A Loss Cause: Fixing the Rules Governing Mutual Fund Losses

Reg. section 1.382-2T(e)(1)(i) defines an owner shift as any change in the ownership of the stock of a loss corporation that affects the percentage of that stock owned by a 5 percent shareholder (defined later, but generally someone who owns at least 5 percent of the loss corporation’s stock at any point during the prescribed testing period) of the loss corporation, including changes resulting from stock issuances and redemptions that affect a 5 percent shareholder’s ownership percentage.

B. Testing Period

The loss limitation rules test whether an ownership change occurs during a testing period. Under reg. section 1.382-2T(d)(1), the testing period generally is the three-year period ending on the testing date. If a loss corporation has been in existence for less than three years, the testing period is generally the life of the loss corporation. The testing period also may be less than three years if a prior ownership change occurred less than three years before the testing date or if the NUBIL or loss carryforwards that made the corporation a loss corporation first arose less than three years before the testing date.¹⁷

C. Turnover in Stock Ownership

Having now established when and over what period to test, we turn to the most complex area of the loss limitation rules: how much turnover in the ownership of a corporation’s stock during the testing period triggers application of the rules. Generally, the loss limitation rules focus on 5 percent shareholders in determining the amount of turnover; if there is too much turnover among these shareholders, an ownership change occurs and the rules will apply. More specifically, under reg. section 1.382-2T(a)(1), an ownership change occurs if the percentage of a loss corporation’s stock owned by one or more 5 percent shareholders as of the close of the testing date has increased by more than 50 percentage points over the lowest percentage of that stock owned by those shareholders during the testing period. A corporation applies the test as follows:

1. Determine the percentage of stock each 5 percent shareholder (that is, a 5 percent shareholder at any point during the testing period) owned as of the close of the testing date.

2. Compare that amount with the lowest percentage of stock that the 5 percent shareholder held during the testing period.

3. For each of the 5 percent shareholders for whom (1) is greater than (2) (that is, whose percentage of stock owned at the close of the testing date exceeds its lowest stock ownership percentage during the testing period), calculate the difference.

4. Add the percentages determined in (3). If the sum exceeds 50 percent, an ownership change has occurred, and the loss limitation rules will apply.

The definition of 5 percent shareholder is therefore critical in determining whether an ownership change has occurred.¹⁸ Moreover, because the definition of 5 percent shareholder is key in determining owner shifts and testing dates, identification of 5 percent shareholders is actually the first step in the overall analysis. Reg. section 1.382-2T(g)(1) provides that a 5 percent shareholder is an individual who, at any time during a testing period, owns 5 percent or more of a corporation, either directly or indirectly (for example, when Individual X owns 10 percent of Corp. A, and Corp. A owns 50 percent of Corp. B, X is a 5 percent shareholder of Corp. B).¹⁹ Further, these regulations provide generally that all of a corporation’s direct shareholders that own less than 5 percent of the corporation’s stock at all times during the testing period are treated as a single entity (a public group) that itself is treated as a 5 percent shareholder (even if this group does not own 5 percent of the corporation’s stock in aggregate).²⁰

Consequently, if no individual or entity directly owns 5 percent or more of a loss corporation’s stock at any time, the regulations treat the corporation as having a single 5 percent shareholder at all times (that is, the public group). Because the public group would own 100 percent of the loss corporation’s stock at all times under those circumstances, there would be no owner shift and thus no testing date.

The identification of 5 percent shareholders becomes more complicated if, at any point during the testing period, one or more entities (as opposed to individuals) directly own 5 percent or more of a loss corporation’s stock. In those scenarios, there may be multiple, distinct public groups, each of which is considered a 5 percent shareholder.²¹ A multistep process is required to identify 5 percent shareholders in this context. To illustrate the mechanics of this process, consider the corporate structure in Figure 1 (for simplicity, assume that the ownership percentages are constant throughout the testing period)²²:

1. Step 1: Classification.

For purposes of this identification analysis, the loss corporation is considered to be at the bottom of a structural ownership chart. The initial step is to classify the entities and shareholders in the ownership structure as follows:

• An entity that directly owns 5 percent or more of the loss corporation at any time during the potential testing period is considered a first-tier entity.²³ In Figure 1, only Corp. A is a first-tier entity.

• An entity that directly owns 5 percent or more of a first-tier entity (but does not necessarily indirectly own 5 percent or more of the loss corporation) at any time during the testing period is a second-tier entity, an entity that directly owns 5 percent or more of a second-tier entity (but does not necessarily indirectly own 5 percent or more of the loss corporation) at any time during the testing period is a third-tier entity, and so on (numerically designated entities above the first tier are referred to as higher-tier entities).²⁴ Corp. C and Corp. D are higher-tier entities.

• A highest-tier entity is a first- or higher-tier entity that is not owned by a higher-tier entity (that is, the entity does not have another entity owning 5 percent or more of its stock directly) at any time during the testing period.²⁵ Corp. D is the only highest-tier entity (although Corp. B is not owned by a higher-tier entity, it is not a highest-tier entity because it is neither a first-tier entity nor a higher-tier entity).

• A 5 percent owner is any individual who, at any time during the testing period, owns 5 percent or more directly of a first-tier entity or higher-tier entity.²⁶ T and U are 5 percent owners. (R is not a 5 percent owner because Corp. B is not a first- or higher-tier entity.)

• A public owner is an individual or entity that, at all times during the testing period, owns less than 5 percent directly of a first- or higher-tier entity.²⁷ S, V, and W are public owners.

• A public shareholder is an individual or entity with a direct ownership interest in the loss corporation of less than 5 percent at all times during the testing period.²⁸ Corp. B, Y, and Z are public shareholders.

An individual who directly owns 5 percent or more of the loss corporation at any time during the potential testing period can be labeled as a 5 percent shareholder at this juncture. The loss corporation’s public shareholders can also be labeled collectively as a 5 percent shareholder (that is, a direct public group) at this juncture, regardless of whether this group owns 5 percent of the corporation’s stock in aggregate at any time during the potential testing period. In Figure 1, X is a 5 percent shareholder, while Corp. B, Y, and Z compose a direct public group that is also a 5 percent shareholder.²⁹

2. Step 2: Constructive ownership.

The next step in the identification process is to apply the constructive ownership rules of reg. section 1.382-2T(h), under which loss corporation stock owned by a first-tier entity is generally treated as owned proportionately by the first-tier entity’s owners. Any loss corporation stock that is thereby treated as owned by a higher-tier entity is generally treated as owned proportionally by that higher-tier entity’s owners, and so on up the structural ownership chart. However, the constructive ownership rules do not apply to all entities. Under reg. section 1.382-2T(h)(2)(iii)(A), an entity that is not (1) a higher-tier entity that owns, directly or indirectly, 5 percent or more of the loss corporation on a testing date, (2) a first-tier entity, or (3) the loss corporation is treated as an individual who is unrelated to any other direct or indirect owner of the loss corporation. Stock owned by an entity described in that sentence therefore is not attributed to any other person (absent actual knowledge). In Figure 1, Corp. B does not fall in any of these listed categories and is therefore treated under reg. section 1.382-2T(h)(2)(iii)(A) as an individual.³⁰

The most basic consequence of reg. section 1.382-2T(h)(2)(iii)(A) applying to an entity is that loss corporation stock owned by that entity is not attributed to any of that entity’s shareholders. This prevents any such shareholder (or group of shareholders) from having to aggregate loss corporation stock that they own indirectly through the entity with any other direct or indirect holding in the loss corporation it (or they) may own. Reg. section 1.382-2T(h)(2)(iii)(A) accomplishes this by treating the entity as an individual who is unrelated to any other owner of the loss corporation (that is, the type of person from whom there is no attribution). As such, the entity should be treated as an individual shareholder of the next-lower-tier entity and, as a less-than-5-percent (indirect) shareholder of the loss corporation, as a member of the next-lower-tier entity’s public group.

If the constructive ownership rules do apply to an entity, attribution will take one of two forms:

1. If an owner of the entity to which the constructive ownership rules are being applied (Entity) is (1) a higher-tier entity (that is, an entity that owned 5 percent or more of Entity at any time during the testing period); or (2) a 5 percent owner of Entity (that is, an individual who owned 5 percent or more of Entity at any time during the testing period), the owner is deemed to own its ratable share of the stock of the loss corporation held by Entity. In Figure 1:

The results of steps 1 and 2 are displayed in Figure 2:

3. Step 3: Identify 5 percent shareholders.

With stock attribution complete, the identification of 5 percent shareholders of the loss corporation (who, as noted previously, cannot be entities) can commence. The identification analysis begins at the top of the structural chart and works its way down³²:

1. Starting at the top of the structural chart, as a result of applying the definitions in step 1, the entity at the top may or may not be a highest-tier entity. Because, by definition, a highest-tier entity must be either a first- or higher-tier entity, an entity will qualify as a highest-tier entity only if it owns 5 percent or more of the loss corporation at any time during the testing period.³³ If the entity does not so qualify, there is no obligation to determine its ownership.³⁴ In this case, the shareholders of the top entity would compose a public group and, as discussed later, would be pushed down to the next-lowest-tier entity. Step 3(I) would then be repeated until a highest-tier entity is identified. In Figure 2 the entity at the top, Corp. D, is a highest-tier entity.

2. If the top entity (or another entity) is a highest-tier entity, any of its shareholders that were identified as a 5 percent owner in step 1 and indirectly own at least 5 percent of the loss corporation’s stock through its interest in the highest-tier entity at any time during the testing period is a 5 percent shareholder of the loss corporation. In Figure 2 T meets these criteria and is therefore a 5 percent shareholder.

