Scholarly article on topic 'Reforming the Taxation and Regulation of Mutual Funds: A Comparative Legal and Economic Analysis'

Reforming the Taxation and Regulation of Mutual Funds: A Comparative Legal and Economic Analysis Academic research paper on "Economics and business"

Share paper
Academic journal
Journal of Legal Analysis
OECD Field of science

Academic research paper on topic "Reforming the Taxation and Regulation of Mutual Funds: A Comparative Legal and Economic Analysis"


economic analysis

John C. Coates IV1


Most Americans invest through mutual funds. An analysis of laws governing mutual funds shows that U.S. mutual funds are taxed less favorably and regulated more extensively than direct investments or other collective investments, including alternatives available only to the wealthy. The structure of U.S. regulation—of 70-year old proscriptive bright-line rules subject to SEC exemptions—makes success of U.S. mutual funds dependent on the resources, responsiveness, and flexibility of the SEC. The legal framework for mutual funds in the E.U. is generally as or more restrictive and inflexible than U.S. law, but competitive pressures force European supervisors to be more flexible in adopting and implementing regulations, and E.U. regulators have greater resources and are more responsive than the SEC. The paper discusses mutual fund legal reforms, including proposals to eliminate unjustified disparities in the treatment of mutual funds and their substitutes and to improve regulatory flexibility and resources.

1. introduction

Over the past 50 years, mutual funds have become the primary way middle class Americans invest:

• 44% of U.S. households own fund shares, up from 6% in 1980, but down from its peak at 48% in 2001 (ICI 2008a, 70; ICI 2007b, 2).2

1 John F. Cogan, Jr. Professor of Law and Economics, Harvard Law School, 1525 Massachusetts Avenue, Griswold 400, Cambridge, MA 02138, Email: This paper benefited from comments and conversations with Sandy Bieber, Chris Christian, Jack Cogan, Buddy Donahue, Susan Ervin, Tamar Frankel, Ron Gilson, Dan Halperin, Terri Hyde, Howell Jackson, Louis Kaplow, Keither Lawson, Si Lorne, Martin Lybecker, Karrie McMillan, Liz Osterman, Gary Palmer, Bill Paul, Deb Pege, Bob Plaze, Bob Pozen, Mark Ram-seyer, Brian Reid, Eric Roiter, Hal Scott, Paul Stevens, Lynn Stout, Eric Talley, Peter Tufano, David Weisbach, and an anonymous referee. The paper also benefited from presentations and discussions at Harvard Law School, Northwestern Law School, and the Mutual Fund Directors Forum. I also thank Mike Doore, Brett Hartman, Ken Leung, Laura McIntyre, Joel Pulliam, and Jason Steinman for research assistance. The paper was requested by but has not yet been reviewed by the Committee on Capital Market Regulation. All errors are mine.

2 Mutual funds manage 23% of household financial assets, up from <3% in 1980 (ICI 2008a, 8).

Summer 2009: Volume 1, Number 2 ~ Journal of Legal Analysis ~ 591

• 33% of all families earning between $25,000 and $50,000 invest in mutual funds (ICI 2007b, 4), and half of households owning funds earn less than $75,000 (ICI 2007b, 1, 5).

• More than 25% of all retirement plan assets are held in mutual funds (ICI 2008a, 99).

With more than $9.6 trillion in assets (down from $12 trillion in 2007) (ICI 2008b; ICI 2008a, 7), U.S. mutual funds are an important channel for investment to flow through capital markets into new businesses around the world. In 2007, for the first time, U.S. mutual funds held more stock in o

U.S. companies than did either individuals or any other type of financial o

institution (Bd. of Governors of the Fed. Res. System 2008). Other forms of e

U.S. collective investments for individuals—private funds, common trusts, o

managed accounts, annuities, real estate investment trusts, and commod- g

ity pools—together held more than $9 trillion in assets in the U.S. in 2007 j

(U.S. Commodity Futures Trading Comm. 2000,.Fig. 3). 0

Most individuals cannot invest without advice, research, execution, and g

similar services from others. The U.S. regulates such services, depending on g

the investment asset and the number and nature of the services. The largest s

asset class for most U.S. investors is real estate (Campbell 2006, Fig. 2)—their /

homes—yet investments in individual homes receive large federal tax subsi- u

dies and are largely unregulated. Investments in securities, by contrast, are reg- V ulated through disclosure and anti-fraud laws enforced by the SEC and other

regulators and through supervision and regulation of the service providers C

and their services (such as broker-dealers, investment advisers, bond inden- g

ture trustees, securities research and analysis, rating agencies, and so forth). C ■

Most individuals cannot cost effectively invest in unique investment J

portfolios, so they invest collectively in standard ways, pooling invest- 7

ments to achieve economies of scale. Collective investment in securities is 0

(with limited exceptions) heavily regulated. Mutual funds, in particular, are 5

governed by several federal laws (Coates and Hubbard 2007; Frankel and Schwing 2001, 2004). The chief laws are

• The Investment Company Act of 1940 (ICA), the primary regulatory statute

• The Internal Revenue Code (IRC), which creates strong tax incentives for regulated investment companies to diversify and distribute earnings annually

Regulation and oversight of mutual funds in the U.S. has generally been good. Relative to other financial sectors, mutual funds have been tainted by few scandals—the hue and cry over late trading in 2001 (SEC 2003b; SEC 2004;

Choi and Kahan 2007) is the exception3 that proves the rule. Funds are generally governed with more oversight and better disclosure than are other collective investments. Fees charged by mutual funds compare favorably with those charged by competitors, including other U.S. collective investments and foreign mutual funds (Kritzman 2008; Khorana, Servaes, and Turano in press, Table 2). Regulatory costs for mutual funds compare favorably with those in some (but not all) developed countries. The rapid growth of exchange-traded funds shows U.S. regulation is not so stringent as to choke off innovation in the industry.

True, the current financial crisis has greatly eroded the value of mutual fund investments—but that has been true of every class of financial asset. Conventional funds marketed to retail investors had a relatively limited role in causing the current crisis and have performed remarkably well compared to other financial sectors, which have either vanished (such as investment banks) or been severely impaired (for instance, banks). To date, the only structural flaw exposed in the basic design of U.S. fund regulation was that money market mutual funds (MMMFs) were permitted to invest in what historically had been safe and liquid securities (such as commercial paper and short-term bonds) that turned out to have much greater liquidity risks than anyone had appreciated, leading the U.S. Treasury Department to create an emergency guaranty program for MMMFs. In retrospect, greater disclosure should have been required of MMMF sponsors that had maintained their share values only with the benefit of subsidies from affiliates, and greater oversight or regulatory reform for MMMFs might be required going forward. Still, even the few MMMFs that "broke the buck" and failed to maintain the traditional $1 price for their shares imposed relatively modest discounts (the largest such fund, the Reserve Primary Fund, repriced to $0.97) and the emergency guaranty program stemmed an incipient "run" on MMMFs despite being available only for assets in place as of September 19, 2008. MMMF assets actually increased in both September and October 2008 (ICI 2008b).

Thus, both historically and in the current crisis, U.S. fund regulation has proven relatively successful. Nevertheless, as in other sectors of U.S. capital

<¡1 t U

3 The scandals affected only a small portion of the U.S. fund industry. The SEC established 26 Fair Funds with $3.3 billion to compensate investors, 0.05% of fund assets in 2001. See SEC (2006a, 23), Coates and Hubbard (2007, Table 1), and ICI (2008a). That figure overstates the harm caused by mutual funds or their advisers. Many Fair Funds received payments from individual investors, brokers, and hedge funds, and in several cases were larger than any plausible loss to affected mutual funds. Rapoport says that losses to investors in auction-rate securities are estimated at $40 billion (2008). The out-of-pocket cost to U.S. taxpayers of the scandals was zero. The 1980-1988 cost of S&L failures to taxpayers is estimated at $160 billion (Nat'l Comm. On Fin. Inst. Reform, Recovery, and Enforcement 1993, 44, 79). The bailout of Fannie Mae and Freddie Mac is expected to cost taxpayers $25 billion (Herszenhorn 2008).

markets, continued U.S. competitiveness cannot be assured, and opportunities to enhance collective investment in the U.S. exist. Current U.S. tax law governing mutual funds, in particular, has a number of unfortunate economic effects, including the discouragement of saving and investment by middle class Americans, misallocation of capital to sectors such as real estate, and the essential walling-off of the U.S. from competition from or with foreign mutual funds in the U.S. or overseas. Part I describes current U.S. tax law and its negative effects. It suggests changes that would improve both the efficiency and fairness of the taxation of investments in U.S. mutual funds and enhance cross-border investment and competition.

In addition, although much less important than U.S. tax law, the current design and implementation of U.S. securities law applicable to mutual funds is inhibiting innovation and growth in the fund sector, again with the effect of reducing investment. Part II reviews U.S. regulation of collective investments, comparing regulation and data on the size and growth of U.S. mutual funds with their major competitors in the U.S. and abroad. The review shows that

• The U.S. fund industry continues to be the world leader, but its growth do

and international competitiveness now lags behind that of its domestic and g

foreign competitors, primarily because of U.S. tax law. 0

• Within the U.S., regulation of mutual funds is more extensive and restric- a tive than for other types of collective investments. C

• Mutual funds are tightly restricted by bright-line rules written into a statute r nearly 70 years ago and are subject to SEC exemptions. This structure of ti"

U.S. regulation makes continued success of U.S. mutual funds dependent G

on the resources, responsiveness, and flexibility of the SEC. °

• Although the high-level formal framework for mutual funds in the E.U. is 0 as or more restrictive and inflexible in most respects than the Investment u Company Act, competitive pressures in the E.U. force supervisors in E.U. 7 countries to be more flexible in adopting implementing regulations, and 0 E.U. regulators (particularly in Ireland and Luxembourg) have greater re- ^ sources and are more responsive than the SEC. The SEC could achieve the

same flexibility and responsiveness through exemptive orders but has been unwilling or unable to do so in a timely fashion.

Part III describes potential improvements in U.S. regulatory oversight of collective investments, including ways to enhance the flexibility and resources of U.S. fund regulators, modifications of the existing ban on asymmetric adviser compensation and the exclusion of foreign funds, and unjustified disparities in the treatment of mutual funds and mutual fund substitutes.

Together, the initiatives in Parts I and III should increase long-term in- 10 vestment and capital formation and reduce risk without reducing returns by expanding investor choice, increasing competition, and encouraging diversification.4

2. u.s. taxation of funds, its effects on competition, and potential changes

U.S. tax law has first-order importance for the competitiveness of the U.S. 11 fund industry. This section reviews U.S. tax law applicable to mutual funds, compares it to taxation of mutual fund substitutes, and describes potential changes in U.S. taxation of funds that could benefit U.S. investors and U.S.-based funds.

A. Current U.S. Taxation of Mutual Funds and Its Effects

U.S. tax law has four sets of effects on U.S. mutual funds investors: 12

1. U.S. tax law taxes U.S. mutual fund investors less favorably than U.S. investors who invest directly in securities or who invest in alternative collective investments, such as private funds, annuities, and common trusts.

2. It essentially prevents U.S. funds from selling to foreign investors.

3. It essentially prevents foreign funds from selling to U.S. investors.

4. It imposes crude, bright-line rule-of-thumb diversification tests that constrain fund investment flexibility and can force funds to liquidate positions, which, ironically, can increase the funds' taxable distributions.

Together, these tax effects undermine the competitiveness of U.S. mutual 13 funds relative to their domestic competitors; wall off U.S. investors from the potential benefits of competition between U.S. and foreign mutual funds, including efficient international diversification; and prevent U.S. mutual funds from competing on a level playing field with foreign funds for foreign investors' assets.

<¡1 t

Overview of IRC Provisions Applicable to Mutual Funds

U.S. mutual funds are structured as corporations or business trusts, which are 14 generally treated as corporations for tax purposes. However, unlike most corporations, a mutual fund is taxed only on the amount of earnings it retains.

4 For evidence that direct investments can reduce investor returns or increase risk, see Barber and Odean (2000); Odean (2000); Odean (1998); Dhar and Zhu (2006); Polkovnichenko (2005); Goetzmann and Kumar (in press).

To get this specialized "pass through" tax treatment, a mutual fund must meet requirements first established under the Revenue Act of 1936 that continue today under Subchapter M of the Internal Revenue Code.5 Subchapter M of the IRC permits mutual funds to avoid triple taxation (tax on portfolio companies, tax on the fund, and tax on fund shareholders) if they distribute their earnings to their shareholders, derive most of their income from investments, and fragment their holdings among at least 12 issuers.6

Specifically, to avoid fund-level tax for any taxable year, a fund must distribute to current shareholders all of its ordinary income and its exempt interest income, as well as all of its net long-term capital gains,7 as dividends on | an annual basis, and it must meet certain other conditions.8 In addition, to

qualify for flow-through tax status, a fund must meet these requirements: f

A. Derive 90+% of its income from dividends, interest, payments for securities g loans, and gains from specified securities transactions j

B. Maintain at the close of each quarter

1. 50+% of the value of its assets in cash, government securities, securities

of other regulated investment companies, or other securities, provided

that securities of any one issuer do not exceed 5% of the value of the .

fund's total assets or represent 10% of the voting shares of the issuer ^

2. Less than 25% of the value of its assets, other than government bonds U

or shares of other regulated funds, in shares of any one issuer, or in g

5 For the history of Subchapter M, see Roe (2004, 102-23) and Fink (2008, 26-29). Roe argues a G

principal reason for the adoption of Subchapter M was to prevent mutual funds from exerting o

control on behalf of investors over portfolio companies. era

6 IRC §§ 851 and 852. Meeting the requirements described in the text permits a fund to take a 3 special deduction from taxable income for all dividends it pays currently. IRC § 852(b)(2)-(9). gT This deduction, combined with special rules allowing funds to flow through the character of 7 certain income (for example, qualified dividends and exempt interest income) and foreign tax 2 credits, achieves a form of "flow-through" tax treatment for mutual funds. 5

7 To qualify for flow-through tax treatment, a fund must distribute 90+% of its ordinary taxable income. It is permitted to retain up to 10% of its ordinary taxable income without affecting its qualification, but the fund must pay tax at the regular corporate rates on the retained amount. The failure to distribute net long-term capital gains does not technically affect a fund's qualification for flow-through tax treatment, but the fund must pay tax on the retained capital gains at the regular corporate rates (generally 35%). Consequently, most mutual funds distribute all of their taxable ordinary income and net capital gains on an annual basis. If a fund does not qualify for flow-through tax treatment (for reasons discussed in the text), the fund would pay tax on all of its taxable ordinary income and capital gains for the year, not just its undistributed income.

8 In addition, a fund that schedules dividend distributions to defer taxation for a calendar year— such as by failing to distribute, with respect to any calendar year, at least 98% of its ordinary income and at least 98% of its capital gain net income (determined using a 12-month period ending October 31)—is subject to a 4% excise tax. IRC § 4982.

shares of 2 or more issuers controlled by the fund and engaged in related businesses

Effects of U.S. Tax Law on U.S. Investors Investing in U.S. Funds

Because mutual funds must essentially pass through all of their taxable profits to maintain favorable tax status, mutual fund investors receive distributions annually and thereby incur corresponding dividend and capital gains taxes. Individual fund investors are taxed on all of the dividends and capital gains realized by a fund each year, even if—as is typical for individual investors9—they automatically reinvest fund distributions in the same fund. In effect, mutual funds escape tax only by forcing their investors to pay capital gains taxes continually as capital gains are realized by the funds, rather than only on sale or other cash realization events with respect to their own shares, as with capital investments generally. The bottom line of these effects is substantial and the average mutual fund investor loses between 15% and 20% in annual returns as a result of current U.S. taxation.

Specifically, mutual fund investors pay tax on a taxable fund's distributed earnings, whether they receive distributions in cash or reinvest them in fund shares. Those distributions take two types—ordinary dividends10 and capital gains—but both types are now taxed at a top rate of 15%,11 although a lower rate applies to some taxpayers for long-term capital gains.12 Because a mutual fund will calculate its gains—which must be then distributed to fund investors—based on its cost basis in the portfolio shares, the capital gains taxes payable by fund investors need not bear any relationship to the performance of the investors' investments in the fund. As a result, it is not unusual for a mutual fund investor to both lose money during the course of a year and still owe capital gains taxes as a result of fund-level capital gains. The mismatch between investor realizations of cash and taxes payable on mutual fund investments can be significant and will generate additional

9 More than 90% of dividends paid by equity mutual funds in 2007 were reinvested (ICI 2008a, Table 28).

10 Dividend distributions come primarily from interest and dividends earned by securities in a fund's portfolio and net short-term gains after expenses are paid by the fund. Dividend distributions are reported as dividends on an investor's tax return.

11 This top rate was enacted in 2003 and was extended through 2010 in 2006. Technically, only qualified dividends are taxed at the 15% rate; non-qualified dividends and short-term capital gains are taxed at ordinary income rates.

12 Long-term capital gain distributions are from a fund's net gains from the sale of securities held for more than one year.

transaction costs not typically borne by investors who invest through individual portfolios.

An Example Here is a simple example.13 Assume an investor has invested all of her savings in 100 shares of a mutual fund, priced at $10 per fund share, for a total of $1,000. Assume that over the course of a year, the fund experiences net redemptions—that is, more of its shareholders sell than buy fund shares, perhaps because existing investors redeem shares to fund their retirements and the overall economy has become riskier so that investors not yet in retirement are worried about investing more until the state of the economy has become more clear. As a result, the fund sells shares of 3

its portfolio companies to generate the cash required by the net redemptions. In so doing, the fund realizes capital gains equal to 10% of the value of its portfolio. U.S. tax law requires the fund to distribute those capital 3

gains to the fund's shareholders, and those gains are treated as short-term gains to the extent the portfolio shares had been held for less than one year, .

even if the fund investor has held her fund shares for more than one year. |

As a result of the fund distribution, the investor receives 10% of $1,000, O

or $100, but because the investor is not in retirement, does not otherwise 3

need the cash, and does not want to reduce her investments, she simply C

reinvests the $100 in the same fund. The fund's share price falls by 10%, t

reflecting the value of the distribution, from $10 to $9, so the reinvested i

$100 buys the investor 11.11 additional shares. Thus, the customer's account now holds 111.11 shares worth $9, or $1,000 (after rounding). The investor's account value is precisely the same as before the distribution, as it should be. However, the capital gain passed along to the fund investor triggers tax for the investor: If the gains are short-term, the investor must pay 35% x $100 = $35. The investor will need to use other cash—which otherwise could have been invested—or redeem fund shares to pay the tax liability, reducing the value of the account to $965. If the fund's shares have risen since the investor bought the fund, that redemption will generate additional capital gains tax. Thus, on an after-tax basis, the investor is obviously worse off than if the fund has not experienced net redemptions and had been able to hold on to its existing portfolio of investments.

Unpleasant Surprises Caused by Fund Taxation Current mutual fund tax rules can create unpleasant surprises for new fund investors. Investors who buy fund shares shortly before a tax law-mandated distribution are effectively treated less well than investors who buy shortly after a distribution.

13 The example is adapted from Lipper (2008, 7).

Distribution of capital gains, even by the same fund, can vary significantly over time, so that even investors who are generally aware of how funds are taxed can be surprised by unexpectedly large taxable distributions. Funds that are older, that are managed to minimize taxable distributions, and/ or have investment strategies calling for less portfolio turnover can accumulate large unrealized capital gains in their portfolios—often called tax overhang. New investors, in effect, buy those potential future taxes when they buy fund shares. Although many investors—particularly the most sophisticated ones—are aware of this tax effect and take it into account when choosing among funds, many investors—particularly those investing for the first time and those with less wealth and education—might not be aware of this tax effect and in effect pay more for a given mutual fund than they would if they were well informed.

An Example of the Effect of Mutual Fund Tax Overhang A simple example of the effect of tax overhang is as follows.14

At time 1, fund XYZ sells 10 shares of stock to three investors, X, Y, and Z, at a price of $1/share. XYZ uses the $30 of cash proceeds to buy three shares of ABC stock for $10/share.

At time 2, ABC stock rises to $30/share, so fund XYZ's share price is $3/ share—this represents an as-yet unrealized capital gain for the fund. A new investor, W, buys 10 new fund XYZ shares at $3/share, and the fund uses the $30 to buy one more share of ABC stock.

XYZ's acquisition costs for its ABC shares are $10, $10, $10, and $30. W in effect "buys" some of the unrealized capital gain when he buys his XYZ fund shares.

At time 3, with the price of XYZ still at $3/share, X and Y redeem their XYZ shares. To meet this redemption, the fund sells shares of ABC that it purchased at $30 and $10 for proceeds of $60 (2 shares x $30/share price). XYZ realizes a capital gain of $20 ($60 proceeds less $40 total cost), which it distributes to the two remaining shareholders, W and Z. Each receives a distribution of $10. 15

In effect, W has had to pay for the realization of gains on ABC shares, despite the facts that XYZ's price has not risen since he bought into the fund and that he has yet to sell any fund shares. As a result of the redemption

14 The example is adapted from Barclay, Pearson, and Weisbach (1998, Fig. 1).

15 The careful reader might wonder where XYZ gets the cash to make this distribution. Because W and Z have precommitted to reinvest all distributions, XYZ knows that it will not actually need cash to make this distribution. In practice, funds maintain cash reserves to make distributions, fund net redemptions, and pay advisory fees.

and distribution, XYZ's share price drops to $2/share ($120 value of ABC shares, less $60 used to redeem X and Y, less $20 distributed to W and Z, divided by W and Z's remaining 10 shares each). Investors W and Z reinvest their distributions, each buying 5 shares each, at $2/share, and pay taxes of 35% X $10 = $3.50. W's total cost of his XYZ shares is $30 + $10 = $40.

At time 4, the price of ABC rises to $40/share, and XYZ's share price rises proportionately, to $4/share (2 ABC shares x $40 value divided by XYZ's 20 outstanding shares). Investor W redeems his 10 shares, receiving $40, and pays no taxes because his proceeds do not exceed the cost of his XYZ shares.

The effect on W is markedly worse than had he purchased his XYZ shares n

at the same time as X, Y and Z. He's effectively had to share in the tax cost of fund activity that both predated and was not triggered by his investment. Although he does not have to pay taxes twice—he gets a credit for his g purchase of embedded capital gains (for example, unrealized appreciation) when he sells his fund shares—he paid tax much earlier than he would have . otherwise done. Time 4, in this hypothetical example, could be (and often | is) many years later than time 3. o

Additional Ways Mutual Fund Investors Are Taxed Unfavorably However, 3

matters are even worse for mutual fund investors because mutual funds are c

not permitted to pass through net operating losses and net capital losses16 t

and individual investors have no control over the way or the extent to which 3

a mutual fund recognizes gains or losses. This asymmetric treatment of capital gains and losses increases the effective tax on stocks held by mutual funds (Mintz and Smart 2002). Most mutual fund investors purchase at least a portion of their shares as a result of automatic reinvestments of annual fund distributions, so the individual accounting and record-keeping burden to calculate and report capital gains and losses triggered by the sale of mutual fund shares is substantial.17

16 In determining the amount of current year income that must be distributed by the fund, however, the fund's operating expenses offset ordinary income and capital losses offset capital gains. In addition, a fund's net capital loss for a year can be carried forward under IRC § 1212(a)(1)(C) to offset capital gains realized in succeeding years. A new operating loss, by contrast, cannot be carried forward to reduce the amount of ordinary income distributions in succeeding years.

17 Many funds provide cost basis information to shareholders or compute gains and losses for shares sold, and bills have been introduced in Congress in recent years to require cost-basis reporting. One difficulty with mandating full cost-basis reporting is that investors have options in how they report redemptions (such as average cost basis or last-in first-out) and can engage in transactions such as wash sales that might affect their tax basis in a fund's shares. Advisors do not observe these choices or transactions, so they cannot know investors' true cost bases.

Actual Taxes Triggered for Mutual Fund Investors Taxes triggered by existing mutual fund tax law are substantial. In the late 1990s, capital gains taxes reduced mutual fund performance by a median of 2.5 percentage points per year, accumulating to more than 250% of an initial investment over a 20-year period (KPMG Peat Marwick LLP 2000). Even in a year such as 2000, when the S&P 500 declined by 9%, mutual fund investors received more than $326 billion in capital gains distributions (ICI 2008a, Fig. 2.5). Ironically, they were forced to pay gains taxes even as their fund account balances fell.

Taxable short- and long-term capital gains distributions from mutual funds have been rising steadily since 2002, as reflected on Fig. 1. The principal reason for the increase in gains distributions is the gradual exhaustion or expiration of fund-level capital losses generated during prior years, particularly the market downturn of 2000-2002. Ironically, the silver lining to the current difficulties in the stock market is that it will provide well-managed funds the opportunity to harvest tax losses and so again reduce gains distributions over the next few years.

Nevertheless, net capital gains resulted in an estimated $22 billion in taxes for long-term, buy-and-hold mutual fund investors in 2007, plus an additional $14 billion in taxes on dividend and interest distributions (Lipper 2008, 1). Nearly all of these taxes have been generated by equity funds (Lipper

<¡1 t U

Figure 1. Estimated Potentially Taxable Distributions from Mutual Funds

I Short-Term Capital Gain I Long-Term Capital Gains Total

2002 2003 2004 2005 2006 2007

Source: Lipper, Taxes in the Mutual Fund Industry (2008).

