is consistent with internal and external reinsurance serving different purposes.")

⁸ See comment letter submitted by Association of Bermuda Insurers and Reinsurers, p. 11 (February 25, 2009).

⁹ ROE is a measure of profitability where after-tax earnings are expressed as a percentage of average equity outstanding at the beginning and end of the year that enables the comparison of firms of different sizes.

¹⁰ Testimony before the Senate Finance Committee (September 26, 2007)

¹¹ "It's not just the climate," Forbes, p. 42 (November 7, 1994).

¹² Dowling & Partners, IBNR Weekly #46, Vol. XIII, p. 3 (November 20, 2006).

¹³ Dowling & Partners, IBNR Weekly #15, Vol. XIV, p. 5 (April 13, 2007).

¹⁴ This illustration only considers federal income tax effects. If State income taxes are also considered, the breakeven profit level is even lower and there is greater incentive to reinsure to an affiliate overseas.

¹⁵ "Cayman and Bermuda Not the Only ACE Place," Risk & Insurance (June 1, 2008), www.riskandinsurance.com/story.jsp?storyId=97688610

¹⁶ E.g., Flagstone Re and Paris Re. See also "Reinsurers not neutral on Swiss advantages, Business Insurance, (September 1, 2008) (describing Flagstone Re's move and quoting A.M. Best senior financial analyst Marc Murray, "I wouldn't be surprised to see more companies set up operations in Switzerland.")

¹⁷ Hard and soft markets are a part of the "insurance market cycle" and are a function of supply and demand. In a hard market, prices increase to what the market will bear because demand exceeds supply. In a "soft market", insurance companies hoping to write new business and to hold onto existing business are likely to offer coverage improvements and/or reduced premiums.

¹⁸ "Don't Blame Bermuda for the Soft Market," National Underwriter Property & Casualty, p. 8 (November 26, 2007)

¹⁹ There are slight differences between the Model Treaty and the Swiss Treaty but they are immaterial for this purpose.

²⁰ In the Swiss Treaty, the phrase is "other or more burdensome."

²¹ See also Notice 87-66, 1987-2 C.B. 376.

²² See I.R.C. sec. 897(i).

²³ See also American Air Liquide, Inc. v. Commissioner, 116 T.C. 23 (2001) (denial of look-through treatment for royalties received from foreign parent not discriminatory because similar treatment applied to any other domestic corporation receiving royalty income from a non-controlled foreign corporation).

²⁴ Article XVII generally requires that a country not treat foreign service providers (located in another member country) less favorably than are domestic suppliers.

²⁵ The Brattle Group, "The Impact on the U.S. Insurance Market of a Tax on Offshore Affiliate Reinsurance: An Economic Analysis." May 1, 2009, p. 8-9.

²⁶ See Lawrence S. Powell and David W. Sommer, "Internal versus External Capital Markets in the Insurance Industry: The Role of Reinsurance," Journal of Financial Service Review, 2007, Vol. 31, p. 186.

²⁷ "Present Law and Analysis Relating to Selected International Tax Issues," Staff of the Joint Committee on Taxation, JCX-85-07 (September 24, 2007), p. 66.

²⁸ See id., pp. 61-62.