3. The remaining shareholders of the highest-tier entity, if any (including the group of public owners from step 2), are considered a single public group.³⁵ If this public group indirectly owns at least 5 percent of the loss corporation in the aggregate on the testing date (rather than at any time during the testing period),³⁶ it is considered a discrete 5 percent shareholder on the testing date. If this public group indirectly owns less than 5 percent of the loss corporation on the testing date, it is pushed down and treated as part of the public group of the next-lower-tier entity (the push-down rule).³⁷ In Figure 2 Corp. D has no shareholders other than T.

4. Step 4.

The analysis in steps 3(II) and (III) is then repeated at the next-lower-tier entity level (replacing highest-tier entity with the higher- or first-tier entity, as applicable), with two modifications:

• When repeating step 3(III) and forming the public group at the next-lowest-tier level, a next-higher-tier entity that is not subject to the constructive ownership rules in step 2 should be considered a “remaining shareholder” that will be included in that public group.³⁸ A next-higher-tier entity that is subject to the constructive ownership rules should not be considered a remaining shareholder and should not be included in the public group at this next-lowest-tier level.

• The public group at this next-lowest-tier level may include members of the public group from the next-highest-tier entity if they were pushed down. If a public group is not pushed down, multiple public groups that are 5 percent shareholders may be identified as the analysis is repeated at each level.

In Figure 2 Corp. C is the next-lower-tier entity in the chain. Repeating step 3(II), U is a 5 percent shareholder. Under step 3(III), Corp. C Public Group is subject to the push-down rule and therefore is treated as part of the public group of the next-lower-tier entity.

The analysis continues all the way down the structure to the loss corporation. In Figure 2 Corp. A is the next-lower-tier entity. Step 3(II) does not identify any new 5 percent shareholders. Under step 3(III), Corp. C Public Group’s holdings are added to those of Corp. A Public Group. Because Corp. A Public Group therefore owns 7.9 percent of Loss Corp., Corp. A Public Group is a 5 percent shareholder.

At the loss corporation level, we have already identified in step 1: (1) any individual who directly owns 5 percent or more of the loss corporation at any time during the testing period as a 5 percent shareholder (X in Figure 2) and (2) the direct public group.³⁹ As noted earlier, the direct public group of the loss corporation (including members of any pushed down public groups) is treated as a 5 percent shareholder regardless of its actual level of ownership.

Therefore, in Figure 2 the 5 percent shareholders are T, U, X, Corp. A Public Group, and Direct Public Group.

Note that despite the general approach to identifying public groups, there appears to be uncertainty about the treatment of an indirect public group that was a 5 percent shareholder during the testing period but owns less than 5 percent of the loss corporation on the testing date, and vice versa. Reg. section 1.382-2T(g)(5)(i) provides that a loss corporation may make specific presumptions⁴⁰ regarding an individual who either becomes or ceases to be a 5 percent shareholder during a testing period:

• Under reg. section 1.382-2T(g)(5)(i)(A), if an individual who owns less than 5 percent of a loss corporation’s stock during a testing period acquires sufficient stock to be a 5 percent shareholder on the testing date, the loss corporation may disregard the interest that the individual owned in the loss corporation before the acquisition and instead treat the interest as owned by “a public group,” which presumably is the public group that the individual belonged to before becoming a 5 percent shareholder.⁴¹ The practical effect of this presumption is to make the individual’s lowest percentage of ownership during the testing period 0 percent. In most cases, this presumption will exaggerate the total increase in ownership by 5 percent shareholders. A loss corporation still may apply the presumption because it allows the loss corporation to avoid expending the effort to determine that individual’s historic ownership (which it previously had no reason to track) and back it out of the public group in determining the public group’s low point during the testing period. In some circumstances, the presumption may reduce the total increase in ownership by 5 percent shareholders.⁴²

• Under reg. section 1.382-2T(g)(5)(i)(B), if a 5 percent shareholder’s⁴³ interest in a loss corporation falls below 5 percent during a testing period, the loss corporation may treat the stock owned by the 5 percent shareholder immediately after the reduction as frozen at that amount for the duration of the testing period (provided the shareholder continues to own less than 5 percent of the loss corporation).⁴⁴ Under the presumption, the frozen stock is considered to be owned by a separate public group, but only for limited purposes.⁴⁵

Reg. section 1.382-2T(j)(1)(v) appears to permit the same assumptions to be made for public groups. Note that if the presumption of reg. section 1.382-2T(g)(5)(i)(B) does lead to the creation of a separate public group (whether for limited purposes or more generally), this result seems to contravene the push-down rule.⁴⁶

An alternative interpretation of reg. section 1.382-2T(j)(1)(v)(B) is that it affects the amount of stock that is pushed down but does not prevent the push-down from occurring in the first place. Although the basic ownership change test in reg. section 1.382-2T(a)(1) requires each 5 percent shareholder to be considered, reg. section 1.382-2T(j)(1)(iv)(A) states that a public group (PG1) that owns less than 5 percent of the loss corporation on the testing date is treated as part of the public group of the next-lower-tier entity (PG2), apparently even if PG1 was a 5 percent shareholder earlier in the testing period. Under this view, when reg. section 1.382-2T(j)(1)(v)(B) provides for “appropriate adjustments” in those circumstances in accordance with reg. section 1.382-2T(g)(5)(B), that adjustment would affect only the amount of stock that is deemed to be pushed down to PG2 and would not prevent the push-down from occurring in the first place.

At this point, the identification of 5 percent shareholders is complete, so it is now possible to determine if a testing date has occurred. Assuming a testing date has occurred, the change in the ownership of the 5 percent shareholders during the applicable testing period is then calculated as described earlier to determine if an ownership change has occurred.

D. Determining Which Losses Are Restricted

If an ownership change occurs, the loss corporation’s ability to use its pre-change losses will be restricted. Accordingly, a loss corporation that experiences an ownership change must identify its pre-change losses. When doing so, it looks to the same four basic categories of losses as it did when determining if it was a loss corporation: (1) NOL carryforwards and CLCs, (2) NUBIL, (3) NOL and net realized loss, and (4) a carryforward of disallowed interest expense described in section 163(j)(2).⁴⁷

• NOL carryforwards and CLCs as pre-change losses: Under reg. section 1.382-2(a)(2), if a loss corporation that experiences an ownership change has any CLCs or NOL carryforwards to the year that includes the date of the ownership change, those carryforwards are pre-change losses and their use is restricted.

• NUBIL as pre-change loss: Under reg. section 1.382-2(a)(2) and section 382(h), if a loss corporation that experiences an ownership change has NUBIL immediately before the ownership change, that NUBIL will generally be treated as a pre-change loss and be restricted to the extent that it is recognized in a tax year any portion of which is in the five-year period beginning on the date of the ownership change.⁴⁸

• NOL and net capital loss as pre-change loss: Under reg. section 1.382-2(a)(2), any NOL or net capital loss allocable to the pre-change part of the year that includes the ownership change is treated as a pre-change loss, and its use is restricted. If a loss corporation has an NOL or a net capital loss for the year that includes the ownership change, it can determine the portion of that loss that is allocable to the pre-change part of the year (and thus is restricted) in one of two ways. The default rule is that the portion of an NOL or net capital loss that is allocable to the pre-change part of the year is determined by ratably allocating an equal portion to each day of the year.⁴⁹ Alternatively, a loss corporation can elect to determine the portion of an NOL or net capital loss that is allocable to the pre-change part of the year as if it closed its books on the date of the ownership change.⁵⁰ If the loss corporation makes that election, its pre-change NOL or pre-change net capital loss will be the lesser of (1) the amount of NOL or net capital loss (as applicable) actually recognized before the ownership change or (2) the total amount of NOL or net capital loss (as applicable) for the entire tax year.⁵¹

E. The Restriction

If a corporation experiences an ownership change, the amount of pre-change losses it can use in any year to offset income or capital gains generally is limited to the product of the value of its stock immediately before the ownership change (less some adjustments) and the long-term tax-exempt rate, which the IRS publishes monthly.⁵² Note that in determining the value of the stock, section 382(l)(4) requires most corporations with substantial nonbusiness assets to reduce their FMV by the amount of those assets. RICs, however, are not subject to this provision.⁵³

F. RICs and the Loss Limitation Rules: Overview

The preceding discussion of the loss limitation rules focused on a static corporate structure. In reality, however, corporate stock ownership is dynamic — shareholders buy and sell stock, and corporations issue and redeem stock. The rules discussed earlier generally address both categories of corporate changes by evaluating changes in a 5 percent shareholder’s ownership over the testing period. The loss limitation rules contain additional provisions addressing stock issuances and redemptions under which these transactions can create additional public groups.⁵⁴ Very generally, if a loss corporation issues or redeems stock, a public group may be segregated into multiple public groups.⁵⁵ Even without delving into the intricacies of the segregation rules, it should be apparent that a RIC would face significant administrative difficulty in applying these rules because, although stock issuances and redemptions are one-off events for most C corporations, they are daily events for most RICs that are open-end investment companies.⁵⁶

Just as the loss limitation rules provide a special rule for RICs for valuation,⁵⁷ so too do they create a special rule for RICs for segregation (the RIC exception).⁵⁸ Specifically, reg. section 1.382-3(k)(1) provides that the segregation rules do not apply to a RIC’s issuance and redemption of its shares in the ordinary course of its business.⁵⁹ The RIC exception is not a panacea, however. As discussed in Section IV, RICs may experience ownership changes as a result of ordinary course redemptions because of how the loss limitation rules’ treatment of some types of retirement plans that are RIC shareholders.

Further, RICs may have multiple public groups as a result of ordinary course issuances and redemptions if the multiple public groups arise because of application of the aggregation rules of reg. section 1.382-2T(j)(1).⁶⁰ As a technical matter, the RIC exception applies to only the segregation rules under reg. section 1.382-2T(j)(2) and not to the aggregation rules under reg. section 1.382-2T(j)(1). Under these segregation rules, a loss corporation’s issuance or redemption of its shares may result in the creation of multiple public groups that would not be multiple public groups under the aggregation rules (for example, multiple public groups that are direct shareholders of the loss corporation).⁶¹ The RIC exception prevents such multiple public groups from arising. In contrast, the provisions creating multiple public groups under the aggregation rules of reg. section 1.382-2T(j)(1) are not subject to the RIC exception. As discussed in Section V, it is these public groups that create potential problems for RICs under the loss limitation rules because of a sponsor’s contribution of seed money.