2008, 10, Tables 2-4). For assets in taxable accounts, "taxes eat up ... at least as much of the gross return [of mutual fUnd investments] as expenses..." (Lipper 2008, 19). Lower returns, of course, result in less investment.

Not all of the gains taxes are attributable to the externalities imposed on buy-and-hold fund investors by other fund investors. Some of the gains arise from normal fund management in actively traded funds, and some arise from issuer-related taxable events, such as cash mergers. However, portfolio turnover—and thus capital gains taxes triggered by turnover— that is attributable to net redemption activity represents a significant share of total portfolio turnover at funds generally. During the last market down- |

turn, investors redeemed more than $13 trillion of mutual fund shares (ICI 2005, 60, Table 2),18 including roughly $1 trillion of redemptions from long-term equity funds (ICI 2005, 83, Table 25) producing $27 billion of g

net outflow of cash from equity funds (ICI 2006, 89, Table 19). As a percentage of average fund assets, shareholder redemptions ranged from 25% . (2004) to 39% (2002) to 57% (1987) for all funds, and from 23% (2005) to | 41% (2002) to 73% (1987) for equity funds (ICI 2006, 97, Table 27). Since | these are annual rates, total redemptions over a multi-year period would | be higher. These are also average rates, so particular funds would have ex- r perienced higher redemption rates even within a single year. Many funds, t in fact, experience net outflows: Barclay et al. found that 25% of a sample i of fund years from 1976 to 1992 experienced net outflows of at least 14% (1998). In 1999 and 2000, nearly half of all mutual fund complexes saw net cash outflows from their long-term funds (Reid 2006, 2, Table 1). In the current crisis year, it is likely that most non-MMMF funds will experience net outflows and that many funds will distribute taxable long-term capital gains despite experiencing record losses.

Taxation of Mutual Fund Substitutes for Wealthy Investors Mutual funds are taxed less favorably than either near substitutes or direct portfolio investments. Substitutes for mutual funds include exchange-traded notes (ETNs), separately managed accounts, variable annuities, commodity pools, hedge funds, and other private funds, common trusts, and exchange-traded funds (ETFs). Although these are taxed somewhat differently from one another, they all enjoy tax advantages over mutual funds (see Table 1).

18 This figure excludes exchange redemptions, which are redemptions followed by an immediate reinvestment in a fund within the same fund complex. Thus, it understates the effect on turnover of total redemptions.

Table 1. Overview of Collective Investments, Regulatory, and Tax Treatment

Type of Collective Investment Issuer/ Sponsor Form of Investor's Investment Example of Investment Assets Tax Treatment

Deferral at the Discretion of the Investor Possible? Tax Externalities? Loss Recognition?

ETN Bank Note issued by bank Note with payoffs linked to S&P500 Yes No Yes

SMA Broker/ dealer Account established and supervised by B/D S&P 500 Yes, although portfolio investments may self-liquidate (such as a bankruptcy or merger) No Yes

Variable annuity Insurance company Separate account within insurance company S&P 500 Yes, although are distributions taxed at ordinary income rates No No

Commodity pool or ETF Commodity pool operator Interest in partnership or LLC Futures on S&P 500 No No Yes

Hedge fund Advisory firm Interest in partnership or LLC S&P 500 No No Yes

Common trust Bank Interest in common trust S&P 500 No No Yes

Mutual fund Advisory firm Shares in business trust or corporation acquired from fund S&P 500 No Yes No

ETF Advisory firm Shares in business trust or corporation acquired on exchange from intermediary S&P 500 No Yes, although less severe than for traditional mutual funds No

Specifically—and as discussed more later—all of the alternatives permit deferral in whole or in part at the discretion of the investor or permit the investor to recognize losses as well as gains, or both. None impose tax externalities—that is, trigger tax for one investor based on the behavior of other investors. Although some of these collective investments (such as variable annuities) are marketed to middle-class investors, many are not. Except for tax advantages, many incorporate features that would make them less-than-ideal mutual fund substitutes for middle-class investors.

An ETN is a note issued by a commercial or investment bank with a return linked to a specified index, ranging from stocks or commodities to n

currencies or real estate, and is typically traded on an exchange (such as the American Stock Exchange). ETNs have generally been treated as prepaid forward contracts under U.S. tax law,19 which means that an investor § realizes no income or gain until the ETN is sold or matures. In contrast, as discussed earlier, investors in mutual funds—even if they are invested in . the same index as an ETN—are taxed annually on gains or income gener- | ated by the investments in that index. Moreover, ETN investors' taxation o is unaffected by the decision of other ETN investors—ETN issuers do not | need to engage in tax-triggering sales and distributions when ETN inves- r tors sell their notes on an exchange. Although ETNs are publicly sold, ETNs t have one fundamental feature that make them inadvisable for middle-class i investors: They represent direct debt claims on the issuer, exposing the in- g vestor to significant and undiversified credit risk, even if the underlying o index is diversified and turns out to perform well. Lehman Brothers issued e

ETNs before it went bankrupt, as did Bear Stearns before being acquired by <§

JP Morgan. §

Similarly, individual investors who invest directly or through separately e

managed accounts (SMAs), which are taxed as individual investment port- 2

folios, are generally able to achieve better after-tax returns with the same in- 5

vestments. Direct investors do not need to worry about redemption activity of other investors, as do fund investors. A direct investor will generally pay capital gains tax only if the investor decides to liquidate a position, or if a portfolio investment is liquidated by the issuer (as in a cash merger or liquidation), and those events also trigger capital gains tax for the mutual fund investor. Second, direct investors who leave their investments to their heirs

19 See IRS Rev. Rul. 2008-1, 2008-2 I.R.B. 1; Notice 2008-2, 2008-2 I.R.B. 2; see also Humphreys, Merali, and Reigersman (2007).

permanently avoid capital gains tax, since the death of an investor generally triggers forgiveness of any embedded capital gains tax, whereas mutual fund investors obtain this benefit only with respect to their shares in the mutual fund. Third, direct investors have complete control over when to realize capital losses, and can coordinate gain and loss recognition to maximize tax efficiency.

Variable annuities20 (including the most recent version, equity index annuities21) can function identically to mutual funds and are sometimes invested through mutual funds (SEC 2007b). Investors who invest through variable annuities are able to defer recognition of capital gains for the life of the investment. After investors start to receive annuity payouts, deferred gains are generally taxed at ordinary income rates. Depending on the relationship between ordinary income and capital gains rates, the age of the investor, the expected return of the investor, and the expected income bracket of the investor over the life of annuity, this combination of tax differences can make annuities more or less valuable to investors than mutual funds investing in equivalent investments. As with ETNs, variable annuities are marketed to middle-class investors. However, they are generally more complex and less transparent than mutual funds, most charge surrender fees (and thus are less liquid than mutual funds), and generally charge higher expenses than equivalent mutual fund investments,22 in part because they also often include an insurance component (the annuitization of the distributions adds administrative expense) that such investors might neither want nor need.

In contrast to mutual funds, a commodity pool or private fund (such as a hedge fund) in the U.S. is typically structured as a true "pass-through entity," such as a limited partnership or limited liability company, to avoid fund-level tax. Similarly, common trusts are taxed as true pass-through vehicles. Such forms of collective investments pay no taxes and do not have to distribute earnings to their investors. Instead, their investors generally

<¡1 t U

20 This discussion covers non-qualified annuities, governed by IRC § 72.

21 See SEC (2008d), a press release announcing the SEC's adoption of rules governing equity index annuities; Nat'l Arch. and Rec. Admin (2008), a SEC release proposing rules governing equity index annuities.

22 "As of Nov. 30, 2002, the average fees for variable annuity products covered by Morningstar was about 2.22 percent per year, whereas the average fee for separate accounts in the size range of our test was about 1.50 percent" (McAfee 2003).

pay tax on their pro rata portions of fund income or gains.23 To qualify for pass-through taxation, a collective investment cannot be registered under the ICA24 or, with some exceptions,25 publicly traded and registered under the 1934 Act.26 Private funds can achieve tax benefits for their investors by allocating tax attributes differently to different investors, provided the economics match the tax allocations, and can be established "off-shore" in a 27master-feeder structure, with one feeder fund set up for foreign or tax-exempt investors and another for domestic taxpayers, which allows optimization of tax outcomes across investors (Hammer et al. 2006, 106; Managed Fund Ass'n 2004). Even though both mutual funds and private funds are nominally "pass-through" vehicles, private funds typically generate lower amounts of taxable gains.

Private funds and commodity pools also currently have one additional tax advantage over mutual funds: A substantial portion of the income of fund managers—termed the "carried interest"—is afforded capital gains treatment for tax purposes, typically taxed at a lower rate (typically 15%) than the ordinary income rate (up to 35%), such as an investment adviser

23 To be fair and clear, partnership investors do pay tax as a result of partnership-level gains, even if they have not sold their partnership interests, just as with mutual fund investors. However, partnerships also pass through net losses and have additional tax flexibility not available to mutual funds. For instance, they can avoid the problem of tax overhang through reverse allocations under IRC § 704(c), and they make an election under IRC § 754 to step up (or down) the "inside basis" of assets in the event of sales or redemptions of partnership interests. Not all partnerships use this flexibility in practice, in part because it adds complexity and costs. It is simply unavailable to mutual funds, and the complexity and costs associated with simply extending these options to mutual funds would likely be cost-prohibitive in the standard, widely held mutual funds through which middle class investors typically invest.

24 26 U.S.C. §§ 851-52 (2006).

25 In general terms, publicly traded partnerships will not be taxed as corporations if 90+% of their gross income in every taxable year is qualifying income, as defined in the IRC, which generally includes dividends, interest, capital gains from the sale or other disposition of stocks and securities, and income derived from energy, real estate, and commodity investments (26 U.S.C. § 7704[c]). Recent high-profile public offerings by Blackstone and Fortress used innovative techniques to transform active income, such as management fees, into passive income, such as dividends, by channeling the income through blocker corporations, which borrowed large amounts of money from affiliated companies to generate tax-deductible interest payments sufficient to eliminate their own corporate tax (Beck 2007).

26 26 U.S.C. 7704(a) (2007).

27 To qualify for reverse allocations under IRC § 704(c), the partnership cannot specially allocate tax attributes other than management fees or "carries." See 1.703-3(e)(3)(iii)(A).

would pay on an advisory fee paid by a mutual fund.28 A mutual fund must pay what is effectively a higher after-tax rate for the same advisory services than would such entities. Private funds and commodity pools are able to produce better all-in after-tax returns than do mutual funds for the same investments.

Most investors in mutual funds have less than $75,000 in annual income (ICI 2007b, 1, 5); do not have access to private funds, commodity pools, or common trusts (SEC 2003a; Zhu 2005, Table 1); and do not have sufficient assets to cost-effectively diversify in an individual portfolio or meet eligibility requirements for a cost-effective SMA (Polkovnichenko 2005, Table 1). Although some ETNs and variable annuities are marketed to retail investors, many are not, and in any event include attributes (credit risk, opacity and complexity, and high fees) that would make them less attractive to middle-class investors. In short, our laws tax the diversified investments of most middle-class American investors more heavily than they do investments of the very wealthy.29

Taxation of ETFs

One sign of the significance of mutual fund taxation is the rapid growth of registered ETFs, discussed in Part II, which have long been promoted to individual investors based on their tax advantages (Gastineau 2002). The entire ETF industry can be viewed in large part as a form of attempted— but not yet clearly successful (Culloton 2006)—tax arbitrage. Among other things, ETFs can keep investor taxes lower because they only redeem shares in large blocks (with investors obtaining liquidity via exchange

<¡1 t U

28 For a discussion, see Sanchirico (2008). It is true that private fund advisers also charge advisory fees, in addition to receiving the carried interest, and private fund advisory fees are taxed less favorably than equivalent fees charged by mutual funds. This happens because mutual funds can deduct advisory fees at the fund level, whereas private funds cannot, and private investors can only deduct advisory fees under IRC § 212, subject to a limits based on adjusted gross income and the alternative minimum tax. Nevertheless, economically, tax-advantaged carried interest generates returns for private fund advisers, and those returns are fungible with advisory fees. For securities law reasons, discussed in Part II, mutual funds cannot provide mutual fund advisers with an equivalent "carried interest" as private funds.

29 Of course, the income tax generally is progressive, applying higher tax rates to higher-income taxpayers, and in theory a change in the progressivity of the income tax could compensate for the distributional consequences described in the text. However, progressivity is not generally thought to compensate for regressivity in the way in which investments are taxed, and because tax rates (and progressivity) are far more salient as political issues than the differences between mutual fund taxation and taxation of private funds, individual portfolios, and SMAs, the former is unlikely to compensate for the latter differences in practice.

transactions), so they face less risk of needing to liquidate portfolio securities to meet cash requirements generated by net redemptions. They can also use in-kind redemptions more cost effectively to meet cash requirements generated by net redemptions—essentially swaps of ETF shares for blocks of portfolio shares (Culloton 2006). Despite their potential tax advantages, relative to traditional mutual funds, ETFs must also distribute gains just like traditional mutual funds. Thus, they are taxed less favorably than private funds, direct portfolios, and SMAs and have not yet shown significant overall tax benefits compared to traditional mutual funds. The creation of ETFs generated significant up-front regulatory and other transaction costs, n

and the transition of investors from index funds to passively managed ETFs generated yet more transaction costs. To the (large) extent that registered passive ETFs are chosen for their tax advantages, all of those up-front and g

transition costs represented a dead-weight social loss induced by current U.S. taxation of mutual funds. .

Tax-Deferred Accounts It is true that middle-class Americans can invest f

on a tax-advantaged basis in mutual funds by establishing individual re- 0

tirement accounts (IRAs); through employer-sponsored 401k, 403b, or 3

similar plans; and through college-savings ("529") plans. Collectively, c

these are called tax deferred accounts (TDAs). TDAs have been attract- t

ing a growing share of mutual fund assets. Roughly half of mutual fund 3

holdings are held in TDAs, and roughly half are held in taxable accounts (ICI 2008a). But TDAs do not provide the equivalent tax advantages of private funds and individual portfolios,30 because investors generally pay income taxes at ordinary rates when they withdraw investments from these plans, which can make them unattractive relative even to taxable investments in mutual funds (Poterba 2004).31 Assets held in TDAs are

30 In addition to the points made in the text, TDAs do not provide the flexibility provided by private funds and individual portfolios. The tax advantages are effectively limited to funds saved for retirement or education. They are limited in amounts; and they impose early withdrawal penalties that make the investments in such plans less liquid than they otherwise would be. This constrains the ability of investors to tap such investments for certain emergencies or basic needs, such as divorce, property losses, certain job losses (such as those triggered by family needs that result in no unemployment benefits), or health costs (up to 7.5% of adjusted gross income).

31 Roth IRAs, which permit tax-free build-up of investments, are the exception to this set of findings, but withdrawals from such plans are permitted only after age 59-1/2, along with limited withdrawals for home purchases. Such TDAs are not available for those earning above $114,000 or for individuals filing singly. Individuals may contribute no more than $6,000 to a Roth IRA per year.

generally required to be withdrawn, triggering capital gains taxes. Direct portfolio investments, or investments held in SMAs or other collective investments, by contrast, can be retained until the investor's death. Heirs have the option to reset capital gains and losses at death, which in essence gives the investor's heirs the ability to avoid tax on any embedded capital gains at the time of death and reset the tax basis (its "cost" for tax purposes) at prevailing market prices at the time of death. More generally, TDAs may hold a range of investment assets, and are not limited to mutual funds, so their tax benefits are spread across a range of types of

investment channels. w

Negative Economic Effects of Mutual Fund Taxation U.S. tax law's treat- 42 g

ment of mutual funds has many unfortunate economic effects, as we have f

seen: g

• It discourages saving and capital formation by a large portion of the Ameri- j can population by taxing the most cost-effective way to invest.

• It discourages diversification, which is most cost-effectively achieved via collective investments, by taxing direct investments less heavily.

• It discourages investment in the safest, most heavily regulated equity channel for savings to be invested in the capital markets (such as mutual funds) / by taxing mutual funds more heavily than bank common trusts and SMAs, U which tend to impose higher fees than mutual funds. e

• It encourages direct investments, which are likely to be mismanaged by unsophisticated investors. 8,

• It exacerbates growing wealth disparities by taxing mutual funds more §

heavily than collective investments available to wealthy investors. o '

• It distorts the allocation of capital by taxing mutual funds more heavily than J

it does direct investments in real estate or passive collective investments in 7

commodities, oil, and gas, which can be invested in through commodity 0

pools, commodity ETFs exempt from the ICA, or energy funds organized as publicly traded partnerships.32

• It generates dead-weight tax-avoidance costs as fund sponsors, advisers, and investors expend real resources to shift funds from taxable mutual funds to tax-advantaged substitutes.

32 See IRC § 7704 (taxation of publicly traded master limited partnerships).

43 In addition, current tax treatment of U.S. mutual fund investors creates socially wasteful incentives for mutual fund advisers:

• Funds incur transaction costs to manage their tax positions by selling investments to realize losses to offset gains, for example. The effects can be significant for

° Older funds, which have held some stock positions for longer, thus

building up greater unrealized capital gains. ° Funds (both active and passive) that mimic an index or benchmark in which firms are frequently changing, such as funds that invest in small- 0

cap, mid-cap, and foreign companies, as well as funds that focus on "spe- d

cial situations" (such as mergers or troubled companies). f

° Funds with significant turnover among investors, such as sector funds in g

a sector downturn, in which investors harvest fund-level losses or real- p

locate their investments.

• Mutual fund taxation creates "tax overhang," which is built-up unrealized gains that will be borne when realized in the future by any investor who buys into the fund. This effect might disproportionately affect first-time and other unsophisticated investors (the best fund disclosure notwithstand- /

ing). It also distorts competition among funds by driving new investors to U

funds whose managers are skilled at keeping tax overhang in check, creat- e'

ing another difficult-to-assess dimension for investors to evaluate, making

comparison shopping harder. 8,

Effects of U.S. Tax Law on U.S. Investors in Foreign Funds 8'

U.S. investors are also tax-disadvantaged if they invest in foreign funds, U

which are generally treated as passive foreign investment companies 7

(PFICs).33 U.S. investors in PFICs must pay U.S. tax on any current distri- 0

butions from the PFIC. In addition, if (as is typical) the foreign fund either does not make distributions or does not comply with U.S. tax reporting requirements, U.S. investors must pay taxes when they sell their fund shares, equal to the tax they would have paid on fund distributions, at the highest rate for ordinary income in each prior year, plus an interest charge on the

33 See IRC §§ 1291-1297. A foreign corporation is a PFIC if 75+% or more of its gross income is passive or if 50+% of its assets "produce or are held to produce passive income," a definition that applies to most foreign funds.

deferred distribution.34 As a result, U.S. investors are generally in a better position if they invest directly in securities of foreign corporations that are not PFICs or invest in U.S. mutual funds that invest in foreign stocks or foreign funds.

The effects of U.S. taxation of foreign funds reinforce the SEC's general refusal to exempt foreign funds from the ICA, described in Part II, and effectively wall off U.S. investors from foreign funds. Both laws reinforce a general tendency of investors to invest close to home, both in terms of assets and methods or channels of investment. Khorana et al.'s study of mutual funds around the world shows that domestic investors invested in foreign funds in 80% of the 20 countries studied—but not U.S. investors (2005, 34, Table 1). Although the U.S. fund industry has many funds that specialize in foreign stocks, they do not offer the same breadth of choice (of either funds or portfolios) available to investors who live in those countries. For example, relatively few U.S.-domiciled funds or ETFs are tracking Japanese securities indexes relative to the number of such funds domiciled in other countries, and the median all-in fees (including amortized loads) of U.S.-domiciled Japan index funds exceeded the fees for the non-U.S.-domiciled Japanese index funds (O'Mary 2007, 29).

Effects of U.S. Tax Law on Foreign Investors in U.S. Funds

In addition to the inequitable and inefficient effects of current U.S. federal taxation of U.S. mutual fund investors, U.S. tax law effectively curtails the capability of U.S. mutual funds to sell shares to foreign investors. Foreign tax law generally does not tax (or require withholding) for distributions from foreign funds until investors redeem fund shares (Partsch, Terblanche, and Malaniuk, 2008; McGeough and Quirke 1992, 39; Jackson and Counihan 2008; "Managing Money" 1990; Gavin 1994). This allows foreign investors in foreign funds to enjoy tax-free asset growth and defer taxes until the fund shares are sold.

<¡1 t U

34 Formally, the IRC permits a shareholder in a foreign fund to elect to treat it as a pass-through entity for U.S. tax purposes, and thus obtain the beneficial tax treatment afforded U.S. private funds., However, this is permitted only if the foreign fund agrees to subject itself to SEC regulation and provides the IRS with data to determine the fund's income and gains, which most retail-marketed foreign funds are unwilling to do (IRS TD 8870 2000). In addition, the Taxpayer Relief Act of 1997 permits U.S. investors in a PFIC traded on a national securities exchange to make a "mark-to-market" election and pay tax based on the market value of the PFIC shares at the end of each year (IRC § 1296). However, any gains from the yearly mark-to-market are as taxed as ordinary income, and ordinary losses are limited to gains previously recognized. In addition, the IRS has proposed regulations that would make it difficult for many foreign funds to qualify for the mark-to-market election (IRS 2002).

Under U.S. federal tax law, dividend distributions by mutual funds are subject to the withholding tax generally applicable to dividends, except that dividends designated as long-term capital gains are exempt.35 This treatment is worse than that generally accorded foreign investors who invest directly in U.S. securities.36 This problem has long been recognized: In its 1992 report, the SEC's Investment Management (IM) recognized the difficulties faced by U.S. funds in trying to market to foreign investors and recommended eliminating the tax disadvantages for U.S. regulated funds being offered overseas (SEC Div. of Inv. Mgmt, 1992, 215-216).

Khorana et al.'s study of mutual funds around the world shows that of 20 |

countries studied, 70% had domiciled funds that were sold in other countries (2005, 34, Table 1). Although U.S. advisers do "clone" their U.S. funds and sell them in other countries (see Part II), the expenses of cloned funds exceed §

those of funds sponsored by advisers from the same jurisdiction (Khorana et al., in press). This reflects foregone economies of scale and the costs of du- .

plicative compliance and regulation. Portfolio investments necessarily reflect |

fund-level cash flows, which will differ between a U.S. fund and its cloned 0

foreign counterpart, so cloned funds never precisely track their U.S. counter- |

parts, reducing economies of scale in marketing. Thus, U.S. tax law directly r

and indirectly penalizes mutual fund investors by hamstringing the fund in- t

dustry and reducing its capability to compete globally for assets. It also results i

in an unnecessarily fragmented global fund industry, reduces geographic di- s

versification, and impedes the integration of global capital markets. The po- o

tential size of the non-U.S. fund market is suggested by Fig. 2, which shows e

that non-U.S. funds are holding an increasingly large share of the global fund <§

market, increasing from 41% in 2000 to more than 54% in 2007. §

Effects of U.S. Tax Law on Fund Investment Strategies §§

Due to the fragmentation requirements described above, U.S. funds are 5

constrained in their investment strategies. Although the fragmentation

35 See 26 U.S.C. § 852(b)(3)(B), 871(a), 881, 1441; Treas. reg. section 1.852- 4(a)(1), (b)(1).

36 This is because foreign investors are generally exempt from withholding on interest payments and capital gains, but not on dividends. U.S. mutual fund distributions are treated as dividends for withholding purposes, even though they mostly represent a pass-through of interest and capital gains (Reich 1998). In 2004, Congress tried to ameliorate this problem by enacting IRC § 871(k) to temporarily allow regulated investment companies to pass through qualified interest and short-term gains to foreign investors tax free, a provision that expired at the end of 2007 (IRC § 871[k][1][C]). Practitioners report that the requirements to obtain this treatment were sufficiently complex, however, that few funds made the necessary elections to do so.

Figure 2. US Mutual Funds' Share of Collective Investments, 2000-2007

13.6% 14.5% 13.0% 13.9%

58.7% \-

477% 45.9%

I Other I EUR I US

2000 2001 2002 2003 2004 2005 2006 2007

Sources: ICI 2008 Factbook; World Federation of Exchanges.

requirements in U.S. tax law are less constraining than those required for a U.S. fund to market itself as a diversified fund for securities law purposes, the fragmentation requirements in U.S. tax law may not be waived by a U.S. regulatory agency, such as the SEC, even if a fund sponsor can make a compelling case that such exemptive relief would benefit fund investors. For funds pursuing conventional broad-based index strategies, the fragmentation requirements generally pose no difficulties, but for ETFs and other funds that attempt to particular sector or geographic indices, the fragmentation requirements can affect investment strategy. For example, if a fund is attempting to invest on a dollar-weighted basis in a telecommunications index, and if a single company (such as AT&T in its heyday) were to make up more than 25% of the value of that index, the fragmentation requirements would prevent the fund from closely tracking the index.