IV. Retirement Plan Shareholders

As discussed in Section III, the loss limitation rules key off the concept of 5 percent shareholders. In general, the more 5 percent shareholders a corporation has, the more likely it is that the corporation will experience an ownership change. A rule that treats an entity as a single individual increases the possibility that a corporation will have 5 percent shareholders. For the most part, the loss limitation rules apply a look-through approach under section 382(l)(3) and reg. section 1.382-2T(h) to entities such as corporations, partnerships, and trusts (the look-through approach may result in the creation of indirect public groups, which may or may not increase the number of 5 percent shareholders). For retirement plans, though, the loss limitation rules treat some types of these plans as discrete individuals.⁶²

Retirement plans, which are major investors in RICs, may be categorized based on several characteristics. For example, retirement plans often are classified as defined benefit plans or defined contribution plans. In a defined benefit plan, a participant’s future benefit is “defined” based on a formula that typically is tied to the participant’s length of service and salary, so that, for example, a participant who works for X years at an annual salary of $Y will receive a defined annual benefit in retirement of $Z. In a defined contribution plan, participants (and sometimes the employer) contribute defined amounts to accounts that are established within the plan for each participant. The benefit available in retirement depends on the investment performance of the assets in a participant’s account.

Retirement plans also may be classified based on who makes the investment decisions. In a participant-directed retirement plan, each participant makes their own decisions regarding the purchase and sale of investments. If a retirement plan is not participant-directed, the plan trustee or plan sponsor makes the investment decisions. It would be highly unusual for a defined benefit plan to be participant-directed. In contrast, a defined contribution plan may or may not be participant-directed.⁶³

In the context of the loss limitation rules, a retirement plan’s most significant characteristic is whether it is a qualified trust described in section 401(a), which, under reg. section 1.382-2T(h)(2)(iii)(B), is treated as an individual who is unrelated to any other owner (direct or indirect) of the loss corporation.⁶⁴ Reg. section 1.382-2T(h)(2)(iii)(C) provides an analogous rule for any U.S. state, any U.S. possession, the District of Columbia, the United States (or any agency or instrumentality thereof), any foreign government, or any political subdivision of any of the foregoing.⁶⁵ That provision may or may not apply to section 457 retirement plans.

The applicability of reg. section 1.382-2T(h)(2)(iii)(C) to a section 457 plan depends on whether, for purposes of that regulation, section 457 plans are considered an integral part of a governmental body or an instrumentality. Section 115, which exempts section 457 plans from tax, provides in part that gross income does not include income derived from “the exercise of any essential governmental function and accruing to a State or any political subdivision thereof” (emphasis added). This suggests that section 457 plans are treated as part of, rather than independent from, a governmental body, assuming the provision of retirement benefits is an essential governmental function. Even if section 457 plans were not subject to reg. 1.382-2T(h)(2)(iii)(C), however, they would be subject to the segregation rules, so that plan participants would be treated as a separate public group under reg. section 1.382-2T(j)(1) if the plan were a 5 percent shareholder. Consequently, section 457 plans may be subject to adverse consequences regardless of whether they are subject to reg. section 1.382-2T(h)(2)(iii)(C).⁶⁶

These exceptions apply regardless of whether the qualified trust or section 457 plan is participant directed. Consequently, a qualified trust or section 457 plan will be treated as a single individual even though each of its participants makes their own decision about whether to purchase or sell stock of a loss corporation.

Section 401(a) defines a qualified trust as a trust that is created as part of a stock bonus, pension, or profit-sharing plan of an employer for the exclusive benefit of its employees or beneficiaries, provided several conditions are met. Many types of conventional retirement plans are qualified trusts, including:

• section 401(k) plans;

• Keogh plans (available to self-employed individuals);

• employee stock ownership plans;

• traditional employer-provided profit-sharing plans; and

• traditional employer-provided defined benefit plans.

Types of conventional retirement plans that are not qualified trusts include:

• IRAs;

• section 403(b) plans (available to employees of some charitable and educational organizations); and

• section 457 plans (available to governmental employees and employees of some tax-exempt organizations).⁶⁷

The adverse consequences of the loss limitation rules’ treatment of qualified trusts and section 457 plans as a single individual are exacerbated by the way these entities select and eliminate RICs as plan investment options. Under ERISA, either the plan sponsor or the plan trustee bears responsibility for selecting investment options available to plan participants (at least some of those options are typically RICs). Each participant then selects their own individual investments.

Under some circumstances, the plan sponsor or plan trustee may determine that the addition or removal of an investment option is in the best interest of plan participants. When an investment option is removed, the plan sponsor or trustee notifies the plan participants, who must select a different investment option in which to reinvest the proceeds from the discontinued option.

If a qualified trust or section 457 plan is a 5 percent shareholder of the discontinued option, and the discontinued option is a loss corporation, the discontinuance will trigger a testing date and may result in an ownership change. Also, if plan participants reinvest the proceeds from the discontinued option by acquiring in aggregate at least 5 percent of a loss corporation, the reinvestment will trigger a testing date and potentially result in an ownership change.

Even if the discontinuance or reinvestment does not result in an immediate ownership change, it may cause one in the future. Suppose, for example, that participants in a participant-directed qualified trust or section 457 plan acquire more than 5 percent but less than 50 percent (by value) of an open-end RIC and that the RIC’s only other 5 percent shareholder during the testing period is its direct public group. The RIC could experience an ownership change during the ensuing three years under the following circumstances:

• The plan increases its investment. The plan continues to invest in the RIC while the direct public group maintains the number of shares it owns, reduces the number of shares it owns, or purchases a disproportionately small number of shares relative to purchases by the plan.

• The direct public group redeems a significant portion of its investment. The direct public group redeems shares while the plan maintains the number of shares it owns, increases the number of shares it owns, or redeems a disproportionately small number of shares relative to redemptions by the direct public group.

• Other investors become 5 percent shareholders. Other investors become 5 percent shareholders through purchases, redemptions by other shareholders, or mergers⁶⁸ in which the RIC is either the acquiring or acquired fund.

• Any combination of the foregoing.

The RIC’s investment manager may be able to take steps to reduce the likelihood that these events will occur, but there is no guarantee those steps will work. For example, the investment manager could close the RIC to new investments, but redemptions by existing investors other than the plan could trigger an ownership change (an open-end RIC generally cannot suspend redemption rights for more than a short period).⁶⁹ The investment manager could not force the plan to redeem all or a portion of its shares. Even if that were possible, the redemption would result in a commensurate increase in the ownership percentage of any other 5 percent shareholders.

When it was adopted, reg. section 1.382-T(h)(2)(iii)(B) may well have been intended to be a taxpayer-favorable rule of convenience for classic defined benefit plans (which were far more prevalent at the time), in which it is impossible to determine beneficial ownership by participants. The rule seems taxpayer-favorable because its application is subject to an antiabuse provision.⁷⁰ Nevertheless, it soon became apparent that the regulation could produce adverse consequences under the loss limitation rules.

In three private letter rulings in the 1990s,⁷¹ the IRS addressed situations in which an ESOP that owned stock in a loss corporation distributed that stock to plan participants. Because an ESOP is a qualified trust, a literal reading of reg. section 1.382-2T(h)(2)(iii)(B) indicates that the distributions represented transfers of stock from one individual (the ESOP) to others (the plan participants) that could trigger application of the loss limitation rules. For example, if an ESOP (or other qualified trust) distributed to its participants stock of a loss corporation that it had acquired more than three years before the distribution, an ownership change could occur even if none of the participants individually acquired 5 percent or more of the loss corporation’s stock.⁷² In results-oriented rulings, however, the IRS concluded in each case that the ESOP participants had been direct owners of the stock all along, so the distributions did not result in an ownership change.

Somewhat curiously, the IRS cited reg. section 1.382-2T(k)(2) in LTR 9122060 and LTR 9652019 to support this ruling (the IRS provided no rationale for its ruling in LTR 9510007). As noted earlier, reg. section 1.382-2T(h)(2)(iii)(B) applies “except as provided in paragraphs (k)(2) and (k)(4)” of the -2T regulation. Reg. section 1.382-2T(k)(2) states in relevant part (and stated at the time of the rulings):

To the extent that the loss corporation has actual knowledge of stock ownership on any testing date (or acquires such knowledge before the date that the income tax return is filed for the tax year in which the testing date occurs) by an individual who would be a 5-percent shareholder, but for the application of paragraphs (h)(2)(iii) . . . the loss corporation must take such stock ownership into account for purposes of determining whether an ownership change has occurred on that testing date. . . . To the extent the loss corporation has actual knowledge on or after any testing date regarding the ownership interest in the loss corporation by members of one public group . . . and the ownership interest of those members in the loss corporation as members in another such public group, the loss corporation may take such ownership into account for purposes of determining whether an ownership change occurred on that testing date.

It is difficult to interpret this provision in a way that supports the IRS’s rulings that effectively disaggregate an ESOP’s holdings to prevent the ESOP from being a 5 percent shareholder. Reg. section 1.382-2T(k)(2) requires aggregation of an individual’s holdings if there is actual knowledge. Further, the provision applies only when an individual would be a 5 percent shareholder as a result of that aggregation. By contrast, the ESOPs in the private letter rulings were 5 percent shareholders without aggregation, and it seems likely that no 5 percent shareholders were created by disaggregating the holdings of those ESOPs.

The IRS and Treasury ultimately recognized that reg. section 1.382-2T(k)(2) provided inadequate support to conclude that ESOP participants were the pre-distribution owners of the plan’s stock. In a highly public transaction in 2003, bankrupt United Airlines terminated its ESOP and distributed its stock to plan participants.⁷³ Noting that “the participants under a qualified plan are not treated as owning any interest in a loss corporation owned by a trust”⁷⁴ because of reg. section 1.382-2T(h)(2)(iii)(B), Treasury and the IRS issued reg. section 1.382-10T (now reg. section 1.382-10). That regulation treats participants who receive a distribution of stock from a qualified plan as if they — not the qualified trust — had acquired the distributed stock on the date the qualified trust acquired it. Consequently, the participants owning less than 5 percent of the stock before and after the distribution are treated as members of the loss corporation’s direct public group both before and after the distribution. As a result, the distribution should not create an ownership change. (Note that the regulation does not address an identical distribution by a section 457 plan; those distributions may be rare, however.)

The willingness of Treasury and the IRS to remedy an inequitable situation for ESOPs is commendable. A similar initiative to remedy the inequitable treatment of qualified trusts and section 457 plans described earlier would be most welcome. The loss limitation rules arbitrarily differentiate retirement plans between those that are treated as a single individual and those that are not. That differentiation is not based on whether participants make investment decisions (which would be a logical way to differentiate among plans) but rather on the status of the employer as private, charitable/tax exempt, or governmental. Consequently, a qualified trust or section 457 plan will be treated as a single individual even though each of its participants makes their own decision about whether to purchase or sell stock of a loss corporation.

To remedy the problem, Treasury and the IRS should amend reg. section 1.382-3 to provide that participants in a participant-directed qualified trust or section 457 plan are treated at all times as direct owners of any RICs they own through the qualified trust or section 457 plan.⁷⁵ Exhibit 1 contains a draft amendment.

The proposed amendment is consistent with both the policy objectives and most of the existing provisions of the loss limitation rules. In addition to the objective constructive ownership rules in section 382(l)(3)(A) and reg. section 1.382-2T(h), reg. section 1.382-3(a)(1) contains a subjective ownership rule that treats a group of otherwise unrelated individuals as a single entity if they have a formal or informal understanding to make a coordinated acquisition of stock. In the absence of such an understanding, however, otherwise unrelated individuals are not to be treated as a group, a point illustrated in Example 3 of reg. section 1.382-3(a)(1)(ii).

Example 3 involves an investment adviser who advises his clients to invest in a particular stock. Twenty unrelated individuals purchase 6 percent of the stock, with each owning less than 5 percent. Because there is no formal or informal understanding among the clients to make a coordinated purchase, all the individuals are treated as part of the direct public group and no owner shift results. The example reaches the same result when a common trustee of several qualified trusts sponsored by unrelated corporations causes the trusts to purchase stock. The IRS has issued rulings amplifying this point.⁷⁶

Considering that aggregation of holdings is not required when individuals receive the same investment advice and separately decide to make the same investment, and that aggregation of holdings is not required when a single trustee makes the same investment on behalf of multiple trusts based on an independent decision for each trust, it seems anomalous to require aggregation of holdings when multiple individuals are making independent investment decisions without receiving any investment advice. Yet this is exactly the treatment that reg. section 1.382-2T(h)(2)(iii)(B) and (C) imposes on participant-directed qualified trusts and section 457 plans, in which the only “advice” participants receive is a menu of investment options available under the plan — a far cry from the explicit recommendation or actual purchase made by the investment adviser and trustee in Example 3.

This treatment stands in marked contrast to the way the loss limitation rules treat participant-directed section 403(b) plans — participants are not aggregated and instead are treated as members of a loss corporation’s direct public group (assuming no participant is a 5 percent shareholder).⁷⁷ Yet in substance, these section 403(b) plans are virtually identical to participant-directed qualified trusts and section 457 plans. In each type of plan, a participant selects from a menu of investment options and makes their own investment decisions. In each type of plan, a participant has a separate account balance and receives periodic account statements. In each type of plan, the assets attributable to a participant are protected from the creditors of the other participants. Given the extent of these similarities, it seems illogical to treat qualified trusts and section 457 plans differently from 403(b) plans in testing for ownership changes.

Viewed in the larger context of the entire loss limitation rules, then, reg. section 1.382- 2T(h)(2)(iii)(B) stands as a curious anomaly whose purpose is not entirely clear. The legislative history to the Tax Reform Act of 1986 does not refer to the exception under reg. section 1.382-2T(h)(2)(iii)(B) treating a qualified trust as an individual who is unrelated to any other owner (direct or indirect) of the loss corporation,⁷⁸ nor does there appear to be any discussion of the exception in the Treasury decisions that that have accompanied the regulations issued under section 382. As noted earlier, the exception may have been intended as a rule of convenience for classic defined benefit plans, since it is impossible to determine the beneficial ownership of participants of such a plan because that ownership depends on many unknown factors. For a participant-directed qualified trust (and section 457 plan), however, each participant receives periodic statements indicating their exact investments. As a result, there is no need in this context for a rule of convenience, particularly considering the draconian results it can produce.

The proposed amendment to reg. section 1.382-3 also generally is consistent with the overriding policy of section 382: the prevention of trafficking in loss corporations. RICs cannot pass through losses to their shareholders, and shareholders therefore cannot use a RIC’s built-in losses or CLCs to shelter income and gains from other investments (although built-in losses a RIC can use may reduce taxable distributions of gains to shareholders). The potential for loss trafficking also does not exist because qualified trusts and section 457 plans are tax deferral vehicles and are generally funded with pretax dollars. As a result, qualified trusts and section 457 plans are not subject to current taxation on investments in RICs,⁷⁹ and participants pay tax on all distributions from a qualified trust or section 457 plan (except to the extent attributable to the recovery of any after-tax contributions), regardless of whether those distributions are attributable to income or gains. Consequently, participants in a qualified trust receive no tax benefit from investing in a RIC with built-in losses and CLCs. In fact, such an investment decreases taxable investors’ shares of these losses, potentially increasing the overall level of taxation associated with the RIC’s income and gains.

The proposed change to reg. section 1.382-3 also creates a measure of parity in declining markets between investors in RICs and investors who acquire securities directly. RICs allow investors to obtain professional investment management and asset diversification that would otherwise be available only to wealthy individuals who invest directly in stocks and bonds (direct investors). Unfortunately, when markets decline, the code and regulations put RIC investors at a significant disadvantage relative to direct investors. First, losses do not pass through a RIC and therefore cannot offset other gains. Thus, a RIC investor with multiple investments may have a net economic loss but a taxable gain in a given year. In contrast, a direct investor can offset losses from one investment against gains from another (unless a special rule applies). Second, a RIC’s ability to use CLCs and built-in losses may be severely restricted if it experiences an ownership change. The concept of ownership change is inapplicable to direct investors who are individuals. Although the proposed change to the regulation would not eliminate the first disparity in treatment, it would at least reduce the likelihood of the second disparity and would therefore create a measure of parity between RIC shareholders and direct investors at times when that parity is most needed.

V. Seed Capital and the Loss Limitation Rules

Section 14(a) of the 1940 act provides that a “registered investment company”⁸⁰ cannot offer its shares to the public unless it has a net worth of at least $100,000. The principal purpose of section 14(a) “is to require the promoters of a new investment company to invest $100,000 of their own funds in the enterprise (or to obtain such funds from a small group of financially responsible persons who have confidence in the promoters) before inviting the investing public at large to entrust its funds in the care of the company’s management.”⁸¹

RICs generally are organized as separate series of a given business trust, with each series treated as a separate corporation under section 851(g). Section 14(a) applies at the trust level, so that a sponsor must provide $100,000 of seed capital only when it establishes a new trust.⁸² Nevertheless, sponsors typically provide seed capital to each new series of existing trusts to ensure, consistent with the sponsors’ fiduciary duty, that a RIC’s assets are adequately diversified before the RIC offers its shares to the public.

Whether contributing seed capital to a new trust or a new series of an existing trust, sponsors typically provide amounts well exceeding $100,000. Still, the subsequent public offering of the RIC’s shares generally dilutes the sponsor’s ownership interest in the RIC significantly, from 100 percent to a much lower number, even if the sponsor continues to retain its interest in the RIC. As detailed later, it is this dilution that increases the risk that the loss limitation rules will apply to the RIC.

A. Scenario 1

1. RIC is a loss corporation at the time of initial investment by the public.

In Scenario 1 (illustrated in Figure 3), a RIC that uses a calendar tax year is organized January 1, 2020. On the same date, Sponsor, a C corporation, contributes seed money in exchange for 100 percent of the RIC’s shares. For simplicity, assume that only individuals own Sponsor’s stock and that no individual is a 5 percent owner of Sponsor. On February 29, 2020, the RIC begins to sell its shares to the public. At the end of the day, public individuals own 95.1 percent of the RIC shares, with no single individual owning 5 percent or more. Sponsor now owns 4.9 percent of the RIC. Also assume that on February 29, 2020, the RIC has NUBIL greater than the lesser of 15 percent of assets or $10 million and that on December 31, 2020, the RIC has no current-year NOL or net realized capital loss.

To determine if the loss limitation rules apply to the RIC, we need to ascertain whether an ownership change occurred February 29, 2020. The potential testing period is January 1, 2020, to February 29, 2020. We next evaluate the turnover in the ownership of the RIC’s stock.

2. Step 1: Classification.

The owners of Sponsor are public owners because each of them individually owns less than 5 percent of a first- or higher-tier entity (that is, Sponsor) at all times during the potential testing period. Sponsor is a first-tier entity because it owns 5 percent or more of a loss corporation (that is, the RIC) at some point during the potential testing period. The individuals who own direct interests in the RIC are public shareholders because each of them individually owns less than 5 percent of the RIC at all times during the potential testing period.