The fragmentation requirements in U.S. tax law do not reflect modern financial theory or research on diversification and are thus hard to justify from the perspective of investor protection. The requirements have the effect of requiring a fund to spread its investments among different issuers, but do not require that the different issuers have uncorrelated returns. Thus, the only "diversification" the requirements insure is the potentially small benefit from moving from one to 12 highly but not perfectly correlated investments. The fragmentation requirements might be

justified, not on investor protection, but on tax policy grounds. U.S. tax law taxes mutual funds differently than other corporations, as described previously, so the law needs to have some way to distinguish between them. The fragmentation requirements provide a crude but simple way to distinguish between corporations that are collective investment vehicles and those that are operating other businesses or are functioning as holding companies for subsidiaries engaged in other businesses. That tax policy goal could be achieved with a variety of means, however. For example, a fund that committed to track a third-party index, as do index funds and passive ETFs, would be clearly and observably functioning as n

a collective investment vehicle, and there is no apparent reason why such g

a commitment could not substitute for the fragmentation requirements f

in the tax code. 3

B. Potential Changes in U.S. Taxation of Funds |

1. U.S. Investors f

A number of potential tax reforms would alleviate or eliminate many of o

these problems. The simplest, most general improvement for taxation of |

mutual funds would be to eliminate taxes on capital gains altogether. Such a r

general change would, of course, have equally general implications, extend- t

ing well beyond collective investments. Another simple but general way to i

improve mutual fund taxation would be to generally eliminate corporate-level taxation. Mutual funds would thus no longer be required to distribute earnings to eliminate corporate-level tax.

A third conceptually simple improvement would be to add mutual funds to the list of publicly held funds that are permitted to use true pass-through tax treatment. This would eliminate many of the negative economic effects of mutual fund taxation. For such a change to be implemented, however, funds, fund advisers, and fund investors would be faced with additional complexity and costs. For wealthy investors in private funds, the benefits of full pass-through taxation are often large enough to warrant the additional tax expense (out-of-pocket, to tax advisers, and opportunity costs). However, for a large number of middle-class mutual fund investors—particularly those in lower tax brackets who invest in large company index funds, which tend to have the lowest net capital gains—it is not clear that the benefits of reduced capital gains taxes would be worth the annual additional tax compliance and record-keeping costs that partnership tax would create.

A fourth possible improvement would be to eliminate the requirement that mutual funds distribute capital gains, or eliminate the requirement that mutual funds distribute long-term capital gains, which would minimize the impact of the change. Instead of distributing gains to comply with the tax law and having 90+% automatically reinvested, mutual funds would simply retain and reinvest capital gains on behalf of their investors. (Any concern you might have about investors' not wanting their money reinvested should be eliminated when you remember that the standard mutual fund issues shares that are redeemable at any time at net asset value per share (NAV).) Eliminating the requirement to distribute gains would preserve mutual funds' hybrid status—neither a true pass-through nor fully equivalent to "C" corporation status—and has the virtue of both conceptual and practical simplicity. It would also reduce the amount of record-keeping required by current rules.

On the other hand, these tax reforms would cause a potentially large short-term reduction in tax revenues, although of course gains will eventually be taxed. They could also open up the possibility of wealthy individuals using mutual funds to accomplish more tax deferral than is currently possible because they could reinvest proceeds of portfolio sales in new portfolio investments without realizing tax. For current funds, this risk seems minimal because investors have little or no control over how proceeds are reinvested, but if this proposal were adopted, wealthy investors could sponsor their own pet mutual funds, soliciting just enough passive investors to obtain favorable tax treatment, and then use the vehicle (which they would control through a controlled subsidiary adviser) to manage the fund's portfolio in their own long-term interest. Rules to combat such use of mutual funds could add substantial complexity, but one simple constraint would be to limit the percentage of shares of a mutual fund that could be held by any one shareholder without that shareholder losing the right to defer gains.

A similar possible improvement is endorsed by the ICI and reflected in a pending bi-partisan-sponsored bill in the 110th Congress—H.R. 2796, the Generate Retirement Ownership Through Long-Term Holding Act of 2005 (the GROWTH Act).37 It would shift the capital gains tax recognition event for individual investors who automatically reinvest fund

37 See ICI (2008c). This bill received the bipartisan backing of 73 members of the House and Senate.

distributions from the fund (for example, distribution of capital gains by the fund) to the fund investor (for example, sale of fund shares or death of the investor). This proposal has the virtue of relative conceptual simplicity, although it will also entail a significant increase in record keeping for individual investors, who will need to track gains distributions by their mutual funds, as currently, but then retain those records for as long as they hold their fund shares. It would be economically equivalent to eliminating the requirement that mutual funds distribute capital gains, discussed previously, and thus it too would cause a potentially large short-term reduction in tax revenues, although of course gains will even- |

tually be taxed. Likewise, the GROWTH Act could be misused, and rules designed to combat misuse of the kind described previously would need to be adopted. 3

Yet another alternative, introduced in another bill in the 110th Congress—H.R. 397—would allow deferral of recognition of reinvested .

capital gains on the first $5,000 for individuals filing singly and $10,000 |

for married couples filing jointly, adjusted thereafter by a cost of living 0

increase. Obviously more modest in scope, it would have a smaller fis- |

cal cost and would provide tax relief for a large number of mutual fund r

investors—roughly 85% of all married taxpayers filing jointly (Saxton, t

2001, 18). This proposal would also have the advantage of greatly reduc- i

ing the incentive to misuse the mutual fund vehicle in the way described previously because the ability of any given wealthy tax payer to defer gains through one or more mutual funds would be capped at the taxpayer level.

Another way of reducing the fiscal cost and limit misuse would be to limit the capability to defer reinvested capital gains to taxpayers making less than a specified amount of adjusted gross income. This was done with respect to dividend income in the Jobs and Growth Tax Relief Reconciliation Act of 2003. That law reduced taxes on qualified dividends generally, but more so for lower-income taxpayers. The rate for qualified dividend income was set at 15% for taxpayers in the top four tax brackets and 5% for those in the bottom two tax brackets (earning less than $32,500 for individuals filing singly and $65,100 for married couples filing jointly). A similar approach to taxation for reinvested capital gains distributions from mutual funds would benefit roughly 25-50% of U.S. households investing in mutual funds (ICI 2007b, 4, Fig. 5).

2. Foreign Investors

The proposed GROWTH Act, H.R. 397, or the means-tested alternative can't level the tax playing field between U.S. and foreign mutual funds, although the GROWTH Act would go much farther in that direction. The U.S. currently derives no tax revenue from foreign fund investors—there are essentially none—so another possible change, with zero revenue cost, would to be exempt bona fide foreign individual investors from taxation on and withholding related to capital gains distributions from U.S. funds. Standing alone, this would not address the disparities in the taxation of mutual funds and other collective investments in the U.S.. But combined with the GROWTH Act or H.R. 397, such an exemption would greatly improve the capability of U.S. funds to compete internationally. It will also have to be accompanied by a willingness on the part of the SEC to exempt foreign funds from the ICA, either through reciprocal treatment, as suggested in Part III, or through more flexible application of current exemp-tive procedures for foreign funds. A concerted effort by the SEC and U.S. trade representatives to secure reciprocal treatment for U.S. funds in other countries will also be needed.

3. Fragmentation Requirements

In addition, Congress should consider shifting the fragmentation requirements set out in the tax law to the securities law, pursuant to which the SEC would have authority to exempt particular funds from those requirements, based on the costs and benefits to investors. This would also have the benefit of consolidating fund regulation in a single set of rules, rather than imposing potentially disparate rules through both the tax and securities laws, and would allow for the requirements to be updated over time through SEC rule-making. To the extent the diversification requirements are thought to satisfy tax rather than securities law policy goals (for example, by preventing operating companies from camouflaging themselves as mutual funds to avoid corporate taxation), those goals would be better served with more flexible tests, as noted previously; by imposing penalties rather than complete disqualification on mutual funds that fail to satisfy the requirements; and by imposing those penalties on the advisory companies, rather than on the funds themselves, which would pass along those taxes to fund investors.

4. Performance Fees

Finally, as noted previously, private fund managers can be paid performance fees that are taxed more favorably (as carried interest) than performance or other advisory fees paid by mutual funds. Current securities law and SEC regulations—discussed in Part II—limit how performance fees can be structured by mutual funds, making them generally unattractive for most mutual fund advisers. If, as proposed in Part III, mutual funds were given more flexibility to pay performance fees, tax law should also be adjusted to permit such fees to be taxed in a manner equivalent to the tax treatment for performance-based returns for private fund managers. The |

details of how precisely to achieve such equivalence are beyond the scope of this report, but the general principle—equivalent taxation for equivalent f

services for collective investments available to middle-class and wealthy |

U.S. investors—is straightforward.

3. securities regulation of mutual funds and 0

other collective investments I

A. Regulation of Collective Investments in the U.S. o

The primary U.S. securities law regulating collective investments in securi- a

ties is the Investment Company Act of 1940 (ICA). As described in more n

detail later, the ICA differs from the dominant mode of federal securities regulation. It goes far beyond disclosure and anti-fraud to forbid and require numerous detailed actions for most collective securities investments, while exempting collective investments sponsored by financial institutions that are separately regulated, and exempting altogether a limited class of private funds.

A number of different types of funds are required to be registered under the ICA, as briefly described in Appendix A. The most important are mutual funds, which do not in specified securities (although they might track a specified index), issue securities redeemable at the funds' NAV per share, and do not list their own securities on an exchange, instead typically offering their shares continuously for sale to the public. An important subcategory is MMMFs, which by regulation are limited to short-term low-risk debt (primarily government securities) to preserve a fixed NAV. More recently, a new type of fund—ETFs—has emerged, which are typically open-end funds that issue and redeem shares only in large blocks to authorized participants, who purchase the units with portfolio securities, break up the

creation units, and sell them to the public. The units then trade as shares on a stock exchange, as would a closed-end fund (McGuire and Helmrich 2008).

Figure 3 presents recent estimates of the share of all collective investments in securities held by funds of various types, including mutual funds, MMMFs, and regulated ETFs.

Figure 4 presents their compound annual growth rate (CAGR) in recent years.

Mutual funds remain the most important type of collective investment, but ETFs registered under the ICA have experienced CAGRs that put them on track to match or exceed the mutual fund industry in 12 years. Even if registered ETFs experience a slowing of growth as their penetration of the collective investment market proceeds, they nevertheless represent a serious rival to the traditional mutual fund in the U.S.

Other types of collective investments that are not covered by the ICA (as detailed in Appendix B) have been growing even more rapidly than mutual funds and ETFs. Broadly speaking, financial sectors in place at the time the

Figure 3. Assets Under Management in US Collective Investments

$53 $29 $5



All amounts in $US billions. All are estimates for 2007, except bank common trust funds (2005), SMAs (2006), and commodity pools (2003). Private equity and venture capital investments equal new 2006 investments in the US, times 5, a common holding period for such investments. Sources: Investment Company Institute; Money Management Institute; Hennessee Group LLC; Morningstar, Inc.; Federal Reserve Board; FDIC; PWC Moneytree; CFTC.

Figure 4. CAGR for Types of US Collective Investments

CAGR = compound annual growth rate for longest period for which data is available (number of years given in parentheses for each type of collective investment).

Estimates for 2001-2007, except bank common trust funds (2002-2005), SMAs (20022003), which present the longest period for which data are available, and exempt ETFs (20042007) and ETNs (2006-2007), which present the period for ehich the products have been in existence.

Private equity and venture captial investments equal investments for each year in the US times 5, a common holding period for such investments. For sources, see Fig. 1.

ICA was adopted have been able to preserve and expand their capability to sell and service collective investments outside the ICA:

• Commercial banks sponsor common trust funds that allow trust accounts (including IRAs) to be collectively invested.

• Insurance companies continue to sell insurance policies and fixed annuities outside the ICA.

• Investment banks sell SMAs (also called wrap or managed accounts) that rely on economies of scale achieved through standardized algorithms for portfolio selection, pooled execution and custody, and shared back-office facilities, as well as ETNs, unsecured debt contracts that pay returns based on specified market indexes (and so track the performance of equivalent ETFs) (, 2006).

<¡1 t U

66 Other collective investments holding specialized assets have been exempted or excluded from the ICA. The collective investment include, for example,

real estate investment trusts (REITs); securitized assets, such as credit card receivables or collateralized mortgage obligations; and commodity pools and ETFs that invest primarily in commodities, foreign currencies, or derivatives, including futures on broad-based stock indexes (see U.S. Commodities Trad. Comm. 2000). Finally, private funds (such as hedge funds, venture capital funds, and private equity funds) are excluded from the ICA's coverage if they limit their services to a small number of investors or to very wealthy investors.

As can be seen from Figures 3 and 4, regulated funds continue to be the 67 single largest category of collective investments in the U.S., but they are losing share to other types of collective investments. The fastest growing collective investments are ETNs, ETFs (particularly exempt ETFs), commodity pools, hedge funds, and SMAs. Among regulated funds, only registered ETFs have been growing faster than the weighted average CAGR for all U.S. collective investments over this period (excluding MMMFs from the average). At estimated recent rates of growth, commodity pools and hedge funds will have more assets than non-MMMF mutual funds in the next 10 to 20 years. The spread of funds of funds, in which mutual funds invest in hedge funds and commodity pools, represents the direct shift of assets from regulated funds to these alternatives.

B. Regulation of Mutual Funds and Other Regulated Funds in the U.S.

In addition to the Internal Revenue Code, the primary statute governing 68 mutual funds and other regulated funds is the ICA. The ICA is heavily proscriptive. It requires and forbids numerous actions in the operation of regulated funds. The ICA

Requires written contracts with its adviser and underwriter approved by a majority of independent directors and by fund shareholders.38 Regulates fund capital structure, requiring equal treatment of and voting rights for investors.

Limits the use of leverage to increase the risk of a fund.

Strictly regulates the custody and use of investor funds and portfolio securities.

Requires detailed books and records.39

38 ICA § 15.

39 ICA §§ 10, 17, 18, 31.

• Effectively requires that a fund have a distinct set of individual directors or trustees charged with overseeing the fund.40

• At least 40% of the directors or trustees must consist of "independent" directors.41

The ICA almost completely bans conflict-of-interest transactions between an adviser and a fund involving portfolio investments, loans, or purchases of fund assets.42 Advisers can thus generally extract value from a fund along only one readily monitored path—the advisory fee—that is itself regulated in a number of ways (Coates and Hubbard 2007).

A final set of legal restraints are embedded in the contracts required by the ICA between fund advisers, funds, and fund shareholders. The contractual provision that is one of the defining characteristics of the mutual fund, redeemable shares, is especially regulated (Schonfeld and Kerwin 1993). Funds that offer redeemable shares must comply with the ICA provisions on redemption and pricing of fund securities. Foreign funds are banned without an SEC exemption,43 which has only rarely been granted (SEC Div. of Inv. Mgmt 1992, 189-90; Roye 2002).

40 ICA § 10. It is sometimes said the ICA requires a mutual fund be established as a separate legal entity (SEC Div. of Inv. Mgmt 1992, 252; Wallison and Litan 2007; ICI 2007a), but mutual funds at their inception were commonly organized as trusts, which do not have the same distinct legal personality as do corporations. A sponsor designates trustees to hold title to fund assets in their name and oversee those assets for the benefit of investors (Fink 2008, 11, and Rounds 2007, 473). (Modern funds more commonly take the form of a Massachusetts business trust, as with Fidelity, or Delaware statutory trust, as with Vanguard, but these trusts do have separate legal entity status.) Insurance companies commonly sell variable annuities, organized as "separate accounts," which are regulated under the ICA. If one views those entities, too, as separate legal entities because their assets are shielded from creditors of the trustees and insurance companies (Hansmann and Kraakman 2000, 416), or because the trustees can sue and be sued in their capacity as trustees (Hansmann and Mattei 1998), the same is true of unit trusts and other non-corporate fund structures used as UCITS-compliant funds in prominent E.U. fund domiciles, such as Ireland and Luxembourg. See Rounds and Dehio (2007, 479), which equates unit trusts in the E.U. with business trusts in the U.S.

41 ICA § 10. The SEC has by rule created incentives for funds to have a majority of independent directors and has mandated that each fund have a compliance officer that reports directly to the fund's independent directors (Role of Independent Directors of Investment Companies, ICA Rel. No. 24816 Jan. 2, 2001). The SEC twice approved a rule effectively requiring that the chair of each fund board and 75% of its directors be independent, but the D.C. Circuit twice struck down the rule based on the process followed by the SEC. Subsequently, the SEC has reproposed, but not yet approved, the same rule (Investment Company Governance, ICA Rel. No. 26520 July 27, 2004). For a critique of the rationale for this rule, see Coates (2008).

42 ICA §§ 10, 17.

43 ICA § 7(d).

Advisers to mutual funds and other investment advisers are also subject 71 to the Investment Advisers Act of 1940 (also called the Advisers Act).44 By comparison to the ICA, the Advisers Act generally imposes relatively light regulation:

• Advisers with assets under management of at least $25 million and foreign advisers operating in the U.S. must register with the SEC. Other advisers must register with a state.45

• An anti-fraud rule is judicially interpreted to require full disclosure to clients.46

• Advisers must maintain records and compliance policies, available for inspection by the SEC.47

One significant restriction on advisers to regulated funds is a general ban on 72 asymmetric advisory fees based on fund performance.48 Directors, trustees, and investment advisers are considered fiduciaries under state law and are subject to duties in providing services to clients.49

Finally, advertisements, sales materials, and sales and distribution fees of un- 73 derwriters of mutual fund shares are subject to rules of the Financial Industry Regulatory Authority (FINRA).50 The ICA also effectively requires management funds to be organized as corporations or trusts, so they are governed by state laws regarding the election and duties of directors or trustees, who are responsible under both state law and the ICA for fund policies and procedures.

<¡1 t

C. Practicalities of Fund Governance

Although the ICA induces advisers to establish U.S. mutual funds as separate corporations with their own boards of directors, mutual funds do not

44 Funds that accept 401(k) and other retirement plans must also provide services that permit plan sponsors to comply with the Employee Retirement Income Security Act of 1974 (ERISA) and regulations of the Department of Labor.

45 Advisers Act § 203.

46 SEC v. Capital Gains Research Bureau, Inc., 375 U.S. 180, 186 (1963).

47 Advisers Act §§ 204, 204A.

48 Advisers Act § 205(a). The ban is subject to an exception for fees that both increase and decrease based on performance over a specified period, known as fulcrum fees, in accordance with SEC rules (Advisers Act § 205[b]).

49 Boss v. La Salle Bank, N.A., 84 F. Supp. 2d 947 (N.D. Ill. 1999); ICI (2007c).

50 For example, FINRA enforces NASD Rule of Conduct 2830, limiting charges in sales of fund shares and banning kickbacks.

function as independent corporations. With few exceptions, funds do not have officers or employees of their own. Rather, by contract, mutual funds outsource their operations to their advisers and other service providers. Advisers or other third parties carry out the most important operations with minimal involvement by fund boards, including investment management, human resources (including hiring, firing, and compensating individual officers and employees), trading, record keeping, reporting, advertising and marketing, distribution, and accounting (Frankel and Schwing 2001, § 1.01[B], § 12.01). As an economic and practical matter, mutual funds are created and managed as portfolios controlled by n fund advisers. Mutual fund boards thus do not manage funds, but instead provide a form of privatized regulatory supervision. In addition, because fund shareholders elect fund boards, the existence of fund boards pro- g vides at least a theoretical way for fund investors to monitor and provide a check on fund advisers. .

In practice, boards are traditionally limited to ICA-mandated, minimal |

activities, all of which are essentially forms of conflict-of-interest over- O

sight.51 It would be possible, of course, for the SEC to perform these tasks. |

In Europe, as discussed more in Part II.D, many of the tasks are carried out r

by national regulators equivalent to the SEC. Those who want the U.S. to t

adopt European models for fund regulation (as a basis, for example, for i

advocating elimination of fund boards [Wallison and Litan 2007]) should consider whether such a move would risk either eliminating the possibility of engaging in transactions now given to independent directors to approve, or shifting authority to approve all such transactions to the SEC, creating more delay and increasing government entanglement in fund management. Even if some items currently subject to fund board approval could be plausibly given to fund advisers, some independent party would need to approve, for example, valuations where market prices are unavailable.

D. Enforcement

76 Both public and private actors enforce these laws. The SEC regulates funds under the ICA and advisers under the Advisers Act and can impose an array of sanctions on individuals or firms who break its rules (Frankel and Schwing 2001). At the SEC,

51 Specifically, fund boards (or independent directors) engage in the activities listed in Appendix C. For a similar list, see SEC Staff (2005, 3-30).

• The Director of the Division of Investment Management (IM) is responsible for administering the ICA and the Advisers Act, and with respect to matters pertaining to collective investments, mandating standards of financial reporting and fair disclosure under the 1933 Act and the 1934 Act.

• The Director of the Office of Compliance Inspections and Examinations (OCIE) examines and inspects mutual funds and investment advisers.

• The Director of the Division of Enforcement has authority over all enforcement activities under all federal securities laws, including the ICA and the Advisers Act.52

Shareholders may initiate private lawsuits under both the ICA and state corporate or trust law (Clark 1986; Rogers and Benedict 1982), which often take the form of class actions or derivative actions (Benedict et al. 2008). State attorney's general and securities commissioners also provide oversight.

E. The Crucial Role of the SEC and the ICA's Exemptive Procedures

Because the ICA and the SEC's existing rules are so restrictive and prescriptive and nearly 70 years old, exemptive procedures are crucial to regulation of U.S. mutual funds. It is likely that, unless the ICA is to be entirely rewritten, significant future innovation in the mutual fund industry is likely to depend on exemptive procedures. The ICA authorizes the SEC to exempt "any person, security, or transaction from any provision" of the ICA "to the extent that such exemption is necessary ... and consistent with the protection of investors and the purposes fairly intended" by the ICA.53 Hundreds of applications are filed each year, and most of the major innovations in the U.S. fund industry over the past three decades have required such relief, as shown in Table 2. The SEC has needed its exemptive authority to permit the creation of index funds, MMMFs, securitizations, and ETFs, to permit regulated funds to offer multiple classes of securities with different sales

<¡1 t U

52 17 CFR 200.19b to 20b.

53 15 U.S.C. § 80a-6(c). In addition, the SEC has specific authority to exempt regulated investment companies from more than30 specific provisions of the ICA (SEC Div. of Inv. Mgmt. 1992, 503 n.2).

Table 2. Examples of Exemptive Orders for Past Mutual Fund Innovations

Innovation Importance of Innovation Need for Exemptive Relief Exemptive Order Exemptive Rule

"Internal" ownership structure used by Vanguard Enabled creation of single-most successful mutual fund complex Prohibition on fund acting as principal to effect any transaction in which company is a joint participant, ICA § 17(d), Rule 17a-1 ICA Rel. No. 15788 (June 9, 1967)

Money market mutual funds Largest new class of mutual funds developed since 1940; helped create pressure to deregulate interest rates on bank savings accounts Impracticality of maintaining constant NAV—which was required for trust and other institutional money fund customers—consistent with the requirement that shares be redeemable at current market value or fair value as determined by fund board, ICA §§ 2, 22, SEC rules 2a-4, 22c-1; ICA Rel. No. 9786 (June 7, 1977) ICA Rels. No. 10451 (Oct. 26, 1978) (permitting penny rounding for debt securities with maturity of less than 60 days); No. 10825 (permitting use of amortized cost method) Rule 2a-7

Waiver or deferral of sales loads (such as permitting contingent back-end sales charges) Enabled development of new sales channels and marketing techniques, reducing costs for investors Requirements that fund shares be sold at NAV and at a price described in the prospectus, ICA §§ 2, 22, Rule 22c-1 ICA Rels. No. 15118 (May 28, 1986), 15455 (Dec. 4, 1986), 15745 (May 19 and June 2, 1987), and 16526 (Aug. 16, 1988) Rule 22d-1

Pooling of cash balances by funds within a complex Facilitated economies of scale within a mutual fund complex, reducing costs for investors Prohibition on fund acting as principal to effect any transaction in which the company is a joint participant, ICA § 17(d), Rule 17a-1 ICA Rel. No. 19158 (Dec. 16, 1992)

Exchange traded funds Largest single new class of mutual funds developed since money market mutual fund Requirement that fund shares be sold at NAV, rather than in large blocks; limit on acquisitions of funds by other funds, ICA §§ 12, 22, Rule 22c-1, 12d-1-2 ICA Rel. No. 19055 (Oct. 26, 1992) Proposed Rule 6c-11

charges and expenses, to impose back-end fees (contingent deferred sales loads), and to permit efficient operation of mutual fund complexes.54

Despite the breadth of the SEC's exemptive power under Section 6(c), the SEC has viewed its authority more narrowly, stating that Section 6(c) is "not blanket authority to waive any provision of the [ICA]."55 The SEC treats a number of provisions in the ICA as core principles and is unwilling to consider exemptive requests that would deviate from those core principles. These core principles have never been formally identified, but would likely include, at a minimum, the anti-fraud provisions of the ICA and structural aspects of the ICA, such as the requirement that a board of individual trustees or directors review and approve advisory fees. Nowhere in the ICA or its limited legislative history is there any clear statement that the SEC should preserve these core principles or what they might be. If anything, the opposite is the case: As the SEC itself has noted, "Congress enacted Section 6(c) to give the [SEC] the flexibility to address unforeseen or changed circumstances in the investment company industry."56

To get an exemption, an applicant must file a detailed application describing benefits for investors and agreeing to conditions to protect investors from risks created by the proposed exemption.57 IM's stated position

54 Valuation of Debt Instruments and Computation of Current Price per Share by Certain Open-end Investment Companies (Money Market Funds), ICA Rel. Nos. 12206 (Feb. 1, 1982) and 13380 (July 11, 1983); Goldman Sachs-Institutional Liquid Assets, ICA Rel. Nos. 17420 (Apr. 11, 1990), and 17479 (May 8, 1990); Freedom Investment Trust, ICA Rel. Nos. 16487 (July 20, 1988) and 16526 (Aug. 16, 1988); WisdomTree Trust et al., ICA Rel. Nos. 28147 (Feb. 6, 2008) (notice) and 28174 (Feb. 27, 2008) (order); Actively Managed Exchange-Traded Funds, ICA Rel. No. 25258 (Nov. 8, 2001); Frankel and Schwing (2001, 31-101 and n. 452.). "Like all funds that were not in existence in 1940, index funds raised a number of problems under the 1940 Act. The SEC's approach to such innovations, as to others, was to allow them but retain the regulation of possible problems that concerned Congress. The operations of these funds to the 1940 Act were usually adjusted by no-action letters. ... These funds have also been dealt with by in exemptions."