3. Step 2: Constructive ownership.

Because Sponsor is a first-tier entity, the RIC stock owned by Sponsor is attributed to Sponsor’s shareholders even though Sponsor does not own 5 percent or more of the RIC on the potential testing date.⁸³ Because all the owners of Sponsor are public owners, however, they do not have to be specifically identified and do not have to aggregate the RIC shares attributed to them in this context with other RIC shares they may own.⁸⁴

4. Steps 3 and 4: Identify 5 percent shareholders using the aggregation rules.

The owners of Sponsor, all of whom are public owners, are a public group (Indirect Public Group). Because Indirect Public Group owns less than 5 percent of the RIC on the potential testing date, it is not a 5 percent shareholder, even though Indirect Public Group did own 5 percent or more of the RIC at some point during the potential testing period.⁸⁵ As a result, Indirect Public Group is pushed down. The direct owners of the RIC (other than Sponsor), all of whom are public shareholders, are a separate public group (Direct Public Group) into which Indirect Public Group is pushed. A direct public group is always treated as a 5 percent shareholder (regardless of whether it owns 5 percent or more of the loss corporation during the potential testing period). In this case, Direct Public Group owns 100 percent of the RIC on the potential testing date. Direct Public Group therefore is the only 5 percent shareholder of the RIC during the potential testing period.

5. Determine if February 29, 2020, is a testing date.

Having identified the RIC’s 5 percent shareholders, we now can determine if February 29, 2020, is a testing date. Because the RIC has NUBIL greater than 15 percent of assets and $10 million February 29, 2020, it is a loss corporation on that date, and because the stock ownership of Direct Public Group (a 5 percent shareholder) changes on that date, an owner shift occurs. Consequently, February 29, 2020, is a testing date, and the testing period runs from January 1, 2020, to February 29, 2020.

6. Calculate the change in ownership of the 5 percent shareholders.

As noted earlier, Direct Public Group, the only 5 percent shareholder, owns 100 percent of the RIC on the testing date. Although it does not affect the outcome under the facts in this example, it is not clear whether the low point of Direct Public Group’s ownership includes the stock owned by Indirect Public Group as a result of the push-down rule or instead excludes it. In the former case, Direct Public Group’s low point would be 4.9 percent (that is, the sum of Direct Public Group’s low point in the absence of the push-down rule during the testing period (0 percent) and Indirect Public Group’s low point during the testing period (4.9 percent). In the latter case, the low point would be 0 percent.

A third possible interpretation, which would affect the outcome of the example, is that Indirect Public Group’s stock ownership and Direct Public Group’s stock ownership are aggregated throughout the testing period in determining the low point. In that case, the low point would be 100 percent and there would be no ownership change (and no shift whatsoever). This interpretation is not inconsistent with the language in reg. section 1.382-2T(j)(1)(iv) creating the push-down rule, which simply states that the indirect public group “is treated as part of the public group of the next lower tier entity” (without indicating whether it is treated as part of the public group of the next-lower-tier entity only after it drops below 5 percent or throughout the testing period). That language could be read to mean that in determining the low point, Indirect Public Group is treated as part of Direct Public Group at all times during the testing period. This is not the correct theoretical approach, however, given that it leads to a conclusion that there was no shift whatsoever in a scenario in which 95.1 percent of the RIC’s stock is changing hands. The remainder of this report assumes that this third possible interpretation is incorrect.

The correct theoretical answer is that the low point is 4.9 percent because ownership of 95.1 percent of the RIC’s stock is changing hands. The regulations contain no language that supports or challenges this result. Reg. section 1.382-2T(g)(5)(i), as made applicable to public groups by reg. section 1.382-2T(j)(1)(v), contains some presumptions regarding stock ownership, but unfortunately, they do not clarify matters:

• The first presumption applies to any public group that is treated as a 5 percent shareholder on the testing date if the public group, at any time during the testing period, was treated as a part of the public group of the next-lowest-tier entity (reg. section 1.382-2T(j)(1)(v)(A)). Indirect Public Group, however, is not a 5 percent shareholder on the testing date.

• The second presumption applies to any public group that is treated as part of the public group of a next-lower-tier entity if the public group, at any time during the testing period, was part of the public group of a higher-tier entity that was treated as a 5 percent shareholder and had a direct or indirect ownership interest in that next-lower-tier entity (reg. section 1.382-2T(j)(1)(v)(B)). Because an indirect public group must own 5 percent of a loss corporation’s stock on a testing date — rather than at any time during a testing period — to be treated as a 5 percent shareholder, the only way the second presumption could apply would be if there were multiple testing dates in the testing period and Indirect Public Group had owned 5 percent or more of the RIC on a prior testing date. That, however, is not the fact pattern described in this scenario.

Regardless of whether Direct Public Group’s low point is 0 percent or 4.9 percent, however, an ownership change has occurred because the increased ownership of 5 percent shareholders (that is, Direct Public Group) is at least 95.1 percent. Note that although reg. section 1.382-2T(e)(1)(ii) provides that “transfers” between members of some public groups are disregarded, this provision does not apply to Scenario 1 even though, in effect, a significant portion of the RIC’s stock is transferred from Indirect Public Group to Direct Public Group. Under reg. section 1.382-2T(e)(1)(i)(C) and (D), “transfers” includes issuances and redemptions that affect the percentage of loss corporation stock owned by a 5 percent shareholder. However, these transfers are disregarded only if the public groups were created under the segregation rules of reg. section 1.382-2T(j)(2) and (3)(iii) (subsection (j)(3) applies the segregation rules in subsection (j)(2) to first- and higher-tier entities). As noted earlier, the public groups in Scenario 1 were created under the aggregation rules of reg. section 1.382-2T(j)(1).

7. Determine which losses are restricted.

Because the RIC experienced an ownership change February 29, 2020, it must identify its pre-change losses. As stipulated in the facts, the RIC had NUBIL greater than the lesser of 15 percent of assets or $10 million immediately before the ownership change. That NUBIL thus represents potential pre-change losses and is subject to restriction (depending on whether and when it is recognized).⁸⁶ The RIC had no NOL or net realized capital loss for the 2020 tax year and had no CLCs or NOL carryforwards to the 2020 tax year, so the NUBIL is the only loss that is subject to restriction.

8. Analysis.

Overall, Scenario 1 illustrates how a sponsor’s contribution of seed money affects ownership changes. In this example, the ownership change is attributable solely to that contribution. If Sponsor had not contributed seed money, the RIC would not have sustained an ownership change, because Direct Public Group would have owned 100 percent of the RIC at all times.

B. Scenario 2

1. Sponsor is closely held.

In Scenario 1 Sponsor has no 5 percent owners. In some cases, however, the sponsor will be closely held. For example, assume the facts in Scenario 2 (as shown in Figure 4) are the same as in Scenario 1 except that Sponsor’s owners include a single individual (Individual) who owns 50 percent of Sponsor’s stock; the remaining 50 percent is owned by individuals, none of whom is a 5 percent owner.

As in Scenario 1, the potential testing period is January 1, 2020, to February 29, 2020, and the potential testing date is February 29, 2020.

2. Step 1: Classification.

Individual is a 5 percent owner of Sponsor. The remaining owners of Sponsor are public owners because each of them individually owns less than 5 percent of a first- or higher-tier entity (that is, Sponsor) at all times during the potential testing period. Sponsor is a first-tier entity because it owns 5 percent or more of a loss corporation (that is, the RIC) at some point during the potential testing period. The individuals who own direct interests in the RIC are public shareholders because each of them individually owns less than 5 percent of the RIC at all times during the potential testing period.

3. Step 2: Constructive ownership.

Because Sponsor is a first-tier entity, RIC stock owned by Sponsor is attributed to Sponsor’s shareholders even though Sponsor does not own 5 percent or more of the RIC on the potential testing date. Because the owners of Sponsor other than Individual are public owners, they do not have to be specifically identified, and the RIC does not have to aggregate the RIC shares attributed to them in this context with other RIC shares they may own. In contrast, Individual is a 5 percent owner of Sponsor and therefore is specifically identified and subject to that aggregation (however, Individual has no ownership interests in the RIC other than her indirect interest through Sponsor).

4. Steps 3 and 4: Identify 5 percent shareholders using the aggregation rules.

Individual is a 5 percent shareholder because she owned more than 5 percent of the RIC (that is, 50 percent of 100 percent) during the potential testing period. The remaining owners of Sponsor, all of whom are public owners, are a public group (Indirect Public Group). Because Indirect Public Group owns less than 5 percent of the RIC on the potential testing date, it is not a 5 percent shareholder (even though Indirect Public Group owned 5 percent or more of the RIC at some point during the potential testing period). As a result, Indirect Public Group is pushed down.

The direct owners of the RIC (other than Sponsor), all of whom are public shareholders, are a separate public group (Direct Public Group) into which Indirect Public Group is pushed. A direct public group is always treated as a 5 percent shareholder (regardless of whether it owns 5 percent or more of the loss corporation during the potential testing period). In this case, Direct Public Group owns 97.55 percent of the RIC on the potential testing date (that is, 95.1 percent, plus 50 percent of 4.9 percent). Direct Public Group and Individual therefore are the only two 5 percent shareholders of the RIC during the potential testing period.

5. Determine if February 29, 2020, is a testing date.

Because the RIC has NUBIL greater than the lesser of 15 percent of assets or $10 million, it is a loss corporation February 29, 2020. Because the stock ownership of Direct Public Group and Individual (each 5 percent shareholders) changes February 29, 2020, an owner shift occurs. Consequently, February 29, 2020, is a testing date, and the testing period runs from January 1, 2020, to February 29, 2020.