55 ICA Rel. No. 14492 (1985). See also First Nat'l Bank, ICA Rel. No. 10680 (1979) (denying application for exemption from all provisions of the ICA); Nat'l Ass'n of Small Bus. Inv. Cos., ICA Rel. No. 5581 (1969) (SEC uses exemptive powers "with circumspection").

56 ICA Rel. No. 19362 (1993). See also Investment Trusts and Investment Companies: Hearings on S. 3580 Before a Subcomm. of the Senate Comm. on Banking and Currency, 76th Cong., 3d Sess. 872 (1940) (Commissioner Healy, a principal author of the ICA, stated that "it seemed possible and even quite probable that there might be companies—which none of us have been able to think of—that ought to be exempted."); id. at 197 (David Schenker, Chief Counsel of the Investment Trust Study, and also a principal author of the ICA, stated that "the difficulty of making provision for regulating an industry which has so many variants and so many different types of activities ... is precisely [the reason that section 6(c)] is inserted.").

57 Applicants must follow SEC Rules 0-2, 0-4, and 0-5 and the guidance in ICA Rel. No. 14492 (1985).

<¡1 t U

(approved by the SEC) is that it "will not support an application that requests relief not adequately justified."58 The SEC's view is that IM must "analyze thoroughly" all issues raised by new proposals before they can be approved.59 The SEC puts the burden of persuasion on applicants60—it will not presume that a proposed exemption has public benefits unless those benefits can be and are articulated in a written application.

This approach is 180 degrees opposite of the approach the SEC takes in respect of prospectuses under the 1933 Act or proxy statements under the 1934 Act. In essence, the SEC's stance toward exemptions is "deny unless convinced to approve," rather than "approve unless there is a reason to 3 deny." This approach means that innovative proposals—where both potential benefits and risks are highly uncertain precisely because the proposals are innovative—might be denied even if they would, if implemented, pro- g vide net benefits to investors. This approach also tends to mean that third-party or systemic benefits (such as the effects of enhanced competition) . will not be given adequate weight in the analysis because the applicant will | tend not to have good information to compellingly present these benefits 0 in their application. The applicant may not necessarily have an interest in | presenting benefits because applications are generally public and might re- r veal sensitive information to competitors. t

IM has no authority to deny applications, but IM can recommend the 3

SEC do so, and the SEC has never approved an application against IM's recommendation. As a result, applicants rarely force IM to recommend against an application actually decided by the SEC. Within IM, the Office of Investment Company Regulation (OICR) reviews most applications.61 OICR has approximately 20 staff attorneys, who divide applications into routine (requesting relief previously granted, on the same conditions) and novel (others), but generally reviews them in the order received.62 Most of these attorneys, and particularly senior attorneys, have spent much of their

58 ICA Rel. No. 14492 (1985).

59 ICA Rel. No. 19362 (1993).

60 Scripps-Howard Inv. Co., 17 SEC 702, 707 (1946).

61 In 2005, OICR reviewed 94% of Section 6(c) filings, including 225 exemptive applications and 158 deregistrations. SEC Inspector General, 40 Act Exemptive Applications, Audit Rep. No. 230 (Mar. 19, 1996). IM's Office of Insurance Products review applications involving insurance products, and Office of Investment Adviser Regulation reviews applications involving investment advisers.

62 Id.; ICA Rel. No. 14492 (1985); ICA Rel. No. 19362 (1993).

careers at the SEC. In part, the detailed and technical nature of the ICA and the SEC's rules and prior exemptions require an extensive training period before attorneys can usefully process exemptive applications. For the same reason, outsourcing has not been a means for IM or OICR to process applications. OICR does not have on its staff non-legal experts, such as economists or public policy experts, who might be in a better position than lawyers to evaluate the business or economic benefits of the product for which an exemption is requested, nor is the SEC organized such that OICR could make practical, extensive use of economists in the SEC's Office of Economic Affairs for this purpose.

The SEC's guidelines require OICR to provide initial comments to applicants within 45 days, after which the applicant has 60 days to respond with modifications or otherwise address the staff's comments, after which an application is deemed inactive until the applicant responds. Applicants commonly file applications in draft,63 creating at least a three-stage process: draft filing and review, formal application and review, and final submission. After IM clears an application, notice is published in the Federal Register, and, unless a hearing is requested, it is generally approved within 30 days thereafter. The Director of IM has had delegated authority to issue notices and to issue orders where a notice has been issued, no hearing has been requested, the matter presents (in his or her judgment) no significant issue not previously settled by the SEC, or otherwise necessitating a hearing in the interest of the public or investors.64 Other orders must find a place on the SEC's lengthy agenda.

<¡1 t U

F. Congressional and SEC Allocation of Resources to Oversight of the Fund Industry

Studies have shown that staffing levels and budgets can have important effects on capital market development (Jackson and Roe 2008; Jackson 2007; Jackson 2006). Currently, the SEC's annual performance and accountability reports make it clear that it dedicates far more effort to enforcing existing laws than to considering their improvement or reform. The SEC's allocation of staffing and budget resources reflects its priorities: first, enforcement and compliance, and a distinct second, promotion of capital markets,

63 In 2005, 142 (63%) of the exemptive applications filed were drafts, and 83 were final applications (SEC 2006b). This is despite the fact that the SEC's guidelines state that draft applications will be reviewed only in "the most extraordinary situations" (ICA Rel. No. 14492 1985).

64 17 C.F.R. § 200.30-5(a).

including processing of exemptive orders and other activities to provide flexibility and respond to innovative fund structures and otherwise develop the fund market. As reflected in Table 3, the SEC is seeking authority to spend 68% of its 2009 budget on enforcement and 9% on the promotion of capital markets. By contrast, the CFTC has its strategic goals in the reverse order: first, to "ensure the economic vitality of the commodity futures and options markets," to which it allocated $37.6 million in 2007, and second, "to protect market users and the public," to which it allocated $35.6 million (CFTC 2007).

The SEC and IM Have No Secure Sources of Funding Despite the fact that the SEC receives a predictable, large amount of fees from companies registering securities each year, the SEC has no secure source of funding. That is because the SEC must turn over all such fees to the U.S. Treasury and must

Table 3. Allocation of SEC Resources Relevant to Mutual Funds

Requested SEC Budget 2009 ($MM) % of Total Requested SEC Staff 2009 (FTE) % of Total

Total $913 3473

SEC Goals

Enforcement goal $620 67.9%

Promote markets $83 9.1%

Foster informed decisions $124 13.6%

Maximize SEC resources $86 9.4%

SEC Divisions

Division of Enforcement $318 34.8% 1093 31.5%

Office of Compliance Inspections and Examinations $206 22.6% 796 22.9%

Division of Corporation Finance $114 12.5% 439 12.6%

Division of Trading and Markets $44 4.9% 174 5.0%

Division of Investment Management $45 4.9% 145 4.2%

Other $230 25.2% 1000 28.8%

SEC Staff Promoting Collective Investments

Division of Investment Management devoted to promoting markets 49 1.4%

<¡1 t U

seek annual approval from Congress for appropriations to fund its operations. In seeking annual funding, the SEC must also go through an annual apportionment process through which the Office of Management and Budget makes specific amounts available to the SEC for specific time periods (usually quarters), activities, projects, objects, or a combination thereof. As a result of this process, the SEC generally is able to spend only a portion of the funds it generates from fees it charges to funds and other companies that register securities with it (SEC 2008a). Moreover, within the SEC, IM has no independent budgetary authority. Its spending is controlled by the Commissioners. In contrast, Congress has provided permanent budget authority to the federal banking agencies, allowing these agencies to use all the funds collected without further legislative action. These agencies are generally not included in the annual appropriations process (GAO 2002). Accordingly, both the SEC and IM have much less secure funding than their counterparts supervising banks. As a result, their capability to make long-term investments in regulatory capacity is less secure. IM's capability, in particular, to hire extensive new staff to consider and review novel exemp-tive applications is severely constrained by its lack of secure funding and spending authority.

The SEC Devotes Relatively Few Resources to the Fund Industry The budget request also shows the relatively small role IM has within the SEC: the Division of Enforcement would receive 35% of the SEC's budget and IM would receive 5%. Of 842 new positions filled in the two years following the Sarbanes-Oxley Act, the fewest (15) went to IM of all the SEC's divisions. IM also had the slowest budget growth from 2002 to 2004 of all of the SEC's divisions—12%, compared to 162% growth for the SEC overall (GAO 2004b, Tables 1 and 2).65 Since 2002, IM's role has remained constant within the SEC, at 4-5% of the SEC's overall budget. Yet, the number of investment advisers supervised by IM "grew substantially, from 7,614 in 2001 to 10,484 in 2006, whereas the number of broker-dealers declined from 5,526 to 5,068" (Hung et al. 2008, xiv).

IM's role in responding to industry requests for innovation is even smaller because IM also plays an enforcement and information review role. The Sarbanes-Oxley Act requires the SEC review all regulated fund disclosures every three years,66 and that the SEC has allocated that responsibility to

<¡1 t U

65 This growth was 33% slower than that forecast by the SEC in 2004 (GAO 2004a).

66 Sarbanes-Oxley Act of 2002 § 408(c).

IM.67 In addition, the SEC's 2009 budget request shows that almost as much (~$12 million) of the amount for IM to pursue enforcement as was requested (~$16 million) for it to promote capital markets.68 The portion of IM's full-time equivalents (FTEs) to be dedicated to enhancing capital markets represents 1% of the total staff positions at the SEC—fewer than 50 staff personnel, roughly half of whom are assigned full time to process exemptive requests. The relatively small allocation of resources to IM might be due, in part, to the fact that the SEC derives no income from its exemptive relief efforts under the ICA. 69

The Time Required to Create New Funds in the U.S. Although the SEC does n

not have formal responsibility under the ICA of approving new funds, so long as they do not require an exemptive order, the SEC's staff does review f

and clear fund disclosures under the 1933 Act. As a result of this review, g

new U.S. funds take up to 15 months to create, and even standardized funds created by established fund advisers take more than 90 days. The U.S. is an .

outlier. Khorana et al. report that fund creation in the U.S. takes almost |

if not quite the longest period of time in a sample of 30 countries sur- 0

veyed by KPMG (2005). The normal time period for approval of standard |

new UCITS funds in Luxembourg is 60-90 days (KPMG 2008a; KPMG r

2008b).70 In Ireland, if the Irish Financial Services Regulatory Authority t

(IFSRA) already knows the manager, the time to create a new fund is no i

more than 60 days (Khorana, Servaes, and Tufano 2005), and interviews with informed practitioners suggests it often can be accomplished in much less time.

The Time Required for Funds to Obtain Exemptive Relief to Innovate More important for ongoing innovation and growth of the fund industry is the time to obtain exemptive relief for novel funds and products. The IM Staff's 1992

67 17 CFR 200.20b.

68 See SEC (2008a) (compare Charts 3 and 4 at pages 9 and 16).

69 Since 1996, applicants have not had to pay any fees to obtain exemptive relief. ICA Rel. No. 22224 (Sep. 17, 1996) eliminates pre-existing fees for exemptive relief, previously adopted pursuant to the Independent Offices Appropriations Act of 1952, 31 U.S.C. 9701. As a result, there is no in-flow to the SEC (or the general budget) as a result of the large number of exemptive requests under the ICA. Staff and other resources required to process ICA exemptive requests must be derived from the SEC's general budget, which in turn derives primarily from fees for registration of securities under the 1933 Act, which at the time the exemptive relief fee was eliminated, had greatly outstripped the expenditures of the SEC as approved by Congress, making further fees under the IOAA unnecessary and, relative to registration fees, less efficient to collect.

70 UCITS Art. 5a requires E.U. member states to approve or deny new management companies within 180 days.

"Redbook" reported that at that time exemptive orders were granted 190 days after filing, on average, down from 205 in 1990. In 1993, the SEC stated that exemptions took "six to eight months from the date an application is received." The SEC's Inspector General reported that in 1994 the time required to obtain an exemptive order averaged 167 days (SEC Div. of Inv. Mgmt, 1992, 505; SEC 1996).71 These periods did not include pre-filing preparation and/or discussions with staff. In 2005, the SEC had "85 requests ... pending for more than a year" ("Donahue Talks" 2006; Wilson 2006). Of 83 non-draft exemptive applications filed in 2005, a bare 13 (16%) received initial comments within IM's self-imposed 45-day guideline (SEC 1996). Current IM Director Andrew Donahue has stated that speeding up the exemptive process is a priority for IM (SEC 1996). IM's turnaround times have markedly improved recently, from a median time of 15 months in SEC fiscal year 2006, to eight months in fiscal year 2007, to six months in the first nine months of fiscal year 2008.

In the case of novel applications, however, the period has been and remains longer (SEC 1996). For the first ETF order, the time was 28 months; for the first actively managed ETF orders, approved in early 2008, the delay was over seven years.72 In its recent release proposing Rule 6c-1, codifying exemptions previously issued for ETFs, the SEC reported that ETF exemptive requests to create a new ETF took between "several months to several years" before approval,73 and that requests by an ETF to permit other regulated funds to invest in the ETF above the "fund of fund" limits in ICA § 12(d)(1) took 15 months on average, or longer if additional relief was requested.74

In its budget request for 2009, the SEC adopted new performance metrics breaking out ICA requests separately and also separating out exemptive requests from no-action relief and interpretive requests.75 This should

71 ICA Rel. No. 19362 (1993).

72 SPDR Trust, Series 1 et al.; Notice of Application, ICA Release No. 18959, (Sep. 17, 1992). 5 The application was originally filed on June 25, 1990 and was amended and refiled four more

times before being cleared by IM. The Matter of SPDR Trust, Series 1 PDR Services Corp. c/o American Stock Exchange, Inc., ICA Rel. No. 19055 (Oct. 26, 1992) is an order granting relief.

73 IC-28193 (Mar. 11, 2008), at 102. Applications alone cost between $75,000 to $350,000 in legal and other fees.

74 Note 71 supra, at n.328.

75 The SEC has for many years annually reported to Congress that it responded to more than 85% of requests for no-action and interpretive relief and applications for exemptions within 60 days, but has not historically separated this data for requests under the ICA from requests under the other securities laws. See SEC Performance and Accountability Reports for 20042006, available at

improve both the public's understanding of how the exemptive process is working and the process itself (SEC 2008a). However, the SEC also set for itself a performance target of giving initial comments to 80% of exemptive requests under the ICA within 120 days—much longer than the 45 days nominally required in IM's own guidelines, which date from 1985.76 Most exemptive requests are not significantly innovative. Thus, the SEC, in setting its own performance target for the first time under the ICA, has essentially proposed that it respond to routine exemptive requests four months after receiving them—more than double the time it takes for an established manager to obtain regulatory approval for a new fund in Ireland, | and 50% longer than in Luxembourg. Standing alone, such a performance goal seems low.77

From time to time, IM recognizes a category of exemptive requests that g

could be handled more efficiently by rule, and the SEC has engaged in rulemaking in part with the goal of speeding the exemptive process. The SEC's . March 2008 ETF rule proposal, for example, will eliminate the need for a | large number of exemptive requests and free up IM staff resources to address 0 other exemptions. Nevertheless, rule-making can be even slower than ob- | taining exemptive relief, in part due to requirements of the Administrative r Procedure Act that apply to rules but not to exemptive orders, and in part t due to the need for the SEC to anticipate litigation challenges to its rules.78 i As a result, substantial numbers of exemptive applications can build up during the rule-making process. In any event, no particular rule can address the general problem of exemptive delay in an environment of continued and accelerated financial innovation, competition, and change.

Summary In sum, the U.S. mutual fund industry is tightly regulated. Supervision and enforcement are robust.79 Indeed, the ICA inverts the conventional securities law paradigm, which largely permits businesses to

76 ICA Rel. No. 14492 (1985); ICA Rel. No. 19362 (1993), at n.15.

77 Although not strictly comparable, it is also worth noting that the CFTC approved 100% of "novel or innovative proposals or requests for CFTC action addressed within six months to accommodate new approaches to, or the expansion in, derivatives trading, enhance the price discovery process, or increase available risk management tools" and 100% of "no-action or other relief completed within six months related to novel market or trading practices and issues to facilitate innovation" (CFTC 2007).

78 See note 200 infra (litigation striking down recent SEC rules).

79 In 2006, the SEC inspected funds holding roughly 40% of the industry's AUM. SEC Annual Report 2006, at 12. Enforcement cases against mutual funds and investment advisers represented roughly 15% of the SEC's enforcement docket in 2006 (SEC 2006b).

design their own securities, governance, and operations, subject to disclosure and market discipline, and replaces it with a regime in which most innovations are prohibited unless the sponsor can obtain exemptive relief. This inverted structure imposes a large and economically significant burden on one division of the SEC—the Division of Investment Management. Yet IM receives fewer resources (in dollars or people) than does the rest of the SEC.

G. Comparison of U.S. Mutual Fund Regulation with U.S. Regulation of Other Collective Investments

Mutual fund regulation is more stringent in most respects than the regulation of any other type of collective investment in the U.S. A comparison of the ICA with regulation applicable to mutual funds' close domestic competitors might provide some insight into whether U.S. mutual funds are overregulated generally, whether it might be worth re-thinking some specific mutual fund rules, and if modes of regulation exist that are useful models for mutual fund oversight. Regulation of substitutes ranges from the nearly unregulated private funds, through the more lightly regulated commodity pools and commodity ETFs, to the differently regulated SMAs, ETNs, and common trusts.

Private Funds Private funds are, as noted previously, not heavily regulated. They are typically organized as limited partnerships, in which investors have fewer control and other rights than mutual fund investors and rights vary among investors. They do not typically have boards dominated by independent directors. Neither they nor their advisers are typically registered under the ICA with the SEC, or are otherwise subject to inspections and examinations. They provide less disclosure to their investors, particularly of their portfolio holdings, and are not subject to 1933 Act or 1934 Act disclosures, such as the proxy rules. They pay performance compensation to their advisers—typically 1- or 2-and-20 fee structures.80 They typically leverage their investments beyond the levels permitted by the ICA, commonly engage in short sales, and their investment strategies are only constrained by contract, in limited ways. They sometimes acquire more concentrated positions than would be permitted by the ICA, and a number of them use

80 1 or 2% asset fees and 20% performance fees are typical of U.S. hedge funds (Goetzman, Ingersoll, and Ross 2003). Performance fees are used by 32% of SEC-registered investment advisers generally, despite the fact that they can only be offered to high net worth clients (Nat'l Reg. Svcs 2007).

those positions to engage in shareholder activism to pressure portfolio firms to change their business or financial strategies. They typically permit only monthly, quarterly, or annual (rather than daily) redemptions by their investors. Advisers to private funds might rely on an exception from the requirement to register under the Advisers Act, if in the prior year they had fewer than 15 "clients" (or funds) and neither hold themselves out to the public as advisers nor act as advisers to a fund registered under the ICA.81 As applied to limited partnerships (the conventional choice for private funds), the SEC had interpreted this provision to refer to the partnership itself as the adviser's "client," rather than to the investors in the hedge | funds.82 All advisers are, however, subject to an SEC rule banning fraud,83 judicially interpreted to require full disclosure to clients. More importantly, private funds are by definition private and cannot be sold to the public g without registering under the ICA.

Common Trust Funds Common trust funds maintained by banks and sav- .

ings and loans (S&Ls) generally function identically to regulated funds, ex- |

cept that they pool assets committed to banks in their capacity as trustees 0

and fiduciaries, rather than directly from individuals (Frankel and Schwing 3

2001, 6-117). As noted previously, common trust funds cannot be adver- £

tised or directly marketed to the public without registering under the ICA, t

and the SEC's interpretation of the ICA is that a bank cannot distribute 3

its trusts to the public as a form of investment. However, a bank can advertise and market its trust services generally, including as a component its capability to invest funds entrusted to it, and it can then collectively invest entrusted funds, achieving the economies of scale that pooling permits. Although national banks are required to provide a written plan to customers who request them, bank common trusts are not subject to an affirmative disclosure regime equivalent to the 1933 and 1934 Acts, from which they are exempt (Freeman 2004; Lybecker 1973).84 As trustees, banks

81 Advisers Act § 203(b).

82 In 2005, the SEC adopted a rule requiring advisers to "look through" their funds in counting their clients, with the effect of requiring the registration of most hedge funds, but in 2007, the rule was struck down as an "unreasonable" construction of the Advisers Act (Goldstein v. SEC, 451 F.3d 873 [D.C. Cir. 2006]).

83 Advisers Act § 206.

84 The Office of the Comptroller of the Currency's rule governing common trusts, 12 C.F.R. § 9.18, applies to national banks and is incorporated into Internal Revenue Code provisions governing trusts. Thus it applies to state banks as well (Freeman 2004, 16).

and S&Ls are subject to fiduciary duty standards that are generally as strict or stricter than those applied to investment advisers. These have the effect of discouraging direct conflict of interest transactions, but indirect conflicts—coinvestments, investments in bank customers, directed brokerage—are subject to considerably less scrutiny than is true for mutual fund complexes. The bank and S&L sponsors of trusts are supervised and examined by one or more federal and/or state bank regulatory agencies, such as the Federal Reserve Board, Federal Deposit Insurance Corporation, or Office of Thrift Supervision. They are also subject to a variety of laws and regulations, which in general terms are designed to safeguard the safety and n

soundness of the institutions. However, the focus of that supervision is not on assets held in trust for others but on deposits and the FDIC's guarantee of those deposits. Office of the Comptroller of the Currency (OCC) regula- g

tions constrain the capital structure of any given trust,85 but they do not require common trusts to have fees annually reviewed or approved by the .

equivalent of independent directors. They are generally free from bright- |

line restrictions equivalent to the ICA's constraints on conflicts of interest, 0

leverage, diversification, or performance fees. |

Commodity Pools and Commodity ETFs Formally, commodity pools and 97 r

commodity ETFs are not regulated, as such, but their sponsors (or op- t

erators) and advisers must register with Commodity Futures Trading i

Commission (CFTC) and the National Futures Association (NFA). An operator or adviser must keep specified books and records and make detailed disclosures to investors and the CFTC, including disclosures regarding fees. These disclosures are under both the Commodity Exchange Act of 1936 (CEA), and, for commodity pools sold to the public, the 1933 Act and the 1934 Act (Markham 1987; Horwitz and Markham 1983).86 As with REITs, commodity pools can be organized as partnerships and can go public without incurring corporate taxation. Thus, although the CEA generally requires an organizer or adviser to organize each pool as a separate legal entity, no restrictions exist on the capital structure or governance of a pool. Pool operators and advisers are generally subject to general anti-fraud standards banning deception, unauthorized trading, misappropriation, churning, breach of a duty of care in handling an account, and limits on the size of speculative positions in the commodities markets. Consistent with the

85 12 C.F.R. § 9.18(b)(3) requires equal proportionate ownership by all trust unitholders.

86 7 U.S.C. §§ 1 to 24 (1996). See generally Markham (1987); Horwitz and Markham (1983).

CFTC's reputation as a "principles-based" regulator, however, it has no detailed rules banning conflicts of interest,87 performance fees, constraints on leverage, diversification, or non-standard investments.88

If the commodity pool is sold to the public, the CEA, the 1933 Act and the 1934 Act all apply. If the commodity pool is not sold to the public, the CEA applies, but not the other two acts. Feel free to re-write as you wish, providing the substance is clear and correct.

Broker-Dealer Products SMAs and ETNs are subject to regulation, as they are typically offered by broker-dealers registered under the 1934 Act and subject to supervision by the SEC and FINRA.89 ETNs sold publicly are |

subject to disclosure requirements under the 1933 and 1934 Acts. In many respects, broker-dealer regulations are similar to the ICA: They forbid fraud, require disclosure and reports, provide for SEC examinations, re- g

quire preservation of books and records, and are subject to general duties designed to benefit their customers.90 For SMAs to be exempt under the .

ICA, brokers must provide for the right to withdraw assets, equivalent to |

redeemable shares of mutual funds.91 SEC and FINRA rules impose soft o

limits on broker fees and spreads,92 which are similar to the annual review a

of advisory fees by fund boards. FINRA also prohibits its members from r

receiving performance fees except as permitted to advisers to wealthy in- t

vestors under the Advisers Act.93 In a few respects, broker regulation goes i

farther than the ICA, requiring broker inquiry into investors' financial positions and sophistication,94 pass professional exams, obtain licenses from

87 Indeed, the annual report of one public commodity pool candidly notes, "The terms of the Management Agreement are not the result of arms-length negotiations and may not represent the best terms that could be obtained from a trading adviser that is not a related party" (Commodity Trend Timing Fund II Form 10-K [Dec. 31, 1996], available at Archives/edgar/data/702655/0000950115-97-000486.txt).