6. Calculate the change in ownership of the 5 percent shareholders.

As noted earlier, Individual and Direct Public Group are 5 percent shareholders of the RIC. Individual does not contribute to any potential ownership change because her holdings on the testing date (2.45 percent) are equal to her ownership low point during the testing period. Direct Public Group owns 97.55 percent of the RIC on the testing date. Although it does not affect the outcome under the facts in this example, it is not clear whether the low point of Direct Public Group’s ownership includes the stock owned by Indirect Public Group as a result of the push-down or instead excludes it. In the former case, Direct Public Group’s low point would be 2.45 percent. In the latter case, the low point would be 0 percent. Whether Direct Public Group’s low point is 0 percent or 2.45 percent, an ownership change has occurred because the increased ownership of 5 percent shareholders is at least 97.55 percent.

7. Analysis.

The result of Scenario 2 is the same as in Scenario 1 — the RIC sustains an ownership change solely because of the seed money contribution. If Sponsor had not contributed seed money, the RIC would not have sustained an ownership change, because Direct Public Group would have owned 100 percent of the RIC at all times.

C. Scenario 3

1. RIC is not a loss corporation at the time of the initial investment by the public but becomes one during the same tax year.

In the preceding scenarios, the RIC is a loss corporation on the date the public begins to invest. Scenario 3 illustrates that sponsor seed money contributions can cause an ownership change even if the RIC is not a loss corporation on that date but becomes one later in the same tax year. Although this effect is not unique to RICs, it exacerbates the inequity arising from ownership changes resulting solely from a sponsor’s contribution of seed money. The facts of Scenario 3 (illustrated in Figure 5) are as follows:

• January 1, 2020: The RIC is formed. It uses a calendar tax year. Sponsor, a C corporation, contributes seed money for RIC shares. Sponsor is owned exclusively by individuals, with no 5 percent owners.

• February 29, 2020: The RIC opens to the public. Public individuals hold 95.1 percent of RIC stock; no individual is a 5 percent owner. Sponsor now owns 4.9 percent of the RIC. The RIC has no NOL carryforwards, CLCs, current-year NOLs, or net realized capital loss, and no NUBIL.

• July 1, 2020: Another investment by public individuals in the RIC causes Sponsor’s ownership of the RIC to fall to 1 percent and the public’s ownership to increase to 99 percent, with no one individual being a 5 percent owner. The RIC has NUBIL greater than the lesser of 15 percent of the FMV of its assets or $10 million.

• December 31, 2020: Sponsor still owns 1 percent and the public owns 99 percent. The RIC has NUBIL greater than the lesser of 15 percent of the FMV of its assets or $10 million. The RIC has no current-year NOL or net realized capital loss.

To determine if the loss limitation rules apply to the RIC, we need to ascertain whether an ownership change occurred February 29, 2020, or July 1, 2020. As a preliminary matter, we can eliminate the former date as a testing date regardless of whether an owner shift has occurred, because the RIC is not a loss corporation at that time. We therefore focus on July 1, 2020, as a potential testing date. The potential testing period is January 1, 2020, to July 1, 2020.

2. Step 1: Classification.

The owners of Sponsor are public owners because each of them individually owns less than 5 percent of a first- or higher-tier entity (that is, Sponsor) at all times during the potential testing period. Sponsor is a first-tier entity because it owns 5 percent or more of a loss corporation (that is, the RIC) at some point during the potential testing period. The individuals who own direct interests in the RIC are public shareholders because each of them individually owns less than 5 percent of the RIC at all times during the potential testing period.

3. Step 2: Constructive ownership.

Because Sponsor is a first-tier entity, the RIC stock Sponsor owns is attributed to Sponsor’s shareholders even though Sponsor does not own 5 percent or more of the RIC on the potential testing date. And because all the owners of Sponsor are public owners, they do not have to be specifically identified and do not have to aggregate the RIC shares attributed to them in this context with other RIC shares they may own.

4. Steps 3 and 4: Identify 5 percent shareholders using the aggregation rules.

The owners of Sponsor, all of whom are public owners, are a public group (Indirect Public Group). Because Indirect Public Group owns less than 5 percent of the RIC on the potential testing date, it is not a 5 percent shareholder (even though Indirect Public Group did own 5 percent or more of the RIC at some point during the potential testing period). As a result, Indirect Public Group is pushed down. The direct owners of the RIC (other than Sponsor), all of whom are public shareholders, are a separate public group (Direct Public Group) into which Indirect Public Group is pushed. A direct public group is always treated as a 5 percent shareholder (regardless of whether it owns 5 percent or more of the loss corporation during the potential testing period). In this case, Direct Public Group owns 100 percent of the RIC on the potential testing date. Direct Public Group therefore is the only 5 percent shareholder of the RIC during the potential testing period.

5. Determine if July 1, 2020, is a testing date.

Because the RIC has NUBIL greater than the lesser of 15 percent of assets or $10 million, it is a loss corporation July 1, 2020. Because the stock ownership of Direct Public Group (a 5 percent shareholder) changes July 1, 2020, an owner shift occurs. Consequently, July 1, 2020, is a testing date, and the testing period runs from January 1, 2020, to July 1, 2020.

6. Calculate the change in ownership of the 5 percent shareholders.

As noted earlier, Direct Public Group, the only 5 percent shareholder, owns 100 percent of the RIC on the testing date. Although it does not affect the outcome under the facts in this example, it is not clear whether the low point of Direct Public Group’s ownership includes the stock owned by Indirect Public Group as a result of the push-down or instead excludes it. In the former case, Direct Public Group’s low point would be 1 percent. In the latter case, the low point would be 0 percent. Whether Direct Public Group’s low point is 0 percent or 1 percent, an ownership change has occurred because the increased ownership of 5 percent shareholders is at least 99 percent.

7. Determine which losses are restricted.

Because the RIC experienced an ownership change July 1, 2020, it must identify its pre-change losses. The RIC had NUBIL greater than the lesser of 15 percent of assets or $10 million immediately before the ownership change, representing potential pre-change losses (depending on whether and when it is recognized) subject to restriction. The RIC had no NOL or net realized capital loss for the 2020 tax year and had no CLCs or NOL carryforwards to the 2020 tax year, so the NUBIL is the only loss that is subject to restriction.

8. Analysis.

The result in Scenario 3 appears anomalous because only a small portion of the RIC’s stock (3.9 percent) was transferred when the RIC was a loss corporation. That is, there does not appear to be any material trafficking in losses that one would expect to be present to trigger the application of section 382. The bulk of the stock transfer (95.1 percent) occurred when the RIC had no NOL carryforwards, CLCs, current-year operating losses or net realized capital losses, or NUBIL.

To prevent an ownership change from occurring under the facts in Scenario 3, the regulations could provide that in determining each 5 percent shareholder’s lowest ownership percentage, one does not look back to a date before the corporation became a loss corporation. If that were the rule, there would be no ownership change on July 1, 2020. The 1987 preamble to the section 382 regulations⁸⁷ seems to agree with this approach, stating, “A testing period does not begin until a corporation is a loss corporation (e.g., a corporation with NOL carryforwards or significant unrealized losses). Shifts in ownership that occur before the date that a corporation becomes a loss corporation thus are disregarded.”

Reg. section 1.382-2T(d), however, which was part of the regulations enacted in 1987, clearly provides for a different result. Under that provision, the testing period generally begins three years before the testing date (assuming there has not been a prior ownership change during that period). However, the regulations also provide that the testing period will not commence before the earlier of:

• the first day of the first tax year from which there is a loss carryforward;

• the first day of the first tax year in which NUBIL exceeding the 15 percent/$10 million threshold arose (that is, not the exact date on which the NUBIL arose); and

• the first day of the tax year in which the testing date occurs.

Consequently, the testing period starts January 1, 2020.

Although contrary to T.D. 8149, this result appears to be consistent with the legislative history to section 382. The Joint Committee on Taxation’s general explanation of TRA 1986 (the 1986 blue book) states that the act contemplates that regulations will permit corporations with NUBIL “to disregard transactions that occur before the year for which such a corporation establishes that a net unrealized built-in loss first arose” (emphasis added).⁸⁸ Thus, the result in Scenario 3, although arguably inequitable, appears to be contemplated.

Under reg. section 1.382-2T(g)(5)(i)(B), if a 5 percent shareholder’s interest drops below 5 percent and stays below 5 percent, the corporation can presume for limited purposes that the 5 percent shareholder continues to own the stock that it owned immediately after dropping below 5 percent, and the corporation can treat the 5 percent shareholder as though it were a separate public group. This presumption is applicable to Scenario 3, but it does not appear to prevent an ownership change from occurring under the facts. On February 29, 2020, the interest held by Indirect Public Group (a 5 percent shareholder) dropped to 4.9 percent, making Indirect Public Group the type of shareholder to which reg. section 1.382-2T(g)(5)(i)(B) applies.

The presumption would prevent an ownership change, however, only if it causes Direct Public Group’s interest in the loss corporation not to change July 1, 2020, in which case there would be no owner shift and therefore no testing date on July 1, 2020. Therefore, there would be no ownership change. The presumption would create that result only if it allows the RIC to (1) presume that Indirect Public Group’s ownership remained at 4.9 percent on July 1, 2020; and (2) use that 4.9 percent number to derive a presumed ownership percentage of 95.1 percent for Direct Public Group on July 1, 2020.

Reg. section 1.382-2T(g)(5)(i)(B) allows the RIC to presume that Indirect Public Group’s ownership is frozen at the level it was at on February 29, 2020. However, the regulation does not address whether the level at which the RIC can presume Indirect Public Group’s ownership is frozen is (1) the number of shares that Indirect Public Group owned as of February 29, 2020; or (2) the percentage that Indirect Public Group owned as of that date (that is, 4.9 percent).⁸⁹ If it is the former, this presumption clearly would not prevent an ownership change from occurring on July 1, 2020.⁹⁰

Also, reg. section 1.382-2T(g)(5)(i)(B) does not state that Indirect Public Group’s presumed ownership level may be used to derive a presumed ownership level for Direct Public Group, such that the ownership levels being used total 100 percent. Further, even if Indirect Public Group’s presumed ownership level may be used to derive a presumed ownership level for Direct Public Group, the application of the push-down rule would bring Direct Public Group’s ownership to 100 percent on July 1, 2020, and may be deemed to cause an owner shift. Thus, the presumption in reg. section 1.382-2T(g)(5)(i)(B) does not seem to prevent an ownership change in Scenario 3.

Under reg. section 1.382-2T(e)(1)(ii), transfers among separate public groups resulting from application of the segregation rules under reg. section 1.382-2T(j)(2) and (j)(3)(iii) are disregarded. If this rule were applicable in Scenario 3, there is a possibility for the transaction on July 1, 2020, to be disregarded, in which case there would be no owner shift and no testing date on July 1, 2020. Therefore, there would be no ownership change. However, Direct Public Group and Indirect Public Group are created under the aggregation rules of reg. section 1.382-2T(j)(1), so reg. section 1.382-2T(e)(1)(ii) is not applicable.

D. Scenario 4

1. Multiple testing dates.

The adverse result that Scenario 3 describes may be eliminated in some cases in which there are multiple testing dates. In Scenario 4 (illustrated in Figure 6), assume the facts are the same as in Scenario 3 except that the RIC has a $100,000 net capital loss for the 2020 tax year (but still had no net realized capital loss as of February 29, 2020).

To determine if the loss limitation rules apply to the RIC, we need to ascertain whether an ownership change occurred February 29, 2020, or July 1, 2020. Because the RIC has a net capital loss for the 2020 tax year, it is a loss corporation for the entire 2020 tax year.⁹¹ As a result, February 29, 2020, cannot be eliminated as a testing date as it was in Scenario 3. Because it is the first potential testing date, we focus initially on February 29, 2020. The potential testing period is January 1, 2020, to February 29, 2020.

2. Step 1: Classification.

The owners of Sponsor are public owners because each of them individually owns less than 5 percent of a first- or higher-tier entity (that is, Sponsor) at all times during the potential testing period. Sponsor is a first-tier entity because it owns 5 percent or more of a loss corporation (that is, the RIC) at some point during the potential testing period. The individuals who own direct interests in the RIC are public shareholders because each of them individually owns less than 5 percent of the RIC at all times during the potential testing period.

3. Step 2: Constructive ownership.

Because Sponsor is a first-tier entity, the RIC stock that Sponsor owns is attributed to Sponsor’s shareholders even though Sponsor does not own 5 percent or more of the RIC on the potential testing date. Because all the owners of Sponsor are public owners, however, they do not have to be specifically identified and do not have to aggregate the RIC shares attributed to them in this context with other RIC shares they may own.

4. Steps 3 and 4: Identify 5 percent shareholders using the aggregation rules.

The owners of Sponsor, all of whom are public owners, are a public group (Indirect Public Group). Because Indirect Public Group owns less than 5 percent of the RIC on the potential testing date, it is not a 5 percent shareholder (even though Indirect Public Group did own 5 percent or more of the RIC at some point during the potential testing period). As a result, Indirect Public Group is pushed down. The direct owners of the RIC (other than Sponsor), all of whom are public shareholders, are a separate public group (Direct Public Group) into which Indirect Public Group is pushed. A direct public group is always treated as a 5 percent shareholder (regardless of whether it owns 5 percent or more of the loss corporation during the potential testing period). In this case, Direct Public Group owns 100 percent of the RIC on the potential testing date and therefore is the only 5 percent shareholder of the RIC during the potential testing period.

5. Determine if February 29, 2020, is a testing date.

Because the RIC has a net capital loss for the 2020 tax year, it is a loss corporation on February 29, 2020. Because the stock ownership of Direct Public Group (a 5 percent shareholder) changes on that date, an owner shift occurs. Consequently, February 29, 2020, is a testing date, and the testing period runs from January 1, 2020, to February 29, 2020.

6. Calculate the change in ownership of the 5 percent shareholders.

As noted earlier, Direct Public Group, the only 5 percent shareholder, owns 100 percent of the RIC on the testing date. Regardless of whether Direct Public Group’s low point is 0 percent or 4.9 percent, an ownership change has occurred because the increased ownership of 5 percent shareholders is at least 95.1 percent.

7. Determine which losses are restricted.

Because the RIC experienced an ownership change on February 29, 2020, it must identify its pre-change losses. The RIC had no NUBIL on February 29, 2020, had no CLCs or NOL carryforwards to the 2020 tax year, and had no NOL for the 2020 tax year. Accordingly, the RIC’s only potential pre-change losses are the portion of the $100,000 net capital loss that is allocable to the period before the ownership change on February 29, 2020. Under reg. section 1.382-6(a), the RIC could choose to determine that portion by ratably allocating an equal portion to each day of the year.

Alternatively, the RIC could elect under reg. section 1.382-6(b) to determine the portion of the $100,000 net capital loss that is allocable to the period before the ownership change as if its books were closed on the date of the ownership change. Under this method, none of the net capital loss would be treated as a pre-change loss, because the RIC did not have a net capital loss as of February 29, 2020. The RIC would thus have no pre-change losses, and the ownership change would have no actual effect. The RIC would presumably elect to use the closing-of-the-books method and achieve this result.

July 1, 2020, is a second testing date (the RIC is a loss corporation on that date, and the stock ownership of Direct Public Group changed on that date), and the testing period runs from March 1, 2020, to July 1, 2020 (that is, it begins on the first day following the date of the previous ownership change). Direct Public Group’s low point during that testing period is 100 percent, and thus there is no ownership change on July 1, 2020.

8. Analysis.

At first blush, the result in Scenario 4 seems odd. The existence of the net realized loss at the end of the 2020 tax year makes the RIC a loss corporation as of February 29, 2020, but the closing-of-the-books election allows the RIC to avoid a restriction on this loss. As odd as that may seem, this result is appropriate from a theoretical perspective. The substantial ownership shift that occurred February 29, 2020, occurred when the RIC had no losses, and thus there does not appear to be material loss trafficking that one would expect to trigger the loss limitation rules. Unfortunately, the loss limitation rules provide for this result only in the narrow circumstances presented in this scenario and thus do not eliminate the broader inequity that Scenario 3 illustrates.

E. Fixing the Problem

As the preceding scenarios demonstrate, a RIC may be subject to the loss limitation rules solely because of its sponsor’s contribution of seed capital. This result is not only inequitable but also unwarranted in light of the purpose of the loss limitation rules, which Congress intended to prevent the possibility “that new shareholders will contribute income-producing assets (or divert income opportunities) to the loss corporation, and the corporation will obtain greater utilization of carryforwards than it could have had there been no change in ownership.”⁹² A sponsor’s contribution of seed capital in no way implicates these concerns; instead, the sponsor makes the contribution because it is required to do so for legal or fiduciary reasons.

The methodological problem with the loss limitation rules’ application to RIC seed capital is that they do not treat the sponsor’s contribution as part of an integrated start-up transaction. If testing dates for RICs could occur, and testing periods for RICs could begin, no earlier than a specified number of days after a sponsor’s initial contribution of seed capital, the problems discussed in this section would be largely eliminated.

The IRS has taken this type of approach in analogous circumstances. In LTR 9142018, a new corporation issued stock to its directors December 29, 1987. On February 29, 1988, the corporation issued additional stock (representing approximately 99 percent of its outstanding stock) to investors and began operations. The corporation was a loss corporation for its tax year ending December 31, 1988. Based on the corporation’s representation that the two share issuances were integrated parts of its formation plan, the IRS ruled that no ownership change had occurred on February 29, 1988. Although not discussed in the ruling, in evaluating future shifts in its stock ownership, the corporation should have treated February 29, 1988, as the earliest date on which a testing period could begin.

The IRS took a similar approach in ILM 201432015. There, a company was a loss corporation at the end of its first tax year. The company had issued stock on several dates during that year. Again, the IRS concluded that it was appropriate to integrate some of the stock issuances because they involved the initial capitalization of the company.

Because an open-end RIC offers its shares continuously, there is no clear point at which the RIC’s initial offering period ends. In Rev. Rul. 73-463, 1973-2 C.B. 34,⁹³ however, the IRS concluded that in deducting stock issuance expenses, an open-end RIC’s initial stock offering period ends 90 days after the date its registration statement is first declared effective under the Securities Act of 1933.⁹⁴ According to the ruling:

The ninety day period was determined by reference to section 14 of the [1940 act], which permits an investment company a maximum period of ninety days after a registration statement of its securities, offered for public sale under the Securities Act of 1933, becomes effective to raise its statutory minimum capital.

This 90-day period seems equally appropriate in applying (or, more accurately, not applying) the loss limitation rules to RICs. Under this approach, none of the four scenarios discussed earlier would result in an ownership change. For instance, in Scenario 4, February 29, 2020, would not be a testing date. Also, because Direct Public Group would be treated as owning 100 percent of the RIC on both February 29, 2020, and July 1, 2020, no ownership change would occur on the latter date.

VI. Nondeductibility of NOLs

Although section 172 generally allows both individuals and corporations to carry forward NOLs for use in future years, RICs may not do so.⁹⁵ The genesis of this rule is unclear, but historically its impact was limited because RICs had few ordinary deductions. Typically, only equity funds investing in stocks with low dividend yields (such as small-cap stocks) had insufficient ordinary income to fully offset their expenses (management fees, custody fees, and similar expenses), and resulting NOLs generally were small. A RIC’s losses from investments were almost always capital losses.

Recent changes in the code and regulations have resulted in RIC investments generating ordinary losses. For example, a passive investment company that marks to market under section 1296, foreign currency losses under section 988, and deductions on notional principal contracts may create ordinary losses. A RIC therefore is far more likely to have an NOL than in the past. There is no compelling policy reason to deny RICs the ability to carry NOLs forward,⁹⁶ and the inability to do so puts RICs at a competitive disadvantage relative to other types of investment vehicles. Consequently, section 852(b)(2)(B) should be repealed and a corresponding fix should be made to section 852(c) (as detailed in Section VII).