88 Of course, commodity pool investments are limited in the sense that if it invests more than 40% of its assets in securities, it will be required to register under the ICA.

89 1934 Act § 15. About 500 firms are both registered broker-dealers and registered investment advisers under the Advisers Act (Hung et al., 2008, xiv).

90 For example, NASD Rule 2320 (best execution).

91 SEC Rule 3a-4.

92 NASD Rules IM-2440-1 (5% mark-up guideline); IM-2310-2 (ban on excessive trading in discretionary accounts).

93 NASD Rule 2330(f). See note 105 supra (SEC Rule 205-3).

94 NASD Rule 2310 (suitability rule applicable to broker-dealers); SEC Rule 3a-4 (inquiry and advice based on customer's situation and objectives required for safe harbor from ICA).

self-regulatory organizations (SROs), and, for SMAs relying on the SEC's ICA safe-harbor, make personnel available for consultation, pass through voting, and other rights.95

In most respects, however, they are less restrictive: Brokers selling ETNs, 100 for example, effectively borrow from their investors, exposing them to direct and undiversified credit risk, a risk generally not involved in mutual funds or ETFs. Brokers need not treat all customers equally and may vary contracts and fees for identical SMAs or ETNs. Although brokers must disclose conflicts of interest and third-party compensation (in general terms) and are subject to anti-fraud standards, they are not subject to a strict ban |

on conflict transactions, as are mutual funds. The fee or other disclosures required of brokers are not as detailed as and comparable to those required of mutual fund and their advisers. Brokers commonly omit fee tables in g

written materials mailed to customers, for example (Hung et al. 2008, 72). Although a broker must provide customers with a copy of Part 2 of its .

SEC Form ADV, these forms are not reviewed prior to use as are mutual |

fund prospectuses under the 1933 Act. They usually set forth fee ranges O

and vague descriptions of portfolio selection criteria, rather than actual |

fees or portfolio composition, which are supposed to vary from customer r

to customer. Nor do Forms ADV disclose standardized returns or after-tax t

returns, reducing the ability of customers to compare SMAs, or to com- i

pare SMAs with funds, even if the SMAs are in fact quite standardized in practice.

Summary This brief comparative review, summarized on Table 4, shows 101 mutual funds are much more heavily regulated than their closest equivalents (hedge funds) but are able to be sold to the public, as the quid pro quo for heavy regulation. Bank common trust funds, which can receive funds indirectly from but cannot be marketed or distributed to the public, are subject to significant regulation. However, this regulation is less burdensome and detailed than the ICA as applied to common trust funds themselves, particularly as regards the ICA's detailed rules on diversification, leverage, indirect conflicts of interest, fund governance, and fees. Other collective investments sold to the public, such as SMAs, ETNs, and commodity pools, are subject to significant regulation, comparable to the ICA in some respects, such as ongoing disclosure and anti-fraud. The regulation is significantly less intrusive and constraining in many other respects,

95 SEC Rule 3a-4.

Table 4. Comparison of Regulation of Mutual Funds and Other Collective Investments

Mutual Funds and Registered ETFs Private Funds Bank Common Trusts SMAs and ETNs Commodity Pools and ETFs

Sold to public Yes No No (although trust services may be) Yes Yes, although most pools are not

Mandatory disclosure Yes No, except as induced by anti-fraud standards No Yes, but more limited Yes

Anti-fraud standards Yes Yes Yes Yes Yes

Books and records Yes No Yes Yes Yes

Regulatory registration and oversight SEC No Bank regulators SEC CFTC

Board with independent directors Yes No No No No

Regulation or negotiation of fees Independent directors No No FINRA regulation No

Performance fees No, except symmetric fees Yes, standard Yes No, except for wealthy clients Yes

Equal treatment of investors Yes No No No No

Detailed ban on conflicts of interest, including co-investments Yes No No, subject only to fiduciary duties No No

Table 4. (Continued)

Mutual Funds and Registered ETFs Private Funds Bank Common Trusts SMAs and ETNs Commodity Pools and ETFs

General fiduciary duty standards Yes Yes Yes No, unless account is discretionary Yes

Restrictions on capital structure No No Yes (equal ownership) No No

Restrictions on leverage 300% asset coverage No Only as required by fiduciary duties No, except net capital rule No

Restrictions on investments Yes No Only as required by fiduciary duties and trust document Only as required by suitability requirements Commodities only (including futures on broad-based market indices)

Detailed rules on redemptions Yes No Yes, but less detailed than ICA and SEC rules No, but safe- harbor for SMAs; client must be able to withdraw assets No

Minimum capital $100,000 None Not for trust fund, but bank sponsor subject to detailed capital rules Net capital (liquid assets) equal to the greater of $250,000 or 2% of aggregate debit items None

particularly the ICA's mandatory rules on fund governance, capital structure, fee negotiation, equal treatment, prohibitions on conflicts of interests and performance compensation, and constraints on leverage.

Of course, the fact that different collective investments are regulated differently, or much less proscriptively, does not directly tell us which regulatory scheme is better. Perhaps the ICA, as implemented by the SEC, should be modified, or perhaps other collective investments should be regulated in ways more similar to the way mutual funds are regulated. But these comparisons do suggest that the SEC should be more willing to be flexible in permitting variation in how specific mutual funds are governed. The comparisons also suggest that n

lawmakers and SROs generally (Congress, the SEC, the CFTC, bank regulators, and FINRA) should consider whether common regulation in certain areas, such as fee disclosure to the public, would be an improvement to permit inves- g

tors to compare similar collective investments across regulatory regimes.

H. Comparison of U.S. Fund Regulation with Major Competing Jurisdictions f

Although the U.S. remains the world's largest domicile for collective invest- o

ments (see Figure 2), it has for several years been growing less rapidly than |

funds in Europe, Asia, and Latin America. Fund regulation is too complex r

to permit detailed, point-by-point comparisons, but U.S. fund regulation t

can be usefully compared in general terms with regulation of collective in- i

vestments in major competing jurisdictions. It should be stressed that the comparison is purely theoretical. Due to U.S. tax law, discussed in Part I, there has to date been no head-to-head competition between U.S. and European funds and fund regulation, and U.S. investors have not been pulling assets out of U.S. funds and putting them into foreign funds. Strong claims about falling U.S. share of worldwide fund assets should also be tempered by the fact that funds have been steadily increasing their role in U.S. capital markets since the 1960s, whereas in many E.U. member states, funds made significant in-roads on banks and insurance companies only starting in the 1990s and still hold less than half the share of E.U. household assets as that held by U.S. funds.96 Nevertheless, such a comparison can be useful to

96 Currently, funds hold 11% of E.U.-wide household assets, compared to 23% for the U.S. Comm. Eur. Communities (2008); Bucks et al. (2006, A10, Table 4); and Bucks et al. (2006) report that U.S. households surveyed in the 2004 Survey of Consumer Finance held 15% of their financial assets in non-MMMF mutual funds and another 32% in retirement accounts. Assuming that, as reported in ICI (2008a, 99), 26% of U.S. households' retirement assets are held in mutual funds, the total held by U.S. households directly or indirectly in mutual funds would be 23%.

• Identify features of U.S. regulation that are unusual, potentially unnecessary or inefficient

• Suggest modes or methods of regulation as guides in reconsidering U.S. fund regulation

• Roughly assess U.S. fund regulatory success by examining those jurisdictions in which the largest and fastest growing fund industries outside the U.S. are located

The analysis focuses on those jurisdictions in the E.U. (Luxembourg and Ireland) that have been successfully competing directly against other fund domiciles within the E.U.

Assets, Growth, and Market Structure Data for the major fund domiciles are reflected in Figures 5-7.97 Most generally, the data show functioning fund regulatory regimes exist in many countries, that investors have enough confidence in those regimes to be committing significant resources in recent years, and that countries vary significantly in the rates at which resources are being invested in funds, even within the E.U.

In assets, the largest foreign fund domiciles are Luxembourg, China/ Hong Kong, France, Australia, and Ireland (Figure 5). In recent growth rates, the U.S. ranks tenth (Figure 6).

The fastest growth in funds has occurred in countries with rapidly developing financial markets (China and Brazil98) or countries that have invested in becoming fund service centers. The UK, a major center for financial services and fund management, is less dominant as a fund domicile; its share of total worldwide fund assets has been constant (at 3-4%) since 2000 (ICI 2008a, 157). France has a long-standing fund industry that is domestically managed and domiciled; its share of worldwide assets has grown modestly, from 6.1% in 2000 to 7.6% in 2007 (ICI 2008a, 157). Germany's fund industry lost assets to Luxembourg in the late 1980s and early 1990s after it adopted a withholding tax—German banks sponsored funds just over the Grand Duchy's border and sold them back to German taxpayers (Bruce 1993). Its share of worldwide fund assets has continued to shrink, from 2.0% in 2000 to 1.4% in 2007.

<¡1 t U

97 For a review of mutual funds worldwide, see Khorana et al. (2005).

98 Brazil is the dominant fund domicile in Latin America, with roughly 84% of all fund assets, mostly domestic (ICI 2008a).

Figure 5. Mutual Fund Assets, by Country, 2007 ($ billions)

Source: ICI 2008 Factbook.

Figure 7. Mutual Fund Assets/Stock Market Capitalization, by Country (2007)

Sources: ICI 2008 Factbook; World Federation of Exchanges.

Luxembourg and Ireland, in particular, stand out as fund exporters, whose funds hold more assets than their domestic securities market (Figure 7).99 From 2000 to 2007, Luxembourg has seen it share of worldwide fund assets almost double, from 6.3% to 10.2%. It is now the second-largest fund domicile in the world, after the U.S. Ireland's share has tripled, from 1.2% to 3.6% (ICI 2008a, 157). Funds in these jurisdictions are primarily sold to foreign investors, hold foreign securities, and are managed by foreign advisers. Ireland and Luxembourg also account for most assets held by funds sold outside their domiciles. (Again, because of U.S. tax law, discussed in Part I, U.S. funds are not sold to foreign investors [Khorana et al. 2005, Table 1].)

"Home bias" (the tendency to invest, consume, trade, and finance locally) is no doubt an important reason that domestic funds continue to dominate most countries' fund markets and the reason that countries such as China and Brazil are experiencing rapid growth in their domestic funds' assets (French and Poterba 1991; Lewis 1999; Sarkissian and Schill 2004; Rauch 1999). However, home bias at the fund level has not been strong enough to prevent Ireland and Luxembourg from rapidly growing their largely exported market shares of global assets under management.

<¡1 t U

99 As another benchmark, in 2007, Luxembourgian funds hold $5.5 million per resident, Irish funds hold $220,000 per resident, and U.S. funds hold $40,000 per resident.


European Fund Regulation: UCITS Member states of the E.U., Ireland, and Luxembourg are subject to dual regulation, E.U. and national. However, due to harmonization of E.U. law through the UCITS100 directives, funds established in any E.U. member state are, if properly structured, freely saleable in theory throughout the E.U. (passported).101 Although hedge funds and other alternative funds are a non-trivial part of the E.U. fund industry, UCITS-compliant funds have held a steady 75-80% of total fund assets over the past several years, even as European funds have grown rapidly (Figure 8). In Luxembourg, UCITS funds represent an even higher share of fund assets (90% in 2007) (ALFI 2007, 25). Funds based in the U.S. and n

elsewhere outside the E.U., by contrast, must comply with each separate European nation's laws to be distributed in Europe.

Partly because of UCITS's regulatory advantage in the E.U., U.S. firms g

are the leading sponsors of funds in Ireland and Luxembourg, where they often clone U.S. funds for sale to European investors (for instance, create .

UCITS-compliant E.U.-based funds that hold investments closely simi- |

lar to their U.S. fund counterparts). As noted in Part I, these clones never 0

precisely track their U.S. counterparts, and do not create the same econo- |

mies of scale as would funds that could be marketed in and out of the U.S. r

U.S. sponsors, such as MFS, were the largest source (~18%) of assets in t

Luxembourg funds in 2006, followed by funds promoted by Swiss (~17%), i

German (~16%), and British (~10%) firms (AFLI 2007, 32). U.S. sponsors, such as Vanguard, are even more dominant in Ireland, where they sponsor a steadily increasing share of funds (44% in 2007), followed by the UK (34%), Italy (3%), and Germany (2%) (Irish Funds Industry Association 2008). The UCITS directives, by opening up the E.U. to funds based anywhere in the E.U., have, perhaps unintentionally, helped open up the E.U. market to U.S.-sponsored and U.S.-managed but E.U.-domiciled funds.

100 UCITS stands for "undertakings for collective investment in transferable securities." The two E.U. directives are 2001/107/EC (the "Management Directive") and 2001/108/EC (the "Product Directive") (1/21/02) [together referred to as "UCITS-III"] and 85/611/EEC (12/20/85) ["UCITS-I"]. (There is no UCITS-II; efforts to revise the UCITS-I in the early 1990s were abandoned. Ernst and Young, ECITS III - A Practical Guide (9/2003), at 2.) In application to fund managers, UCITS-III cross-references the Investment Services Directive, 93/22/EEC (5/10/93), which was replaced by 2004/39/EC (4/21/04), effective 11/1/07, the Financial Instruments and Markets Directive.

101 Implementation of the UCITS Directives continues, with ongoing consideration of another directive (UCITS IV) that would facilitate the passporting of management companies, rather than just funds, as well as cross-border fund mergers and asset transfers. See Bucks et al. (2008, 75).

Figure 8. Growth in and Share of UCITS, 2002-2007 (billions)


Source: EFAMA State of European Investment Fund Industry (3/08).

The E.U. passport, home bias, or the comparative maturity of the U.S. fund industry are not the sole explanations for the declining U.S. domiciled funds' share of fund assets, however. The E.U. passport and home bias provide UCITS-compliant funds with no advantages in other parts of the world, such as Asia and Latin America, where UCITS has been growing its market share in the past several years.102 The comparative maturity of the U.S. fund industry would, if anything, provide it with an advantage in selling into new markets, but it is hindered by U.S. tax law (discussed in Part I) and, to a lesser extent, by U.S. securities regulation (discussed in Part II.B).

Comparing U.S. Regulation with E.U. Regulation In the E.U., the UCITS Directives set minimum standards for publicly sold (retail) collective investments addressing diversification, authorization, structure, permissible activities, and disclosure.103 Appendix D to this report compares those standards, as implemented in Ireland and Luxembourg, with the ICA (Coates and Hubbard 2007; Frankel and Schwing 2001; Rounds

€ 9,000

€ 8,000

€ 7,000

€ 6,000

€ 5,000

€ 4,000

€ 3,000

€ 2,000

102 Asia-based assets now represent roughly 15% of all assets in E.U.-based UCITS funds and 6% of all fund assets. In Latin America, 3% of all fund assets are in E.U.-based UCITS funds

(EFAMA 2008b, 3; ICI 2008a, 157; KPMG 2007, 15; Johnson 2006, 3).

103 Formally, the UCITS Directives are not law applicable to funds, but are part of the treaty apparatus that constitutes the E.U. and apply to member states, requiring them to implement its contents at the national level. Member states have some flexibility to not implement a directive in every detail.

and Dehio 2007; Schonfeld and Kerwin 1993; KPMG 2008a; KPMG 2008b; PriceWaterhouseCoopers 2007a; PriceWaterhouseCoopers 2007b; UCITS Directives; Thompson and Choi 2001; IOSCO 2006; Selman 1992; McGeough and Quirke 1992; Jackson and Counihan 2008; Partsch, Terblanche and Malaniuk 2008). The Directives allow for different methods of achieving its minimum standards, and countries may subject their domestic funds to more or less rigorous requirements.104 UCITS-compliant funds are passported, meaning that funds organized under the laws of one E.U. nation must only comply with the marketing, advertising, and tax laws of another member nation in which they are doing business. The effect of n

this passport is that the minimum standards set out in the UCITS Directives are de facto general standards. Member states continue to experiment with alternative fund regimes, which do not adhere to UCITS minimums, but g

sponsors can sell shares in those alternative funds only in the home country and those host countries that exempt or affirmatively authorize the alterna- .

tive fund (for example, a hedge fund sold only to institutional or qualified |

purchasers). Because of this and because the data reviewed previously show 0

that UCITS-compliant funds continue to dominate the E.U. fund market, |

the focus here is on the UCITS framework. r

As reflected in Appendix D, UCITS and the ICA are similar in many t

respects: £

• Each specifies disclosures to the public and provides for regulatory review

of those disclosures. g

• Each requires at least one third party with obligations to oversee fund r

operations on behalf of investors. O

• Each requires securities to be held by a custodian separate from the U adviser. g

• Each imposes constraints on leverage, diversification, and non-standard 0 investments.

• Each imposes either regulatory or board/shareholder approval requirements on changes of advisers or advisory contracts.

104 Irish funds include variable capital companies (VCCs), equivalent to U.S. mutual funds organized as corporations, and unit trusts, equivalent to a U.S. business trust. Luxembourg funds include Sociétés d'investissement a capital variable (SICAVs), equivalent to U.S. mutual funds organized as limited liability companies (~44% of funds, by number), and fonds commun de placement en valeurs mobilieres (FCPs), similar to a U.S. tenancy in common (~55% of funds, by number) (AFLI 2007).

• Each permits funds to be established without requiring separate legal entities.

• Each permits multiple share classes.

Each provides for ongoing supervision and enforcement by a public agency. 115

• Each permits active and passive management; equity, bond, and hybrid investments; investments in derivatives; and, with restrictions, funds of funds (including funds of hedge funds).

True, significant differences do exist. One overarching difference in the approaches of the ICA and UCITS is conceptual. In the E.U., the adviser or management company is the focus of regulation and control, and funds are viewed as (and often are) simply products of the sponsor, which generally owns the management company. In the U.S., the funds are the focus of regulation and control, and are viewed (by lawyers, at least) as separate companies, whereas investment advisers as such are lightly regulated. That conceptual difference shows up in a number of areas, most prominently in fund oversight. U.S. funds are overseen by boards of individual directors elected by fund shareholders, and fund boards are similar in design (although not function) to boards of industrial companies. In Europe, funds are typically overseen by a combination of an institutional trustee or custodian and the board of individual directors of the management company. However, the conceptual difference also shows up in other areas:

• In the U.S., funds must have minimum start-up capital; in the E.U., management companies must meet minimum capital requirements.

• In the U.S., funds must have compliance policies; in the E.U., management companies must have risk management and compliance policies.

• In the U.S., funds are subject to extensive conflict of interest rules that apply by extension to the advisers; in the E.U., management companies are limited in what other activities they can undertake.

Any simple comparison of fund regulation on its own, without comparing 117 manager/adviser regulation, would thus be misleading.

A second conceptual difference is that the UCITS model preserves a 118 central role for the member states in the E.U. Regulators in Luxembourg and Ireland, for example, are the primary supervisors and enforcers of the UCITS rules for cross-border funds, not a regulator in Brussels (Selman

1992).105 In the U.S., the SEC has largely displaced the states as the regulator of mutual fUnds and their advisers. More specifically, the ICA is tighter in four major respects and the UCITS regime is tighter in six (Table 5). 106

In the U.S., private rights of action and frequent class and derivative actions supplement public enforcement of fund regulation (Benedict et al. 2008; Murphy and Bassel 2005; Benedict et al. 2004; Coates and Hubbard 2007). More generally, the heavily litigious environment in the U.S. leads fund sponsors to draft disclosure in a defensive way and leads the SEC to move more slowly and cautiously than it otherwise would in its rule-making.107 In the E.U., private shareholder litigation is rare, fund investor litiga- | tion even more so, and the litigation environment is less fraught, although this might be beginning to change (Armour et al. 2007; Dougherty 2006). The ICA tightly constrains conflict transactions and the form of adviser g compensation, essentially banning performance fees. In the E.U., conflict transactions are constrained only by adviser obligations to deal fairly and . at an arm's-length basis with their funds. Performance fees are common | (Kruithof 2005), but the E.U. more tightly constrains funds in terms of 0 diversification, non-standard investments, and leverage (Bucks et al. 2006, | 12). In the U.S., fund boards and shareholders must annually approve r changes in advisers, advisory contracts, and adviser compensation. In the t E.U., regulators approve such changes, and although no annual review is i required, regulators are obliged to make substantive findings, such as that the management company and depositary directors are of good repute and have sufficient experience and competence to carry out their jobs.108 The U.S. imposes no significant formal requirements of local agents, whereas

105 The UCITS rules are also minimum standards; member states may and some do apply higher standards to funds domiciled in those states (Selman 1992). However, because Ireland and Luxembourg adhere closely to the UCITS minimums and because a UCITS-compliant fund can be sold throughout the E.U., the UCITS standards are also effectively the E.U. standards for cross-border funds.

106 U.S. regulation requires more extensive disclosure (Paul 1992, 86; SEC 1990, n. 28).

107 For example, Goldstein v. SEC, 451 F.3d 873 (D.C. Cir. 2006) strikes down a SEC rule requiring registration of hedge fund advisers under Advisers Act); Financial Planning Assoc. v. SEC, No. 04-1242 (D.C. Cir. Mar. 30, 2007) strikes down a SEC rule exempting broker-dealers from Advisers Act despite receiving "special compensation" if "incidental" to brokerage; PAZ Securities, Inc. v. SEC, No. 051467 (D.C. Cir. July 20, 2007) strikes down a SEC order affirming expulsion of a NASD-member firm and barring its president from the securities industry for failing to comply with various examination requests.

108 UCITS Directives Art. 4, 5.

Table 5. Summary of Significant Formal Differences Between U.S. and E.U. Fund Regulation

U.S. is "Tighter" E.U. is "Tighter"

Private litigation enforcement Minimum capital requirements

Constraints on performance-based compensation for adviser Tighter restrictions on leverage, diversification, and non-standard investments

Annual board/shareholder approval of renewal or changes in adviser, compensation, and advisory contracts Regulatory approval of fees or for changes in adviser, compensation, or advisory contracts

Extensive ban on direct and indirect conflict transactions Activity restrictions on management (advisory) companies

Discretionary regulatory approval for new funds

Local economic subsidies (such as a mandate to rely on local employees)

formal local economic subsidies are accepted in the E.U. Although both the U.S. and E.U. impose minimum capital requirements, the E.U. requirements, which are imposed on the management company (and not the fund, as in the U.S.), are more stringent.

It is unclear how practically important these formal statutory differences are. Although private litigation generates costs for U.S. funds, these costs are not large when measured against the vast scale of the U.S. fund industry. U.S. fund advisers can and do pay performance-based compensation to individual portfolio managers, and the link between fund in-flows and performance is strong enough to provide advisory firms with an form of performance compensation, albeit one that is indirect and weaker than would be the case if performance fees were permitted in the U.S. (Barber, Odean, and Zheng 2005; Chevalier and Ellison 1997; Edelen 1999; Gruber 1996; Ippolito 1992; Patel, Zeckhauser, and Hendricks 1994; Sirri and Tufano 1998; Baumol et al. 1990). Weaker investment and diversification constraints matter less if funds are constrained by tight conflict-of-interest standards. Vice versa, a simple arm's-length standard for conflict transactions is easier to enforce if a fund may only invest almost exclusively in securities with verifiable market values. Regulatory approvals for changes in compensation and contracts are not likely to matter much if the regulatory officials can be counted on to act quickly and flexibly. Board and shareholder approvals for such changes in the U.S. can fairly be seen as direct substitutes for such regulatory approval in the E.U. The detailed, proscriptive nature of the ICA, with its extensive

<¡1 t U

ban on even indirect conflict transactions, partly substitutes for no formal requirement of regulatory approval for new funds, as most innovative funds require an exemptive request. Local subsidies are in the first instance simply redistributive and only affect the overall efficiency of a fund if the local service providers do not face competition. With both Ireland and Luxembourg offering much the same bundle of services, it's unlikely that these subsidies produce a large negative efficiency effect. Minimum capital requirements can be a real barrier to entry, which is reinforced by merit or reputational clearance of individual management company directors by regulators in the E.U., but there is no direct evidence on the effect of the E.U.'s requirements on n

management company start-ups.

Nevertheless, the fact that the U.S. is an outlier in a number of respects— both more and less permissive—should spur Congress and the SEC to re- g

evaluate the current U.S. fund regulation. In Part III, specific suggestions are made about two rules:

The U.S. ban on asymmetric incentive compensation

The U.S. insistence on the ICA as the sole authorized regulatory scheme and

the SEC as the sole authorized regulator of mutual funds sold in the U.S.

Detailed empirical country comparisons of funds, fund costs, and fund t

regulation are few. In one of the few studies to date, Khorana et al. show i

that fund fees are, on average, both higher and lower for funds domiciled in the U.S. than for funds domiciled in Luxembourg and Ireland, depending on the measure (Khorana et al. in press). Their findings also suggest that, without making assumptions about investor holding periods and the number and identity of countries in which a fund is sold, it is hard to make strong assertions about overall fund fees in either the U.S., on the one hand, or European jurisdictions, such as Luxembourg or Ireland, on the other hand. When Khorana et al. examine the effect of regulation specifically, they find that, across all countries, having a regulator approve a fund's disclosure (as in the U.S.) is correlated with lower fees, however measured, but that having a regulator approve both a fund and the fund's disclosure (as in Luxembourg and Ireland) is correlated with even lower fees. They also find that management fees are lower in countries that require independent custodians and constrain conflicts of interest, but in their data (which they derive from informal OECD and the IOSCO surveys), the U.S., Ireland, and Luxembourg do equally well on these measures.

They do report two findings with implications for U.S. regulation. First, 123 they find that an additional small but non-trivial average fee of 2-6 basis points is associated with funds sponsored by foreign advisers, as is the case with all U.S.-sponsored clone funds sold in the E.U. This finding suggests that any back-up regulation provided by the fact that U.S. sponsors (such as Vanguard or Fidelity) would be unlikely to engage in activities through their European funds that would be in violation of fundamental regulations in the U.S. does not provide any marginal benefit to their E.U. investors in the form of lower fees. This finding also suggests that removing tax and other regulatory barriers to sales of U.S. funds overseas could provide a n

meaningful benefit to both the sponsors of and investors in those funds by eliminating the need to incur duplicative regulatory costs in both the U.S. and the E.U. Second, they find that total expense ratios are correlated with g

set-up costs for new funds. As those set-up costs are largely a function of legal and regulatory costs and because the U.S. fares less well on this mea- .

sure than its competitors,109 this finding has potential implications for U.S. 0

regulatory reform. 0

Comparison of Regulatory Incentives, Procedures, and Resources in the U.S. 124 |

and Europe After tax, the biggest difference between regulation in the U.S. r

and the E.U. is not a matter of formal statutory law but practical statutory t

law. The differences between the U.S. and E.U. regimes lie in two differences i

at the regulatory level. The first difference is that in the E.U., the UCITS statute provides a minimum set of rules to which all member states must comply, but the amount and nature of detailed implementing regulation is left to supervisors in each member state. The states have incentives to compete with each other to attract funds to organize in their jurisdictions. In the U.S., the ICA is implemented by a subunit of a single agency (the SEC's IM) with no competitive rival for new mutual funds. The IM has adopted detailed rules and policies that, over time, have significantly increased the restrictiveness and rigidity of the ICA. The second difference is that in the E.U., relative regulatory resources are considerably greater,

109 Analysis of data from KPMG (2008a and 2008b) shows set-up costs in the U.S. are 2-3 times

higher than the average of the other top 10 fund domiciles for which KPMG reports information (such as the UK, France, and Australia), and higher than any other top 10 domicile. KPMG does not report set-up costs for Ireland or Luxembourg. Franks, Schaefer, and Staunton (1997) shows that ongoing U.S. mutual fund regulatory costs per employee are half those in the UK, but twice those in France.

which permits a much greater speed of regulatory clearance for new fund set-up or exemptive requests or interpretations of existing rules.

The success of Ireland and Luxembourg within the E.U., after all, cannot be attributed to formal E.U.-level statutory framework, as UCITS creates the potential for a level playing field for funds based in all E.U. member states. Although both nations' labor markets and tax regimes played important roles,110 much of the fund-exporting nations' success can fairly be attributed to the flexible, innovative, and promotional approach their regulators have taken with regard to collective investments. Both nations' regulators are generally viewed as responsible and protective of investors |

while well disposed toward fund managers and promoters. In particular, they have been willing to make accommodations for the fund industry in a fairly rapid fashion where investor protection was not put in serious jeop- §

ardy. Ireland was the first country in Europe to authorize ETFs, was one of the first to authorize MMMFs, and in the 1990s, established a flexible .

set of regulatory tiers for different types of hedge fund investors and fast- |

track approval for those reserved to the wealthiest investors, thus emerging 0

as the leading hedge fund domicile in Europe (Neylin 2004). Luxembourg |

was the first E.U. state to implement UCITS-I in 1988 ( r

index.php?id=27) and updated its law to reflect UCITS-III in less than a t

year.111 As or more important, the E.U. regulators are constrained by com- i

petition from adding substantial regulation or costs beyond that required §

by the UCITS directives. By comparison, the SEC has little direct incentive o

not to add regulations or costs if it is politically expedient to do so. U.S.- e

based mutual funds have no choice but to register with the SEC, non-U.S. <§

funds cannot sell shares to U.S. investors, and the SEC derives no additional §

110 Khorana et al. (2005) attribute the success of Luxembourg to strict bank secrecy laws, tax laws §§ favoring investors (relative to neighboring Germany), and its central location. They attribute 0 the success of Ireland to laws favoring the investment industry and an educated workforce. 5 Both, they also note, benefit from E.U. harmonization. They also study the success of countries in attracting assets to their mutual fund domiciliaries and find that larger fund market

shares correlate with a stronger judiciary (or accounting standards), regulatory approval requirements for a new fund and/or its disclosures (but lower set-up costs), more fee-related disclosures, weaker insider trading laws, higher GDP/capita, and a more educated/literate population. They exclude Luxembourg and Ireland from their regressions because their success in attracting fund assets from other countries makes them qualitatively different from other countries (Table 4). Independent custodians and their measure of conflicts regulation, derived from Thompson and Choi (2001), does not correlate with a country's fund industry size.

111 See The Law of 20 December 2002, available at Luxembourg/Legal%20and%20Regulatory%20Environment/Lux%20loi%200PC %20201202%20FR.pdf, implementing UCITS-III.

revenue from the registration of new mutual funds because the fees associated with new mutual funds are paid over by the SEC to the U.S. treasury and only allocated back as part of a budget process in which the SEC's IM plays a relatively small role.

In contrast to the U.S., the governments of Ireland and Luxembourg 126 more generally view the fund sector as an important priority. As with Delaware in the U.S. with respect to corporate law (Romano 1993), the employment and tax revenues provided by the fund business generate keen incentives to both preserve and extend the fund regulatory franchises they have developed for Ireland and, even more so, for Luxembourg, with |

one fund for every 50 residents. Corporate franchise taxes provide 7% of Luxembourg's tax revenues, and taxes on income of individuals in the professional service and investment management industries provide another g 7% of tax revenues, 6% of the employment, and 4% of the GDP (Com. Dév. Place Fin. 2007). In Ireland, the national government invests substan- . tially in the IFSRA, funding roughly 50% of its expenses out of general tax | revenues (Ireland Financial Services Regulatory Authority 2007). In the O U.S., by contrast, the SEC generated 1.75 times as much revenue as it was | allocated by Congress, with the excess subsidizing the general budget of r the U.S. (see SEC 2008a, 5). 1

Both Ireland and Luxembourg have frequently developed and pursued 127 i

legislation allowing for non-UCITS funds and addressing difficulties in pre-existing law. Ireland, for example, eliminated the risk of cross-liability among sub-funds not separately organized as distinct legal entities112. In Luxembourg's 2002 law implementing UCITS III, it included a number of provisional "derogations"—reservations of flexibility—from the investment restrictions required by the E.U.-level directive.113 By 2004, both nations' regulators had established new rules governing funds of funds (mutual funds that invest in hedge funds) two years ahead of their counterparts at the SEC, allowing European funds of funds to go public before their U.S. competitors (Der Hovanesian 2006, 13; von Cottier 1997, 67).114 This was

112 Investment Funds, Companies and Miscellaneous Provisions Act 2005, Section 25.

113 See The Law of 20 December 2002 at Art. 49.

114 Fund of Fund Investments, ICA Rel. Nos. 33-8713, IC-27,399, 71 Fed. Reg. 36,640 (effective July 31, 2006) (adopting Rule 12d-1 permitting fund-of-funds).

despite the fact that Congress had given the SEC authority to approve funds of funds in 1996, more than ten years earlier.115

Reflecting these incentives, Ireland and Luxembourg also devote relatively more regulatory resources to their fund industries than does the U.S. As reflected in Table 6, both countries' regulatory staffs dedicated to their fund industry are larger than the SEC's IM, relative to assets under management in regulated funds. The SEC has roughly 1/5 the staff the IFSRA does and roughly 1/7 the staff the CSSF has devoted to the fund industry, per dollars of assets under management.

The SEC also devotes far more resources to enforcement relative to funds 1

than does either of its competitors. The SEC has one person in IM for every 7.5 persons in the Division of Enforcement; in Ireland and Luxembourg, these allocations are reversed, with 5.5 and 7.1 persons supervising funds g

for every one person in their enforcement units, respectively.116 Of course, regulatory resources alone are not sufficient to preserve and extend a regu- .

lated industry—they could just as easily choke one off with excessive regu- |

lation or bureaucratic inertia. Part III returns to the question of regulatory O

design and incentives. a

4. five sets of reforms to enhance regulation of

Five sets of suggested reforms are discussed in this part. The first and most sweeping set concerns the structure and funding of fund regulation and oversight in the U.S. (All of the suggested reforms assume that the fundamental structure of U.S. financial regulation remains largely intact. If current proposals to implement dramatic regulatory change by consolidating the SEC, CFTC, bank regulators, and so forth into a new financial regulatory body are implemented, the suggested reforms would need to be reconsidered, but most should remain generally applicable.) The second set concerns the SEC's

u.s. mutual funds n

115 National Securities Markets Improvement Act of 1996, P.L. 104-290, § 202(4), 110 Stat. 3416, 3427 (1996); H.R. Rep. No. 622, 104th Cong., 2d Sess., at 43-44 (1996). It discusses new § 12(d) (1)(J) of the ICA that gives the SEC exemptive authority under ICA § 12(d)(1), which prohibits a registered investment company from acquiring >3% of any other registered investment company, investing more than 5% of its assets in any other registered investment company, or investing >10% in registered investment companies.

116 The number for the SEC enforcement staff only includes staff in the Division of Enforcement, even though significant staff in the OCIE and IM are devoted to enforcement and compliance functions.

Table 6. Regulatory Resources Devoted to Fund Industry

Total Staff for Funds Fund AUM ($U.S. Bn) (2008) Staff per $1 Trillion AUM Ratio of Funds Staff to Enforcement Staff

U.S. 145 $12021 12 1:7.5

Ireland 60 $951 63 5.5:1

Luxembourg 92 $2685 34 7.1:1

Sources: ICI (2008a); SEC (projected for 2009), IFSRA (2007), CSSF (2006).

exemptive process under the ICA and include initiatives that would represent a marked departure from the SEC's approach to the ICA. They have the potential to energize and expand the fund industry, as well as modest improvements that build on IM staff recommendations made more than 15 years ago. The third set proposes new rule-making initiatives on adviser compensation, reciprocity with fund regulators in other countries, and disclosure. The fourth set addresses SEC examination and inspections, and the fifth set is a list of new information and reporting suggestions for the SEC and studies that would inform future regulatory initiatives.

Structure and Funding of Fund Regulation/Oversight

The most important steps are to ensure that the agency (or subunit of the agency) responsible for the U.S. fund industry has the capability to make long-term investments in and commitments to regulatory and supervisory capacity that are

• Required to preserve and enhance the U.S.'s role as a leader in the fund industry

• Required to play the regulatory role created by the ICA, which is distinct from the regulatory and enforcement tasks required by other federal securities laws

• Comparable to those that have already been made and continue to be made by the IM's counterparts in Ireland (IFSRA) and Luxembourg (the Commission de Surveillance du Secteur Financier [CSSF]).

As discussed in Part II, the SEC and IM currently do not have the resources to carry out those tasks. Several possible means can be used to accomplish those goals. The boldest would be for Congress to move IM out of the SEC altogether, either into a new agency modeled on the Federal Reserve Board

or the Public Company Oversight Accounting Board (PCOAB) (Coates 2007), or into a self-regulatory organization (SRO) modeled on FINRA (Seligman 2005). Each of those agencies, crucially, has a secure industry-based funding mechanism outside the general U.S. budget. Having such a secure source of funding would allow the agency to invest in a staff and infrastructure that would enable it to carry out large-scale studies, regulatory reforms, and exemptive experiments. Such a structure would give the IM and the U.S. fund industry the best chance to compete in the long-term.

Less bold, but still potentially effective, would be for Congress to either

move the exemptive and product approval tasks of IM out of the SEC into n

a new agency or SRO, or create a dedicated funding mechanism for IM,

coordinated with increased flexibility in how it carries out some of the

work required for innovative product approvals. Any of the solutions will g

no doubt be controversial and take time to develop and implement. In the

short-run, the SEC should seek and Congress and the SEC should allocate .

funds and staff for IM at least equal to its counterparts at the IFSRA and/ |

or CSSF, appropriately benchmarked by the size of the industry in those 0

countries. 1

Spinning out IM (or the part of IM charged with evaluating exemp- r

tive requests) from the SEC might at first glance seem to cut against the t

grain of current reform proposals that seek to consolidate the SEC and i

other regulatory agencies into one or a small number of financial regulatory agencies with broad scope—similar to the UK's Financial Services Authority. A spin-out of IM is suggested, however, only in light of the overall current regulatory framework. As long as the unit of any consolidated agency responsible for supervising mutual funds and mutual fund advisers has access to a secure funding source and has authority to use the flexibility already built into the ICA, the basic goals of this suggested reform would be achievable. The goal of the reform suggested here is not about the formal overall shape of regulatory organization, but about the reliability of regulatory resources and the authority of those regulatory officials who know most about the fund industry to trade off the costs and benefits of regulatory choices. Currently, neither the SEC nor IM has secure funding, and IM has insufficient delegated authority or independence from political influences to effectively supervise the fund industry. If overall financial regulatory reform were to change these facts, the reason for spinning out IM would fall away.

Any changes along these lines should be coupled with clear direction 136 from Congress that the increased and/or dedicated funding source should be used primarily to pursue innovation and development of funds and fund products and not for enforcement and routine supervision. Both are important goals but both are already much better funded and staffed. Ideally, the direction should also make it clear that whatever the SEC's historic view as to the key provisions and protections of the ICA, it should treat no such provision as sacrosanct and off-limits for exemptive requests or rule-making proposals. The comparisons of alternative regulatory regimes reviewed in Part II should make it clear that few features of the ICA are absolutely n

crucial to a functioning collective investment scheme. It should also be clear that variation and innovation should be viewed as affirmative public goods, worthy of support and regulatory facilitation, provided that regula- g

tory risks associated with innovation are appropriately circumscribed, as discussed more in Part III.B. .

In addition—both as part of the re-focusing of the fund regulator on 137 |

innovation and development and as an independent goal—Congress 0

and/or the SEC should direct that IM staff should be significantly ex- |

panded with economists or policy analysts, ideally not those with long r

prior careers as lawyers. Alternatively, the SEC's Office of Economic t

Analysis (OEA) could be expanded and a unit of OEA seconded to IM. i

The SEC should recruit these new professional staff members from the same pool of top talent that currently is the target of hiring by the Federal Reserve Board or the U.S. Treasury and pay them commensurately, as permitted by the Investor and Capital Markets Fee Relief Act117 and the Accountant, Compliance, and Enforcement Staffing Act of 2003.118 The Capital Markets Fee Relief Act exempted the SEC from general federal pay restrictions and provided the agency with the authority to bring salaries in line with those of other federal financial regulators. The Accountant, Compliance, and Enforcement Staffing Act of 2003 provided the SEC with relief from competitive hiring requirements to expedite the hiring of accountants, economists, and examiners so that the agency could more quickly fill new positions. These new personnel should be involved in all significant new or amended rules and all novel and significant exemp-tive orders. All new rules and significant enforcement initiatives should

117 Pub. L. No. 107-123, 115 Stat. 2390 (Jan. 16, 2002).

118 Pub. L. No. 108-44, 117 Stat. 842 (July 3, 2003).

continue to be subjected to cost-benefit analysis, as recommended by the ICI (ICI 2007a) and already required by law for most other government agencies.119 As discussed later, cost-benefit analysis should be brought to bear on exemptive request denials and approvals.

Exemptive Requests

As discussed in Part II, the SEC's exemptive powers under Section 6(c) are crucial to the functioning of the ICA, particularly now that the statute is nearly 70 years old. Yet the SEC's approach to exemptive relief has remained largely intact since the statute was enacted. Although its procedures have |

been improved somewhat in the past few years, exemptive orders are still difficult and costly to obtain and still take longer to obtain than it takes to start a new fund in the E.U. The SEC should rethink its entire approach to g

Section 6(c).

SEC Approach and Philosophy Most generally, the SEC should abandon .

two shibboleths that are now constraining the effective regulation of the 0

fund industry: 0

• The applicant should bear the burden of persuasively anticipating and ad- S dressing every risk to investors that an innovative exemption presents. /

• The ICA has key elements that are non-starters for those seeking exemptive U relief. v

Nothing in the ICA affirmatively requires either of these principles to stand in the way of SEC exemptions. Although Section 6(c) requires the SEC to make certain findings before granting relief, it has no discussion in it as to who should bear the burden of persuasion on exemptive requests, or as to how plausible a regulatory risk must be before the applicant must propose a solution to the risk, or how certain the proposed solution will be to work. Different ways of addressing risks exist: One is to never take them, which seems essentially to be the SEC's current philosophy under the ICA. Another is to take them, but to take steps to prevent them from causing large or unstoppable harms. Another way of putting this general philosophical point is that there is a large but rarely acknowledged risk in not allowing exemptive requests entailing moderate but controlled risks to

119 See Executive Order No. 12,291, 46 F.R. 13193 (Feb. 17, 1981); Executive Order No. 12,498, 50 F.R. 1036 (Jan. 4, 1985); Executive Order No. 12,866, 58 F.R. 51735 (Sept. 30, 1993); Executive Order No. 13,258, 67 F.R. 9385 (Feb. 26, 2002).

be approved. This risk is that the U.S. fund industry will lose its competitive position, and investors will, over time, choose to opt out of the system altogether by investing directly or through collective investments that fall outside the scope of the ICA. None of this is to suggest wholesale repeal of the ICA and its protections or that the interests of public investors should not be considered. Rather, the current mindset of "do nothing until com-pellingly convinced otherwise" is itself a way of ignoring the interests of public investors. Risk should be balanced against risk, and specific risks can be addressed in more moderate ways than adamant refusal to consider a request until the risks have been completely or nearly eliminated (Sun-stein 2005).

1992 Staff Recommendation and IM Performance Reporting Turning to process and detail, the SEC should, at a minimum, implement the SEC IM Staff's 1992 Redbook recommendation to accelerate orders where precedent exists. The Redbook proposal was 120 days after filing, versus the then average of 190 days, and as proposed by the SEC in 1993, it was 90 days, limited to relief that tracked the most recent relevant order in full.120 Indeed, these time periods seem generous in light of improvements in communication and information technology since 1992. This is reflected in the fact that in 2003 the SEC required registrants under the 1934 Act to speed up their public filings on those grounds.121 Implementation of electronic filings of exemptive applications, as currently planned by the SEC, will facilitate both applications and staff review.

The Redbook recommendation (which was from SEC staff, as opposed to the SEC) came first - it was 120 days; then the SEC itself proposed a rule - it was 90 days, limited to relief that tracked the most relevant order in full.

The proposed rule was withdrawn in 1996,122 based on concerns that IM's staff and resources would be diverted from innovative filings to comply with the amended rule, a concern apparently shared by both industry participants and IM staff (SEC 1996). The SEC had taken a different position in 1993, however, stating that it "believed that while some ... resources [would] be devoted to meeting the time periods imposed by the proposed

<¡1 t U

120 ICA Rel. No. 19362 (Mar. 26, 1993)

121 SEC 1934 Act Rel. No. 46464A (Mar. 3, 2003).

122 See ua/ web_fetch_doc?dataset=ua&db=agendaApril1996& doc_id=x3235-AF56&query=and (last visited July 20, 2008).

amendments, such resources would be minimal because, among other reasons, its staff will be reviewing applications and draft notices that have been marked to show changes from previous applications and notices." Indeed, if the accelerated relief were to be limited to applications that truly tracked prior orders, the black-lining requirement would seem to eliminate the need for review by professional staff altogether. Paralegals or other less expensive staff could carry out the minimal review and verification required. Alternatively, the rule proposal could be modified to eliminate staff review of precedent-based applications, as in a later suggestion. More generally, however, concerns that speeding up exemptive processing will demand n

more resources are better handled with more resources than by keeping the process slow—the goal of the preceding recommendations.

The failure to pursue the staff's 1992 recommendation has had observ- §

able effects in slowing innovation in the U.S. fund industry. After the SEC granted exemptive orders permitting MMMFs to use alternative valuation .

methods (amortized cost and penny rounding) starting in 1977, similar |

orders were subject to significant delays until the SEC adopted Rule 2a-7 in 0

1983 (six years later).123 After granting exemptive relief for multiple share |

classes in 1990, approximately 200 orders for similar exemptive relief were r

subject to substantial delays before the SEC finally adopted Rule 18f-3 in t

1995 (five years later).124 The first exemptive order for an ETF was issued i

in 1992 and the second was not issued until 1995. The SEC began granting §

blanket exemptive relief to particular ETF sponsors beginning in 2000 and o

finally proposed a rule to eliminate the significant delays typically associ- e

ated with orders for new ETF sponsors in March of 2008, more than 15 c§

years after the first ETF order and seven years after the first application for §

an actively managed ETF exemption. This happened despite the fact that, as e

the SEC noted recently, "All ETFs trading today operate in a similar way."125 2

123 Money Market Management, Inc. et al., ICA Rel. No. 9967 (Oct. 21, 1977); Valuation of Debt Instruments and Computation of Current Price Per Share by Certain Open-End Investment Companies (Money Market Funds), ICA Rel. Nos. 12206 (Feb. 1, 1982) (proposing Rule 2a7) and 13380 (July 11, 1983) (adopting final rule).

124 Goldman Sachs-Institutional Liquid Assets, ICA Rel. Nos. 17420 (Apr. 11, 1990), 55 FR 14541, and 17479 (May 8, 1990); ICA Rel. No. 20915 (Mar. 2, 1995) (adopting final rule).

125 SPDR Trust, Series 1, ICA Rel. Nos. 18959 (Sept. 17, 1992) (notice) and 19055 (Oct. 26, 1992) (order); Diamonds Trust, ICA Rel. Nos. 22927 (Dec. 5, 1997) (notice) and 22979 (Dec. 30, 1997) (order); Barclays Global Fund Advisors, ICA Rel. Nos. 24394 (Apr. 17, 2000) (notice) and 24451 (May 12, 2000) (order); SEC Rel. Nos. 33-8901; IC-28193 (Mar. 11, 2008) (proposing Rule 6c-11).

Similar delays followed initial exemptive orders for foreign finance subsidiaries (Rule 3a-5) (ten years),126 asset-backed vehicles starting in 1987 (Rule 3a-7) (five years),127 and research and development companies in 1993 (Rule 3a-8) (ten years).128

In addition, the IM should separately track exemptive request processing times for (a) routine, (b) novel but not materially different (and therefore within the IM Director's delegated authority), and (c) novel and materially different fund products (and therefore requiring SEC approval). It should report summary statistics on these times annually and commit to maintaining modest, steady improvement in these times. Breaking out these different types of requests in its performance reports will allow the public to determine whether future delays in exemptive requests are due to the nature of the requests sought, understaffing or inefficiency in IM, or the need for requests to compete for space on a lengthy SEC agenda. Each result would have different implications for future reform. If delays were due to large numbers of routine requests, further changes in IM's procedures or use of non-professional staff might be warranted. If delays were due to novel but not materially different requests, further changes in or additions to the IM's resources might be warranted. If delays were due to the need for SEC approval, further changes in the Director's delegated authority might be warranted.

Automatic Approvals Based on Precedent More ambitiously, the SEC should by rule provide for automatic approvals of exemptive requests, effective on filing, where precedent exists, as certified by counsel in good standing. As in the SEC's 1993 proposal, precedents could be limited to the most recently approved order, in its entirety, to prevent cherry-picking, and IM should retain the discretion to "de-precedent" a prior order if problems

<¡1 t U

126 General Electric Overseas Capital Corporation, ICA Rel. Nos. 8624 (De. 27, 1974) (exemptive order); ICA Rel. No. 12679 (Sep. 22, 1982) (proposing Rule 3a-5); ICA Rel. No. 14275 (Dec. 14, 1984) (adopting final rule).

127 Shearson Lehman CMO, Inc., ICA Rel. Nos. 15796 (June 11, 1987), 52 F.R. 23246 (notice) and 15852 (July 2, 1987), 38 SEC Docket 1403 (order); Exclusion from the Definition of Investment Company for Structured Financings, ICA Rel. No. 18736 (May 29, 1992) (proposing Rule 3a-7); ICA Rel. No. 19105 (Nov. 19, 1992) (adopting rule).

128 ICOS Corp., ICA Rel. Nos. 19274 (Feb. 18, 1993) (notice) and 19334 (Mar. 16, 1993 (order); Certain Research and Development Companies, ICA Rel. No. 26077 (June 16, 2003) (adopting rule). "The most prominent symbol of past SEC lethargy has been its failure to adopt a three-year old rule proposal on manager of managers approvals. If implemented, the rule—which would allow fund firms to change subadvisers without the need for a shareholder vote—would by itself do away with many incoming applications" (Wilson 2006).

arose subsequent to its issuance. Expanded reliance on precedent can be modeled on its use under Section 4(c)(8) approvals under the BHCA, made automatic by the Gramm-Leach-Bliley Act, which has generated no significant controversy or difficulties in the 10 years since it was implemented.

Although the IM staff expressed concerns about automatic approvals in 1992, they were again based on limited IM resources, and the recommendations made in the preceding section (if implemented) should remove those concerns. In addition, fund competitors would have an incentive to blow the whistle on exemptive requests that were not based on precedent. Alternatively, or in addition, a limited private right of action could n be developed, similar to the one used under Section 16 of the 1934 Act, to provide a cost-effective means of disciplining sponsors in their reliance on this precedent-based system. In 1993, the SEC also expressed the concern g that "an applicant that has received [exemptive] order gains a measure of protection from liability, even if the applicant made misrepresentations in . its application," but the only support provided for this implausible legal assertion was inapposite.129 Finally, the SEC worried that because ICA § 6(c) o "specifically requires the [SEC] to grant exemptions only if the exemption a is necessary or appropriate in the public interest and consistent with the r protection of investors ... passive granting of exemptive relief could call into t question whether the [SEC] was meeting its obligations under the ICA,"130 i but this, too seems hard to credit if the automatic approvals were limited to precedent-based applications. The SEC had already made an active determination of public benefit for these applications, and the SEC had determined pursuant to a rule-making process that it was in the interests of investors for future funds to be able to form to compete with the first mover.

If automatic, precedent-based approvals were granted, the SEC might consider committing to a one-time delay (of, say, one year) after a significant,

129 ICA Rel. No. 19362 (1993) at n.41. The SEC stated that "Under section 38(c) of the [ICA], no liability attaches 'to any act done or omitted in good faith in conformity with any rule, regulation, or order of the [SEC], notwithstanding that such rule, regulation, or order may, after such act or omission, be amended or rescinded or be determined by judicial or other authority to be invalid for any reason,'" but the very language quoted, which requires "good faith," would not be applicable to a intentional misrepresentation, and if automatic approval required an applicant to blackline the application against precedent, any claim the applicant did not intend the misrepresentation would not be credible. Moreover, as the same release went on to note, Section 34(b) of the ICA makes it unlawful to make any untrue statement of a material fact in "any registration statement, application, report, account, record, or other document filed ... pursuant to" the ICA.

130 Id.

novel order is implemented by a requesting party before it could be used as precedent. This would allow for an innovation to generate returns and to give some time for the market to respond to innovations. Without such a practice, innovators would plausibly have too little incentive to bear the significant up-front cost of pursuing a novel exemption because competitors could come in with very low-cost precedent-based applications and begin competing in rapid fashion. In addition, the delay would allow some time for potential problems not anticipated during the review process to surface, which could then be addressed in subsequent applications. The commitment to the one-time delay would promise some temporary rents for innovators, similar to the way that patents function. Although incentive schemes based on such quasi-entitlements raise difficult judgment issues (how long should the period be, should it be adjusted from time to time, and how large a rent is sufficient to induce innovation without choking off subsequent innovation), these issues have been dealt with adequately in other areas of regulation.

Finally, the SEC could consider adopting (or disclosing, if already adopted)

• Staff protocols for determining when a sufficient number of pending ex-emptive requests raise similar or related issues as to warrant rule-making initiatives

• Ways of outsourcing or de-professionalizing some of the work required for exemptive requests to alleviate staffing burdens for large surges in exemp-

. . 131

tive activity

Although training and quality control would present issues for some tasks, much of the basic administrative review and summarization of exemptive orders for further internal processing at the SEC could be done by external and/or non-professional staff.

Controlled, Experimental Exemptive Procedures with Sunsets Most boldly, the ICA's exemptive procedures could be completely revised to encourage controlled experimentation and innovation while monitoring its effects on investors. A sketch of such a revision is as follows:

New funds and fund products would receive automatic exemption upon simple notice filing. This exemption would not be permanent, but

<¡1 t U

131 To be effective, outsourcing will need to be structured to not entail duplicative work by IM staff, as happened with Fair Funds independent distribution consultants. For instance, external service providers will have to be given real delegated authority.

would have a five- to ten-year sunset. The sale of these new funds or products would be subject to mandatory disclosures under the 1933 Act and to anti-fraud rules under the 1934 Act. However, the new funds would not be subject to ongoing 1934 Act reporting. Instead, periodic disclosure would be determined under the fund's contract with investors, subject only to a mandatory five-year, after-the-fact end-of-period accounting, with full disclosure of portfolio holdings, to permit ex-post review of the risks taken and returns generated by the fund or product, as measured against relevant benchmarks. The new funds and products could be publicly advertised and would not be subject to the ICA's requirements concerning | governance (such as boards). They would be subject only to minimal requirements for custody, auditing, books and records, and internal controls. The amount and structure of adviser and distributor compensation g would be unregulated, subject only to disclosure and contract. Because transactions with fund affiliates would still be governed by the ICA, this . would not provide a blanket exemption for a fund family, only for new | funds or products. 0 To ensure that the experiments were unaffected by state regulation and | to help ensure that the new funds' disclosure and contract obligations were r met, a new private right of action would be created under federal securities t law specifically for violations of contracts and/or disclosure obligations un- i der the 1933 Act, which would preempt state laws governing new funds and products. Other than SEC enforcement of the 1933 Act and minimal governance obligations, the SEC would have no authority to regulate these new products. The SEC's IM, however, would be mandated to inspect the sponsors of the new products on a periodic basis, with unannounced, surprise, confidential inspections, subject to strong examination privileges similar to that enjoyed by banks. After five years, the SEC staff would report on the new funds or products and any problems or abuses they documented via inspections and examinations, which would be due 270 days prior to the end of the sunset period. If the specific exemption was not renewed, the existing funds or products would be permitted to maintain assets in place, provided no fraud or other crimes were committed, but would be permitted to continue to raise new assets only with new SEC authorization. If, however, the funds and products had performed well, the structures would be deemed to be exempt permanently as alternative funds or products compliant with the ICA.

Substantive Statutory Changes and Rule-Making

The following specific substantive statutory changes and rule-making initiatives should be considered by the Congress and the SEC.

First, again following up on the SEC IM Staff's 1992 Redbook recommendations and building on the SEC's recent initiatives in broker-dealer regulation (Tafara and Peterson 2007; SEC 2008a), reciprocity should be afforded to funds organized and supervised by well-supervised foreign countries, such as the member states of the E.U. (This would only be viable if U.S. tax laws were changed, as discussed in Part I, but the SEC should move forward without attempting to wait for the Congress to act on taxation.) The Redbook recommendation was that Congress amend the ICA to reduce the burden the SEC perceived the ICA to impose on both itself and foreign fund applicants for exemptive relief under Section 7(d) of the ICA. It is fair to say, however, that the requirements in Section 7(d) are sufficiently malleable and vague that a determined SEC would be able to make those findings. Although the SEC's recent rulemaking efforts under the ICA and Advisers Act have not all been successfully defended against judicial appeal, there is no reason that the SEC could not sustain those findings, particularly given the reduced role for court review of SEC exemptive relief. Nevertheless, Congressional moderation of Section 7(d)'s requirements would also be beneficial in reducing the administrative burden on the SEC in pursuing reciprocity. Reciprocity could and should be conditioned on receiving equivalent UCITS-style treatment for U.S. funds in the E.U.— this, coupled with U.S. tax reform, would have the effect of allowing U.S.-domiciled funds to be sold through the E.U., without the duplicative and burdensome need to engage in fund cloning. For investors, the advantage of reciprocal treatment is that it allows for a natural experiment in which funds and fund products developed under a number of different but (from an investor's perspective) roughly comparable regulatory regimes could compete on a level playing field, revealing information about the relative value (as perceived by investors) of those regimes.

Second, fund advisers must compete for investors and have their fees approved by independent fund boards. Unfortunately, as noted in Part II.B, fund advisers may only be paid based on assets or in symmetric fulcrum fees. Although advisers can and do pay performance fees to individual portfolio managers of mutual funds, the advisory companies themselves cannot be paid by the mutual funds with performance fees. In less competitive subsectors of the fund industry, advisory companies

<¡1 t U

might have weak incentives to improve fund performance by, for example, reducing portfolio turnover or brokerage commissions. As noted in Part I.C., advisers to hedge funds and commodity pools might and commonly do receive compensation based both on assets and returns, and performance fees are not generally symmetric. When properly measured, moreover, typical hedge fund compensation generates total costs for the fund that are no higher than asset-based compensation used by nearly all mutual funds, but better align to the incentives of portfolio managers of actively managed funds (Kritzman 2008). As noted in Part II.D, UCITS funds in Europe do not face restrictions on performance fees, and funds n

commonly pay performance fees without evident harm to investors. |

Although some regulatory checks on mutual fund fee structure are sen- f

sible, current rules are out of date and do not reflect modern economic g

thought. On behalf of the SEC, Hung et al. (2008), in early 2008, surveyed finance and economic research on adviser compensation and concluded .

that the articles reviewed "generally report that a bonus-compensation structure, in which the adviser is paid a bonus (either a fixed sum or a О

percentage) if the portfolio return exceeds a predetermined benchmark, |

is the optimal contract from the client's point of view." The SEC should r

revisit these rules to better improve the incentives and performance of t

mutual fund advisers (and, at a minimum, provide ready exemptive relief i

to funds that seek to pay performance fees).

Third, consistent with the recent memorandum of understanding between the SEC and the CFTC (SEC 2008c) and consistent with the U.S. Treasury Blueprint's call for more coordination among financial regulators generally (U.S. Dept. of Treas. 2008), the SEC, the CFTC, and the bank regulators should jointly undertake a review of disclosure and other requirements for collective investments, along the lines begun in Part II. There is no particular reason that similar collective investments sold to the public should have disparate mandatory content or formats. Although there may be reasons to exempt particular classes of investments from particular requirements—such as when they are sold exclusively to wealthy investors—the current overlapping regulatory regimes provide for different wealth- or sophistication-based exemptions for legally different but economically similar collective investments sold by firms subject to different regulatory regimes to investors who cannot distinguish among them. A 2008 RAND study commissioned by the SEC, for example, found that "investors typically fail to distinguish broker-dealers and investment advisers

along the lines that federal regulations define" (Hung et al. 2008, xiv). More troubling, it found that disclosures required by current law do not help customers understand the distinction or its implications for regulation, intermediary duties, or customer rights (19). Disclosure regarding fees, expenses, conflicts of interest, and after-tax returns should be harmonized across mutual funds, registered ETFs, ETNs, SMAs, variable annuities, commodity pools, and commodity ETFs that are marketed to the public, and a single, simple bright-line test for distinguishing between wealthy or sophisticated investors for all classes of collective investments should Q

be adopted. Disparate disclosures create investment biases and perverse n

incentives for brokers and financial advisers and distort competition. g

Supervisory Examinations and Inspections

At a minimum, the SEC should continue and expand the OCIE pilot program that uses dedicated teams of two to four examiners to provide more continuous and in-depth oversight of the largest and most complex groups of affiliated investment companies and investment advisers (SEC 2007a, 10). In addition, the SEC should ensure that inspections and examinations are coordinated, recognizing that many funds and advisers are part of integrated networks located in multiple SEC regions, as recommended by ICI (ICI 2007a). Finally, the SEC should consider moving inspection and examination authority for regulated funds back to IM from OCIE, as recommended by the ICI (ICI 2007a). OCIE should not function as an arm of enforcement, but as an arm of the more informed and forward-looking division of the SEC responsible for the fund industry, part of a prudential system of oversight more akin to bank supervision than to conventional SEC enforcement (Wasserman 2005). Enforcement should be aimed at those who intentionally and clearly break the law. OCIE and the information that an effective examination process produces should, first, be used to improve ongoing regulatory oversight, rule-making, exemptive request processing, and secondarily to prepare the way for changes in the law. Use of OCIE-generated information for enforcement purposes obviously runs the risk that the OCIE/adviser/fund relationships will be fraught, confrontational, and ultimately less productive. Evidence of intentional law-breaking, of course, should not ignored or swept under the rug, but examinations and inspections that are essentially fishing expeditions with enforcement in mind will tend to be both wasteful and counterproductive.

Reports and Studies

Last, the SEC should consider a number of self-imposed performance reporting measures that have the potential for improving its operations over time. First, the SEC should report, as planned, ICA exemptive request processing times separately from no-action and other relief requests, but also expand it, as described previously, to distinguish between standard, novel but not material, and materially novel requests. Second, the SEC should benchmark regulated fund assets as it currently does against the prior year as a performance measure, but it should also benchmark the growth rates in fund assets against growth rates of non-regulated and foreign funds, g

such as UCITS funds. Third, the SEC should report SEC staff turnover by department. f

Finally, with the assistance of the OEA, IM staff should do detailed g

studies comparing current U.S. regulation of funds to U.S. regulation of non-regulated funds, such as commodity pools and SMAs, and to E.U. I

regulation of UCITS funds. Then it should provide, where possible, quan- f

tified support for elements of U.S. regulation that are more stringent and/ 0

or burdensome on U.S. regulated funds than their non-regulated and g

E.U. counterparts. In addition, again with the assistance of OEA, the SEC 0

should pursue rule-making experiments that can produce scientifically re- i

liable information about the value of the rules that should be designed. g

Specifically, randomized trials might be feasible for granting exemptive re- g lief from existing rules regarding governance, compensation, and conflicts

of interest on a randomized basis to a subset of applicants that would be e

matched (based on relevant criteria) with existing advisers or funds that <§

would not be given such relief. This would provide a reliable basis for as- g

sessing the effects of the relief over time. g


Armour, J., B. Black, B. Cheffins, and R. Nolan. 2007. Private enforcement of corporate law: An empirical comparison of the US and UK. Working paper.

Association of the Luxembourg Fund Industry. 2007. Annual report 2006-2007: The Luxembourg fund industry in figures. Available at StatRA_2006-2007_GB.pdf.

Barber, B., and T. Odean. 2000. Trading is hazardous to your wealth. J. Fin. |


Barber, B. M., T. Odean, and L. Zheng. 2005. Out of sight, out of mind: The

effects of expenses on mutual fund flows. J. Bus. 78:2095-2119. g

Barclay, M., N. Pearson, and M. Weisbach. 1998. Open-end mutual funds

and capital gains taxes. J. Fin. Econ. 49:3-43. .

Baumol, William J., Stephen M. Goldfeld, Lilli A.Gordon, and Michael F. |

Koehn. 1990. The economics of mutual fund markets: Competition versus o

regulation. New York: Springer. a

Beck, S. 2007. The transformers. Am. Lawyer 29:94-100.

Benedict, J., M. Dulka, and K. W. Price. 2004. A perfect storm: Analyzing the

recent explosion of mutual fund litigation. Sec. Litig. Rep. 1:20-23. i

Benedict, J., S. Murphy, C. N. Gray, R. Miller, and Milbank, Tweed, Hadley and McCloy LLP. 2008. Recent developments in litigation involving mutual funds and investment advisers. ALI-ABA Course of Study.

Board of Governors of the Federal Reserve System. 2008. Ownership of equities and corporate mutual bonds. Retrieved Dec. 12, 2008 from the Insurance Information Institute's website at savings/investments/.

Bruce, R. 1993. Luxembourg thrives as an Eden for funds. Int'l Herald Tribune, Nov. 13.

Bucks, B. K., A. B. Kennickell, and K. B. Moore. 2006. Recent changes in U.S. family finances: Evidence from the 2001 and 2004 Survey of Consumer Finances. Fed. Res. Bull. A1-A38, available at PUBS/oss/oss2/2004/bull0206.pdf.

Burman, L. 1999. The labyrinth of capital gains tax policy. Washington, DC: Brookings Institution Press.

Campbell, John. 2006. Household finance. J. Fin., 61:1553-1604.

Chevalier, J., and G. Ellison. 1997. Risk taking by mutual funds as a response to incentives. J. Pol. Econ. 105:1167-1200.

Choi, Stephen, and Marcel Kahan. 2007. The market penalty for mutual fund scandals. B. U. L. Rev. 87:1021-1059.

Clark, R. 1986. Corporate law. New York: Aspen Publishers.

Coates, John C. IV. 2007. The foals and promise of the Sarbanes-Oxley Act. J. Econ. Persp. 21:91-116.

Coates, John C. IV. 2008. A review of economic theory and evidence on mutual fund governance. Working paper.

Coates, John C. IV, and R. Glenn Hubbard. 2007. Competition in the mu- n

tual fund industry: Evidence and implications for policy. Journal of Corporation Law. 33:151-222.

Comité pour le Développement de la Place Financière. 2006. Etude d'impact g

de l'industrie financière sur l'economie. Etude_impact_2006.pdf. .

Commission of the European Communities. 2008. Impact assessment of |

the legislative proposal amending the UCITS directive. Commission 0

Staff Working Document. COM (2008) 458 Final; SEC (2008) 2264. 3

Culloton, D. 2006. Are ETFs really more tax-efficient than mutual funds? r

Morningstar Advisor. Feb. 14. t

Deli, D. N. 2002. Mutual fund advisory contracts: An empirical investiga- i

tion. J. Fin. 57:109-133.

Der Hovanesian, Mara. 2006. A peek into the hedges. Bus. Wk. Dec. 11.

Dhar, R., and N. Zhu. 2006. Up close and personal: An individual level analysis of the disposition effect. Mgt. Sci. 52:726-740.

Donahue talks about speeding up exemptive relief. 2006. Fund Action. June 26.

Dougherty, C. 2006. Collective shareholder lawsuits reach European courts. Int'l Herald Trib. Dec. 1.

Edelen, R. 1999. Investor flows and the assessed performance of open-end mutual funds. J. Fin. Econ. 53:439-466. 2006. What are exchange-traded notes? http://www.

European Fund and Asset Management Association. 2008a. Annual asset management report: Facts and figures. Brussels: EFAMA.

-. 2008b. UCITS as a global brand. Brussels: EFAMA.

Fink, M. 2008. The rise of mutual funds: An insider's view. New York: Oxford University Press, U.S.A.

Frankel, Tamar. 2004. The regulation of money managers: Supplement 2004. New York: Aspen Law and Business.

Frankel, Tamar, and Ann T. Schwing. 2001. Regulation of money managers: Mutual funds and advisers, 2nd. Ed. New York: Aspen Publishing.

Franks, J. R., S. M. Schaefer, and M. D. Staunton. 1997. The direct and compliance costs of financial regulation. J. Banking and Fin. 21:1547-1572.

Freeman, Jr., D. F. 2004. Bank common trust funds and collective investment funds. Inv. Lawyer 11:15-18.

French, K. R., and J. M. Poterba. 1991. Investor diversification and international equity markets. Am. Econ. Rev. 81:222-226. 3

Gastineau, G. 2002. The exchange-traded funds manual. New York: Wiley.

Gavin, J. 1994. Ireland gears up for U.S. funds. Int'l Fin'l L. Rev. 13:18-19.

Goetzmann, W. N., J. E. Ingersoll, and S. A. Ross. 2003. High-water marks g

and hedge fund management contracts. J. Fin. 58:1685-1718.

Goetzmann, W., and A. Kumar. In press. Equity portfolio diversification. .

Rev. Fin. ^

Gruber, M. 1996. Another puzzle: The growth in actively managed mutual O

funds. J. Fin. 51:783-810. 3

Hammer, D., C. Reiser, G. Haynes, K. Koren, A. Caldwell, J. Frolik, C. Mick- r

elson, B. Sacks, J. Milani, and H. Dunn. 2006. U. S. regulation of hedge t

funds. Washington, DC: American Bar Association. £

Hansmann, H., and R. Kraakman. 2000. The essential role of organizational law. Yale L. J. 110:387-440.

Hansmann, H., and U. Mattei. 1998. The functions of trust law: A comparative legal and economic analysis. N. Y. U. L. Rev. 73:434-479.

Herszehorn, David. 2008. A mortgage rescue strains calculations. New York Times. July 23.

Hoffman, D. 2008. ETFs are enjoying steady growth. Inv. News, available at article?AID=/20080414/REG/343787601/1025/ETF.

Horwitz, D. L., and J. Markham. 1983. Sunset on the commodity futures trading commission: Scene II. Bus. Law. 39:67-100.

Humphreys, T., S. Merali, and R. Reigersman. 2007. The U.S. tax treatment of exchange-traded notes—The brewing battle between mutual fund advisers and ETN sponsor. Available at files/13102.html.

Hung, A. A., N. Clancy, J. Dominitz, E. Talley, C. Berrebi, and F. Suvanku-lov. 2008. Investor and industry perspectives on investment advisers and broker-dealers. Washington, DC: Rand Institute for Civil Justice. Internal Revenue Service. 2002. Electing mark to market for marketable

stock. Fed. Register 67:49634-49643. International Organization of Securities Commissions (IOSCO), Technical Committee. 2006. Examination of governance for collective investment schemes. and Investment Company Institute. 2005. 2005 investment company fact book. n g

-. 2006. 2006 investment company fact book. http://www.icifactbook. f

org/2006/. g -. 2007a. Review of the regulatory structure associated with financial :

-. 2007c. The differences between mutual funds and hedge funds. http://

-. 2008a. 2008 investment company fact book, 48th Edition. http://

-. 2008b. Trends in mutual fund investing: October 2008. http://www.

-. 2008c. One sensible solution.


Ippolito, R. 1992. Consumer reaction to measures of poor quality. J. L. and Econ. 35:45-70.

Ireland Financial Services Regulatory Authority 2007. Financial regulator. 07.pdf.

Irish Industry Funds Association 2008. Industry statistics. http://www.

Jackson, H. E. 2006. Regulatory intensity in the regulation of capital markets: A preliminary comparison of Canadian and U.S. approaches. Research study commissioned by the Task Force to Modernize Securities Regulation in Canada. pdf.

institutions. Submission to the Department of the Treasury. http://www. . ^ -. 2007b. Trends of ownership in mutual funds in the United States,

2007. Research Fundamentals. 16 (5):1-12. a

-. 2007. Variation in the intensity of financial regulation: Preliminary

evidence and potential implications. Yale J. Reg. 24:253-292.

Jackson, H. E., and M. J. Roe. 2008. Public and private enforcement of securities laws: Resource-based evidence. Working paper.

Jackson, H. E., and E. L. Symons, Jr. 1999. Regulation of financial institutions. New York: West Publishing.

Jackson, M., and D. Counihan. 2008. Investment funds: Ireland.

Johnson, S. 2006. How UCITS became a runaway success: Despite their clumsy name, the kitemarked funds are now a force to be reckoned with across the world. Fin. Times. Nov. 27. 1

Khorana, A., H. Servaes, and P. Tufano. 2005. Explaining the size of the mutual fund industry around the world. J. Fin. Econ. 78:145-185.

-. In press. Mutual fund fees around the world. Rev. Fin. Stud. g

KPMG. 2007. State of the investment management industry in Europe. http:// .

pdf. O

-. 2008a. Ireland regulation. Funds and Fund Management 2008. http:// O 3 lation.pdf. h -. 2008b. Luxembourg regulation. Funds and Fund Management

2008. i


KPMG Peat Marwick LLP. 2000. Tax-managed mutual funds and the taxable investor.

Kritzman, M. 2008. Who charges more: Hedge funds or mutual funds? J. App. Corp. Fin. 20:121-123.

Kruithof, M. 2005. Conflicts of interest in institutional asset management: Is the E.U. regulatory approach adequate? Working paper. http://papers.

Lee, C., A. Shleifer, and R. Thaler. 1991. Investor sentiment and the closed-end fund puzzle. J. Fin. 46:75-109.

Lewis, K.K. 1999. Trying to explain home bias in equities and consumption. J. Econ. Lit. 37:571-608.

Lipper. 2008. Taxes in the mutual fund industry. http://www.lipperweb. com/research/ResearchStudiesOverview.asp?doc=7334-42121796 ResearchStudy20080415efi-2&lang=ENG&c=US.

Lybecker, M.E. 1973. Regulation of bank trust department investment activities. Yale L. J. 82:977-1003.

Managed Fund Association. 2004. Survey of selected substantive and technical issues raised in SEC comment letters re: registration of hedge fund advisers. 1990. Managing money: A legal guide to the world's investment fund markets. Int'l Fin. L. Rev. Markham, J. 1987. Commodities regulation: Fraud, manipulation, and other

claims. Deerfield, IL: Clark Boardman Callaghan. McAfee, W. 2003. Photo finish.

research/Publications/BandE/bande50/50n3/photo.html. McGeough, D., and L. Quirke. 1992. Ireland. Int'l L. Rev. |

McGuire, W. J., and R. F. Helmrich. 2008. The SEC's ETF rule proposal: |

Key issues for the future exchange traded products. Investment Lawyer f

15:1-13. §

Mintz, J., and M. Smart. 2002. Tax-exempt investors and the asset allocation

puzzle. J. Pub. Econ. 83:195-215. .

Murphy, S., and C. Bassel. 2005. Mutual funds under scrutiny: An overview |

of recent litigation. Andrews Sec. Litig. and Reg. Rep. 10. 0

National Archives and Records Administration. 2008. Part VI: Securities 1

and exchange commission. Fed. Register. 73:37751-37774. r

National Commission on Financial Institution Reform, Recovery and En- t

forcement. 1993. Origins and causes of the S&L debacle: A blueprint for i

reform: A report to the President and Congress of the United States. Wash- g

ington, DC. 0

National Regulatory Services, and Investment Adviser Association. 2007. e

Evolution revolution: A profile of the investment adviser profession. http:// <§ §

and_Bro chures/IAA-NRS_Evolution_Revolution_Reports/evolution_ f

revolution-2007.pdf. §§

Neylin, D. 2004. Regulatory frameworks: A key challenge for the hedge fund 5

industry. Finance Dublin. PriceWaterhouseCoopers. Odean, T. 1998. Are investors reluctant to realize their losses? J. Fin. 53:1775-98.

-. 2000. Do investors trade too much? Am. Econ. Rev. 89:1279-1298.

O'Mary, M. A. 2007. Mutual fund regulation in the United States: The cost

of the status quo. Working paper (on file with author). Partsch, T., J. Terblanche, and B. Malaniuk. 2008. Investment funds: Luxembourg.

Patel, J., R. Zeckhauser, and D. Hendricks. 1994. Investment flows and performance: Evidence from mutual funds, cross-border investments and new issues. In Japan, Europe and the international financial markets: Analytical and empirical perspectives, ed. R. Satlk, R. Levitch, and R. Ram-achandran. New York: Cambridge University Press.

Paul, P. 1992. The European community's UCITS directive: One model for United States regulatory change in a globalized securities market. Vand. J. Transnat'l L. 25:61-96.

Polkovnichenko, V. 2005. Household portfolio diversification: A case for

rank-dependent preferences. Rev. Fin. Stud. 18:1467-1502. |

Poterba, J. M. 2004. Valuing assets in retirement savings accounts. Nat'l Tax J. 57:489-512.

PriceWaterhouseCoopers. 2007a. Ireland hedge funds. 3

--. 2007b. Luxembourg hedge funds.

ins-sol/publ/pwc_luxembourg_hedgefunds.pdf. .

Rapoport, L. 2008. Auction-rate crackdown widens; UBS faces charges in |

New York, as scrutiny of Wall Street's role intensifies. Wall Street Journal. 0

July 25. J

Rauch, J. 1999. Networks versus markets in international trade. J. Int'l Econ. r

48:7-35. t

Reich, Y. 1998. Taxing foreign investors' portfolio investments: Develop- i

ments and discontinuities. Tax Notes 79:1465-1472.

Reid, B. 2006. Competition in the mutual fund business. Washington, DC: Investment Company Institute. rc_competition.pdf.

Roe, M. 1996. Strong managers, weak owners. Princeton, NJ: Princeton University Press.

Rogers, W., and J. Benedict. 1982. Money market fund management fees: How much is too much? N. Y. U. L. Rev. 57:1059-1125.

Romano, R. 1993. The genius of American corporate law. Washington, DC: The AEI Press.

Rounds, C.E., Jr., and A. Dehio. 2007. Publicly traded open end mutual funds in common law and civil law jurisdictions: A comparison of legal structures. N. Y. U. J. L. and Bus. 3:473-518.

Roye, P. 2002. Remarks at the global challenge in investment management regulatory and legal issues. U.S. Securities and Exchange Commission.

Sanchirico, C.M. 2008. The tax advantage to paying private equity fund managers with profit shares: What is it? Why is it bad? U. Chi. L. Rev. 75:1071-1079.

Sarkissian, S., and M. Schill. 2004. The overseas listing decision: New evidence of proximity preference. Rev. Fin. Stud. 17:769-809.

Saxton, J. 2001. The taxation of mutual fund investors: Performance, savings and investment. Washington, DC: Joint Economic Committee.

Schonfeld, V. E., and T. Kerwin. 1993. Organization of a mutual fund. Bus. Lawyer 49:107-162.

Seligman, Joel. 2005. Should investment companies be subject to a new stat- |

utory self-regulation organization? Wash. U. Law. Rev. 83:1115-1126.

Selman, T. 1992. Model for reform: The offer and sale of UCITS in the f

United States. Int'l Law. 26:623-666. 3

Sirri, E., and P. Tufano. 1998. Costly search and mutual fund flows. J. Fin.

53:1589-1622. .

Strauss, L. C. 2006. Tailor made. Barron's. Feb. 20. f

Sunstein, Cass R. 2005. Laws of fear: Beyond the precautionary principle. 0

Cambridge, UK: Cambridge University Press. |

Tafara, Ethiopis and Robert J. Peterson. 2007. A blueprint for cross-border 0

access to U.S. investors: A new international framework. Harv. Int'l L. J. t

48:31-68. §

Thompson, J., and S. -M. Choi. 2001. Governance systems for collective investment schemes in OECD countries. OECD Financial Affairs Division Occasional Paper 1.

U.S. Commodity Futures Trading Commission. 2000. CFTC/SEC agreement to reform Shad-Johnson Accord. opasynposis.htm.

-. 2007. Performance and accountability report fiscal year 2007. file/2007performacctreport.pdf.

U.S. Department of the Treasury. 2008. Blueprint for a modernized financial regulatory structure. regulatory-blueprint/.

U.S. General Accounting Office. 2002. SEC operations: Implications of alternative funding sources. GAO-02-864. d02864.pdf.

-. 2004a. SEC operations: Oversight of mutual funds industry presents

management challenges. -. 2004b. Securities and Exchange Commission review of fiscal year

2003 and 2004 budget allocations. cgi-bin/getdoc.cgi?dbname=gao&docid=f:d04818.pdf.

U.S. Securities and Exchange Commission. 1990. Request for comments on reform of the regulation of investment companies. Fed. Reg. 55:25,322-25,339.

-. 1996. 40 act exemptive applications. Audit Report No. 230. http://

-. 2003a. Implications of the growth of hedge funds. http://investor. n


-. 2003b. SEC Chairman Donaldson releases statement regarding initia-


-. 2006b. Performance and accountability report2006. http://www.sec.


-. 2007a. Steps being taken to make examination program more risk-based

and transparent. GA0-07-1053. pdf.

-. 2007b. Variable annuities: What you should know. http://www.sec.


-. 2008a. FY2009 congressional justification in brief. http://www.sec.


-. 2008b. SEC, Australian authorities sign mutual recognition agreement.

Press release: 2008-182. htm.

-. 2008c. SEC, CFTC sign agreement to enhance coordination, facilitate

review of new derivative products. Press release 2008-40. http://www.sec. gov/news/press/2008/2008-40.htm.

-. 2008d. SEC improves protections for seniors and other investors in

equity-indexed annuities. Press release 2008-298. news/press/2008/2008-298.htm.

tives to combat late trading and market timing of mutual funds. http://www. 3

-. 2004. Summary of SEC initiatives in response to scandal. http://www. o

pdf. o

-. 2006a. IM exemptive application processing. a

U. S. Securities and Exchange Commission Division of Investment Management. 1992. Protecting investors: A half-century of investment company regulation. Washington, DC: U.S. Securities and Exchange Commission.

U.S. Securities and Exchange Commission Staff Report. 2005. Exemptive rule requirements of2004: The independent chair provision. http://www. von Cottier, Philipp. 1997. Hedge funds and managed futures. Bern, Germany: Haupt Verlag AG. Wallison, P. J., and R. E. Litan. 2007. Competitive equity: A better way to or- n

ganize mutual funds. Washington, DC: AEI Press. Wasserman, Craig M. 2005. Dear prudence: Finding a proper balance in the

post-Enron regulatory and enforcement arena. Yale Law School Center g

for Law, Economics and Public Policy Res. Paper No. 325. http://papers. .

Wilson, S. 2006. SEC to usher in fast exemptive relief. Fund Action. Oct. 16. |

Zhu, N. 2005. Cost of time and household choice between direct and del- 0

egated investment. Working paper. |

cfm?abstract_id=686569. r

<¡1 t U

appendix a: types of regulated funds

For regulatory purposes, the ICA divides collective investments into two types:

• Unit investment trusts (UITs) are governed by contract, do not have directors, and issue redeemable shares representing undivided interests in specified securities.132

• Management companies133 do not invest in specified securities, are organized as legal entities, and have separate boards of directors. Management companies are further divided into

° Open-end funds that issue securities redeemable at the funds' net asset

value per share (NAV). g

° Closed-end funds that issue non-redeemable securities and typically trade on an exchange.134

Conventional open-end funds—generally known in the U.S. as mutual funds—do not trade on a stock exchange and most continuously offer shares for sale.

Regulated funds also differ by the primary nature of the investment they с/

hold, such as equity funds, bond funds, or hybrid funds. One important и

type, a money market mutual fund (MMMF), holds low-risk short-term e

debt (primarily government securities) to preserve a fixed NAV, and so func- y

tions like a bank account. Introduced in 1971, MMMFs in 2007 held 26% S

of mutual fund assets and 31% of the cash of U.S. businesses.135 Another §

important type—an index fund—by contract invests only (or primarily) in o

securities held in a published market index of some kind (for instance, the J

Standard and Poor's 500). Because the selection of investments is made by 7

132 For a description of UITs in operation, see United States Trust Co. v. Alpert, 10 F. Supp. 2d 290, 297 (S.D.N.Y. 1997), aff'd sub nom. United States Trust Co. v. Jenner, 168 F.3d 630 (2d Cir. 1999). In Europe, unit trusts are common, but differ in that the securities in which they invest are not fixed but can vary over time.

133 In the U.S. law, the phrase refers to a fund that is managed; in Europe, the phrase "management company" generally refers to the company that manages a fund, which in the U.S. would be referred to as an advisory company. Except for this paragraph, this report adopts the more general European usage.

134 ICA § 6 further divides management companies into diversified and non-diversified companies.

135 ICI 2008 Factbook, at 112, 11.

third party sponsors of the relevant indexes and not by portfolio managers working for the adviser, index funds are known as passively managed funds. Mutual funds that retain or contract for specific investment selection by portfolio managers are generally called actively managed funds.

More recently, a new type of fund—exchange-traded funds (ETFs)—has emerged. ETFs are typically open-end funds that issue and redeem shares only in large blocks (for instance, 50,000 shares), called creation units, to authorized participants. They purchase the units with portfolio securities, break up the creation units, and sell them to the public. The units then trade as shares on a stock exchange, as would a closed-end fund (McGuire n

and Helmrich 2008).136 Because arbitrageurs can purchase shares of an ETF and redeem them, or purchase shares from an ETF and sell them on the market, there is typically little or no difference between the NAV of an ETF g

and its share price, unlike closed-end funds.137 To date, nearly all ETFs approved by the SEC have been passively managed—linked to an index like . an index fund—but in February 2008 the SEC approved four fully trans- | parent actively managed ETFs, seven years after seeking comment on the 0 concept.138 |

<¡1 t U

136 Eight of 601 ETFs registered as of December 2007 were organized as UITs. ETFs have converged on the open-end company as the standard model in recent years. The SEC's proposed rule authorizing standard ETFs would not apply to UIT-ETFs. See IC-28193 (Mar. 11, 2008) at 22.

137 Lee, Shleifer, and Thaler (1991) note that discounts of 10-20% as normal for U.S. closed-end funds. The average difference between the daily closing price and the daily NAV of ETFs that track domestic indexes is generally less than 2% (Vanguard U.S. Stock ETFs, Prospectus 56-59 Apr. 27, 2007).

138 WisdomTree Trust et al., ICA Rel. Nos. 28147 (Feb. 6, 2008) (notice) and 28174 (Feb. 27, 2008) (order); Barclays Global Fund Advisors et al., ICA Rel. Nos. 28146 (Feb. 6, 2008) (notice) and 28173 (Feb. 27, 2008) (order); Bear Sterns Asset Management, Inc. et al., ICA Rel. Nos. 28143 (Feb. 5, 2008) (notice) and 28172 (Feb. 27, 2008) (order); PowerShares Capital Management LLC et al., ICA Rel. Nos. 28140 (Feb. 1, 2008) (notice) and 28171 (Feb. 27, 2008) (order); Actively Managed Exchange-Traded Funds, ICA Rel. No. 25258 (Nov. 8, 2001).

appendix b: exempt or excluded collective investments

The ICA does not apply to all collective investments sold to individuals. In the first instance, it purports to apply very broadly,139 but Section 3(c) of the ICA excludes, among others, funds sold to

• Fewer than 100 investors

• Brokers and underwriters

• Commercial banks, insurance companies,140 or bank common trust funds °

• Commercial lenders 1

• Factors and real estate investment companies d

• Qualified purchasers ($5+ million of investments) 0

As suggested by this list, the ICA reflects a political bargain in which fi- :

nancial institutions in regulated sectors have preserved their businesses .

without complying with the ICA. Many services offered by these institu- 0

tions are not close substitutes for collective investments in securities (for 0

example, bank checking accounts or traditional life insurance policies), |

and investment banks, commercial banks, and insurance companies are r

subject to extensive regulatory schemes of their own (Jackson and Symons t

1999). Other collective investments are similar in design and function, but i

hold specialized assets, such as real estate investment trusts [REITs], or se-curitized assets, such as credit card receivables or collateralized mortgage obligations.

In addition to sponsoring and advising mutual fund complexes subject to the ICA, excluded institutions have developed collective investment vehicles that are economically closer substitutes for regulated funds but are subject to different regulatory regimes. Bank common trust funds allow for trust

139 Section 3 of the ICA generally provides an investment company is any person who issues any security, including any investment contract or interest in a security, and is (a) primarily in the business of investing in securities or (b) is in that business and has 40+% of its assets invested in non-government securities.

140 In the 1950s, insurance companies introduced variable annuities economically similar to mutual funds, while also having insurance functions, such as insuring against death or an unusually long life. Companies exclusively selling such products must register under the ICA, and insurance companies that sell primarily traditional insurance products but segregate variable annuity assets in separate accounts must register those accounts as registered investment companies (SEC v. Variable Annuity Life Ins. Co., 359 U.S. 65 1959; In re Prudential, 41 S.E.C. 335; Prudential Ins. Co. v. SEC, 326 F.2d 383 1964). Typically, separate accounts are organized as UITs, but they can also be registered as open-end investment companies (SEC 1992, 374, n.7).

accounts (including individual retirement accounts [IRAs]) to be collectively invested and managed.141 Brokerage firms142 have developed separately managed accounts (SMAs) (also called wrap or managed accounts) that purport to and sometimes do provide individually tailored advice and other services, but which depend on economies of scale achieved through collective investment, reflected in what can be relatively standardized algorithms for portfolio selection,143 pooled execution and custody, and shared back-office fa-cilities.144 Since 2005, investment and commercial banks have also sold more than $5 billion in exchange-traded notes (ETNs), unsecured debt contracts that pay returns based on specified market indexes and so track the perfor- |

mance of equivalent ETFs (see ETNCenter 2008; Hoffman 2008). These notes are general claims on the issuer, do not represent ownership of investments tracked by the index, and expose investors to the issuer's credit risk—made 3

vivid by the fact that Bear Stearns was a major issuer of ETNs before its forced

141 Investment Company Institute v. Conover, 790 F.2d 925, 938 (D.C. Cir. 1986), cert. denied, 479 | U.S. 939 (1986). Generally, trust beneficiaries may not withdraw funds, as they can from mutual S funds, and trustees retain investment discretion (subject to fiduciary duties and bank regula- r tions), rather than committing to a particular investment strategy, as is typical for mutual funds. 1 The Gramm-Leach-Blilely Act of 1999 clarified that common trusts sponsored by banks would U be subject to the ICA if they are "advertised" or "offered for sale to the general public," 1999 S. U.S.C.C.A.N. 113 Stat. 1338, 1401-02, amending ICA § 3(c)(3) (effective May 12, 2001), codify- S ing a long-standing SEC interpretive position that trusts distributed to the public as a form of 3 investment are not exempted from the ICA. Some commentators have suggested that because S the same statute does not clearly address common trusts sponsored by S&Ls, they remain out- e side the ICA (Frankel and Schwing 2001, 6-119). However, IRC § 584, which governs common ¡ trust funds, specifically refers to the OCC's Regulation 9, which governs national banks. 0

142 Eighty percent of SMAs are sold through "major brokerage firms" such as Smith Barney and U Merrill Lynch (Strauss 2006, 27). 3

143 The standardization of investment advice has grown substantially since the advent of portfolio 0 theory, with its emphasis on diversification, and the efficient capital market hypothesis, with its 5 implication that most investment managers are unable to systematically outperform the market. The widespread (if partial) acceptance of these ideas has led to the growth of passively managed mutual funds and ETFs, an emphasis on low costs rather than selection of investments,

and increasingly standard advice for at least a large portion of a given client's portfolio.

144 The SEC has adopted a "non-exclusive safe harbor" for programs offering discretionary investment advisory services, provided that "each client's account in the program is managed on the basis of the client's financial situation and investment objectives," information about which has been obtained prior to opening the account; subject to reasonable restrictions imposed by the client; the sponsor reports and inquires of the client on a periodic basis as to changes in the client's situation and objectives; the sponsor's personnel and management are reasonably available to clients for consultation; the client has the right to withdraw assets; and the sponsor passes through votes, confirmations, and rights to sue to the clients (SEC Rule 3a-4).

sale to J.P. Morgan (ETNCenter 2008).145 None of these forms of collective investment are subject to detailed, proscriptive regulations akin to the ICA.

The ICA's 3(c)(1) and 3(c)(7) exclusions provide private funds (hedge funds, venture capital funds, and private equity funds) with a means of remaining almost entirely unregulated, provided they limit their services to a small number of investors or to very wealthy investors.146 Private funds are typically actively managed and not subject to any detailed federal regulation, other than an anti-fraud rule adopted in 2007 that makes it a fraudulent practice for any investment adviser for a collective investment to make false or misleading statements or otherwise defraud investors.147 Also in |

2007, prominent private equity and hedge fund managers Blackstone and Fortress sold interests to the public under the 1933 Act but avoided the need to register under the ICA by selling shares in their management and g

advisory companies, rather than in the portfolio companies in which their affiliated private funds have invested (Beck 2007). .

The ICA also does not apply to funds or companies that invest in some- |

thing other than securities, such as commodities or foreign currencies, or O

derivatives related to commodities or foreign currencies, as the courts have |

interpreted those terms.148 Commodity pools invest primarily in commod- r

ities, derivatives (including futures on broad-based stock indexes [CFTC t

2000]), or other commodity pools and are generally not covered by the i

ICA. Conventional commodity pools are actively managed and limited to a small number of wealthy investors, and a smaller number have sold shares publicly. Since the advent of ETFs, a growing number of commodity pools have been listed for trading as passively managed ETFs. In 2004, commodity pools held more than $600 billion in assets. From their inception in 2004, commodity ETFs have grown rapidly to hold more than $28 billion, representing 5% of all ETF assets (ICI 2008a, 40).

145 Growth in ETNs has slowed of late, in part due to credit market turmoil and in part because of the possibility that their current tax treatment, which is favorable relative to mutual funds and registered ETFs, might be lost. IRS Revenue Ruling 2008-1 and Notice 2008-2 (Dec. 7, 2007) concludes that a single currency-linked ETN should be taxed as a debt instrument, rather than a prepaid forward contract, resulting in less favorable tax treatment.

146 To avoid registration under the 1934 Act, private funds must not have 500+ beneficial owners (1934 Act §§ 12[g], 15[d]).

147 Rule 206-(4)-8; see Adviser Act Rel. 2628 (Aug. 7, 2007).

148 For example, Chicago Mercantile Exchange v. SEC, 883 F.2d 537, 550 (7th Cir. 1989) and Bd. of Trade of the City of Chicago v. SEC, 187 F.3d 713, 726 (7th Cir. 1999).

appendix c: activities of mutual fund boards

Mutual fund boards participate in the following activities:

• Approve advisory, underwriting (distribution), and custody contracts and fees

• Approve other contracts with the adviser or its affiliates, such as for ancillary services (such as transfer agency, shareholder accounting, or brokerage services)

• Oversee other conflict-of-interest transactions with the adviser or its °

affiliates 1

• Determine whether a redemption fee is appropriate for the fund d

• Select independent auditors 0

• Review and approve fidelity bonds and/or insurance policies t

• Value securities for which market prices are not available j

• Approve fund mergers or liquidations

• Determine the credit quality of certain debt securities

• Retain independent counsel to advise the board and/or independent directors

• Approve compliance polices and procedures /

• Oversee the fund's chief compliance officer U

appendix d: regulation of mutual funds and other publicly traded collective investments, by major jurisdiction

U.S. Luxembourg Ireland

Disclosure at initial offering Mandatory (1933 Act) Mandatory Mandatory

Continuous disclosure, including portfolio Annual and quarterly reports to SEC, semi-annual reports to shareholders (1934 Act, ICA § 30; SEC Rules 30a-1, 30b-1-5) UCITS: Annual audited and semi-annual reports to shareholders (Art. 27); additional information reported monthly to regulators

Regulatory approvals Regulatory approval of prospectus; mandatory advisory contract terms, subject to exemptions (1933 Act, Advisers Act § 205) UCITS: Prior approval required for fund, management company (MC) and its directors, depositary and its directors, fund rules, including fees, and prospectus disclosure and marketing materials; approval of new MC can take six months (Arts. 4, 5)

Individual agents required in home country No, except minimal requirement for agent for service of process (state law) Fund administration, MC, custodian must be in Luxembourg (Art. 4) MC (or self-managed VCC) must have two Irish directors; custodian must be in Ireland

Minimum capital $100,000 for regulated fund privately placed before registration (ICA § 14); third-party custodian must have sufficient financial resources under SEC rules (ICA § 17) UCITS: Manager must have capital equal to the greater of (a) 125,000 E.U.R for MC or (b) 0.02% of assets more than 250 MM E.U.R, up to 10 MM E.U.R, or, for self-managed funds, 300,000 E.U. (Art. 5a)

1.25 MM E.U.R for self-managed SICAV 635,000 E.U.R for all fund sponsors

Fund is separate legal entity Corporation: Yes Business or unit trust or separate account: No SICAV: Yes FCP: No VCC: Yes Unit trust: No

Third parties with duties to investors Mandatory board of directors, >40% independent (IDs) (ICA §§ 10, 16); separate custodian required (ICA § 17) UCITS: Mandatory depositary (institutional trustee or custodian) (Art. 4), separate from MC, required to act independently and solely in the interests of investors (Art. 10)

Appendix D. (Continued)

Adviser duties to investors Yes (ICA § 36(b), state law) MC answerable to investors for non-or improper fulfillment of obligations (Art. 15) Trustees or board of VCC have duties to investors (S.I. No. 211 Regs. 20, 39; common law)

Leverage restrictions ICA § 18: May borrow only For investment, 300% asset coverage, for instance, less than 33% debt-to-assets ratio For temporary purposes, < 5% of assets UCITS: < 10% of net assets on temporary basis, not for investment (Art. 36)

Board/shareholder/regulatory approvals Shareholder elect boards (ICA § 16); both approve Changes to fundamental policies Rule 12b-1 plans Adoption, amendment, and assignment of management contracts UCITS: Boards/shareholders not required, but regulators must approve any change in MC, depositary, or fund rules (Art. 4[4])

SICAV: Yes FCP: No Unit Trust: 75% unitholders approve significant changes in investment policies and manager compensation; VCC: Shareholders elect board and voting rights as in Unit Trust

Manager compensation and activity constraints No performance fees, except symmetric fulcrum fees (Advisers Act § 205); advisers may engage in other activities UCITS: MC compensation in "fund rules," approved by regulator (Art. 43); performance fees permitted; MC activities limited to fund management, administration, and related activities (Art. 5, Annex 2)

Diversification/con-centration constraints Diversified funds (ICA § 6): < 25% of fund assets in any one issuer < 25% of fund assets in 5+% blocks < 10% of shares issued by one issuer For all funds (IRC §§ 851-852): >50% in cash, gov't securities, and regulated fund securities < 5% of fund assets in single issuer < 10% of shares issued by any issuer UCITS (Art. 25): < 10% of net assets invested in any one issuer < 40% of net assets invested in 5+% blocks No significant influence over any issuer < 10% of nonvoting shares of debt of any issuer

Appendix D. (Continued)

Restrictions on non-standard investments ICA § 12(d)(1), subject to Rule 12d-1: < 3% of shares issued by other RIC < 5% of assets in other RIC < 10% of assets in all other RICs ICA Rels. No. 5847 (Oct. 21, 1969) and 18612 (Mar. 12, 1992): < 15% of net assets in illiquid securities UCITS (Art. 22): < 10% in non-listed or illiquid securities May not invest in commodities or hedge funds < 20% in any other fund < 30% in non-UCITS funds May not invest in another fund which invests > 10% in funds (no "funds of funds of funds")

Redeemable shares Not required (although regulated if used) UCITS: Required (Arts. 1, 2)

Multiple share classes Yes Yes

Risk management Disclosure of risks but no policy required UCITS: Policy required (Art. 21)

Compliance Compliance policy; chief compliance officer reports to fund board (SEC Rule 38a-1) UCITS: No comparable requirement; Member states require policies (for example, IFSRA UCITS Notices Art. 35)

Conflict of interest constraints Strict prohibition, subject to regulatory exemptions (ICA § 17) UCITS: MC may only engage in fund management and ancillary activities and must be "structured and organised" to minimize conflicts of interest (Art. 5)

No detailed prohibitions; MC must act in "exclusive interest" of investors (Art. 14); MC to ensure clients are "fairly treated" (Art. 86); fund industry has a non-binding code of conduct No detailed prohibitions; common law duties may apply

Public enforcement SEC, DOJ, states CSSF IFSRA

Private enforcement Yes (ICA § 36(b), state law) No Not in practice