VII. CLCs, NOLs, and RIC’s Accumulated E&P

An entity can qualify as a RIC only if, among other requirements, the deduction for dividends paid equals or exceeds 90 percent of its “investment company taxable income” (and 90 percent of its tax-exempt income less allocable expenses).⁹⁷ Any investment company taxable income exceeding 90 percent and any net capital gain that a RIC does not distribute and for which a RIC does not receive a dividends paid deduction is subject to entity-level tax.

The dividends paid deduction therefore is critical to a RIC’s passthrough status. Because shareholders include the corresponding dividends in taxable income (unless they are tax exempt or tax deferred), a reduction in the number of RIC shares outstanding because of redemptions can cause both a timing and character detriment to shareholders. To take an overly simplified example, assume that on January 1, 2019, A and B each invest $10 in a newly created calendar-year RIC and receive one RIC share. The RIC invests the $20 in various assets and recognizes $4 of gain and receives $2 of dividends through November 30, 2019. Also assume that as of that date the RIC has no unrealized gain or loss in its assets and has made no distributions, so its net asset value per share is $13. If Shareholder B redeems its share November 30, 2019, it will recognize a $3 short-term capital gain, which represents its share of the dividends received and gains recognized by the RIC.

Assuming the RIC recognizes no additional income, gain, or loss in 2019, the RIC must distribute a $6 dividend in December to Shareholder A, its remaining shareholder, to reduce its taxable income to zero for 2019.⁹⁸ Shareholder A therefore pays tax not only on its $3 share of income/gain but also on Shareholder B’s share of income/gain. Assuming Shareholder A does not reinvest the dividend, it will continue to own one RIC share with a basis of $10 and a value of $7 (that is, the $13 net asset value of the share immediately before the distribution (assuming the value of its assets do not change after November 30) minus the $6 distribution). Shareholder A therefore has an unrealized $3 loss to offset its inclusion in income of Shareholder B’s share of the RIC’s income/gain, but the loss will be capital while the inclusion is ordinary, and recognition of the loss will be deferred until Shareholder A redeems its share of the RIC.

To prevent this result, section 562(b)(1) allows RICs to claim a dividends paid deduction for accumulated E&P “distributed” to redeeming shareholders in their redemption proceeds. Section 562(b)(1)(A) provides generally that “in the case of amounts distributed in liquidation, the part of such distribution which is properly chargeable to E&P accumulated after February 28, 1913, shall be treated as a dividend for purposes of computing the dividends paid deduction” (emphasis added). The flush language in section 562(b)(1) states that for purposes of section 562(b)(1)(A) a “liquidation” generally includes a redemption to which section 302 applies.⁹⁹

This imputed deduction, known as equalization, is designed to reduce or eliminate the amount of income/gain a RIC would have distributed to redeeming shareholders. To oversimplify, the RIC in the example might claim a dividends paid deduction for Shareholder B’s $3 share of income/gain, so that the RIC would have to distribute only $3 to Shareholder A to reduce the RIC’s taxable income to zero. Equalization therefore functions in a manner analogous to a partnership’s election under section 754.¹⁰⁰

Although section 316(a)(1) and (2) provides that a corporate distribution is treated as a dividend if the corporation has either current or accumulated E&P, section 562(b)(1)(A) provides that for equalization purposes, only the portion of redemption proceeds attributable to accumulated E&P gives rise to a dividends paid deduction. If a RIC has a deficit in accumulated E&P but positive current-year E&P, it will be unable to use equalization when it pays its current-year distribution. For example, if the RIC in the preceding example had been launched in 2018 and had a $40 NOL, it would have a deficit in accumulated E&P in 2019 (but $6 of current E&P), so it would be unable to use equalization to reduce its $6 distribution requirement (because none of the proceeds used to redeem Shareholder B would be attributable to accumulated E&P).

Although this result may not appear unduly harsh, it is exacerbated by special E&P rules that apply only to RICs. Because of those rules, in some fact patterns a RIC arguably may start every year with a deficit in accumulated E&P that it can never eliminate. Instead, each year the RIC will need to “earn out” of the deficit before it can use equalization. The deficit will then return at the beginning of the following year.

This issue is partially attributable to code provisions that predate the Regulated Investment Company Modernization Act of 2010 (RIC Modernization Act). Before its amendment, section 851(c)(1) provided that a RIC reduced its accumulated E&P — but not its current E&P — for any amount that the RIC could not deduct, most importantly NOLs and CLCs.¹⁰¹ Under this provision, a RIC with a current-year capital loss and current-year ordinary income could meet its 90 percent distribution requirement from current E&P. For example, a RIC in a year before 2010 with a current-year net capital loss of $30 and current-year investment company taxable income of $40 could pay a $40 dividend and reduce its investment company taxable income to zero because it had current E&P of $40 (because the nondeductible capital loss would not affect its current E&P).

Because distributions typically reduce E&P,¹⁰² one might expect that the $40 dividend would create a $30 deficit in accumulated E&P that the RIC would carry into the subsequent year (that is, $40 of investment company taxable income minus $30 capital loss minus $40 dividend). Reg. section 1.852-5(b) provides that a distribution cannot alone create a deficit in accumulated E&P. The regulation provides the following example:

If a corporation would have had E&P of $500,000 for the tax year except for the fact that it had a net capital loss of $100,000, which amount was not deductible in determining its taxable income, its E&P for that year if it is a regulated investment company would be $500,000. If the regulated investment company had no accumulated E&P at the beginning of the tax year, in determining its accumulated E&P as of the beginning of the following tax year, the E&P for the tax year to be considered in such computation would amount to $400,000 assuming that there had been no distribution from such E&P. If distributions had been made in the tax year in the amount of the earnings and profits then available for distribution, $500,000, the corporation would have as of the beginning of the following tax year neither accumulated earnings and profits nor a deficit in accumulated earnings and profits, and would begin such year with its paid-in capital reduced by $100,000, an amount equal to the excess of the $500,000 distributed over the $400,000 accumulated earnings and profits which would otherwise have been carried into the following tax year.

This regulation, however, does not mean that CLCs arising before the RIC Modernization Act never created deficits in a RIC’s accumulated E&P. Instead, those deficits arose to the extent they were not attributable to a RIC’s distributions. This principle is illustrated by Rev. Rul. 76-299, 1976-2 C.B. 211, in which a RIC had a net capital loss of $800x in its 1969 tax year and no other income, expense, or loss. The IRS concluded that the RIC started 1970 with an $800x deficit in its accumulated E&P.¹⁰³

For its 1970 tax year, the RIC had $10,000x of ordinary income and a $1,000x net long-term capital gain, and it distributed $11,000x — $1,000 of which it designated as a capital gain dividend. The IRS ruled that the RIC would not decrease its accumulated E&P when it used the CLC to offset a portion of the capital gain because the loss giving rise to the carryover was reflected in the accumulated E&P at the beginning of the tax year of the carryover. As a result, the entire $11,000x distribution was paid from E&P from 1970. Because the RIC had to net the $800x CLC against the $1,000x capital gain, however, only $200x of the distribution qualified as a capital gain dividend; the remaining $10,800x was an ordinary dividend. The ruling also concluded that the RIC had an $800x deficit in its accumulated E&P January 1, 1971 (because it started 1970 with an $800x deficit and then earned and distributed $11,000x during that year).

At first blush, this revenue ruling may appear to suggest that a pre-RIC Modernization Act CLC could create a permanent deficit in accumulated E&P in all cases. But that is not the case. The only reason that the RIC in the ruling had a deficit in its accumulated E&P January 1, 1971, was because it chose to distribute more than was required; the RIC could have avoided entity-level tax by paying an ordinary dividend of $10,000x and a capital gain dividend of $200x. In that case, it would have had a zero balance for accumulated E&P January 1, 1971. That is, by using the CLC (and not overdistributing), a RIC could earn its way out of the deficit in accumulated E&P.

Problems could arise, however, if a RIC did not use a CLC. Before being amended in 2010, section 1212(a)(1)(C) provided that a RIC’s CLC could be carried forward only eight years, after which it expired. To illustrate the problem this creates, assume that the RIC in the revenue ruling was unable to use any of its $800x CLC in its 1970-1977 tax years. Assuming the RIC distributed all its ordinary income in those years, any increase in its accumulated E&P attributable to that income would have been offset by a corresponding decrease attributable to the distribution of that income. Thus, at the beginning of its 1978 tax year, the RIC still would have had a deficit of $800x in accumulated E&P. Assume that the RIC had a net capital gain of $1,000x in its 1978 tax year. Because its CLC had expired, the RIC would have to distribute $1,000x to its shareholders to avoid entity-level tax, so the $800x deficit would remain at the start of the RIC’s 1979 tax year and in every subsequent year.¹⁰⁴ The RIC therefore could not start to use equalization in a given year until its income and gains exceeded $800x.

The RIC Modernization Act eliminated this problem for newly incurred CLCs. First, it amended section 1212 so that a RIC’s CLCs no longer expire. Second, and more important, the act amended section 852(c) to provide that a net capital loss reduces neither current nor accumulated E&P. It is unclear, though, whether pre-act losses that expired unused continue to create a permanent deficit in accumulated E&P or instead may be removed from the calculation of accumulated E&P.

This ambiguity stems from the effective date language for the changes. The amendment to section 1212 (eliminating the eight-year expiration rule) applies to net capital losses for tax years beginning after December 22, 2010. In contrast, the amendment to section 852(c) (providing that net capital losses do not reduce current or accumulated E&P) applies to tax years beginning after — not to net capital losses incurred in tax years beginning after — December 22, 2010. Arguably, this language means that the new E&P rule applies in tax years beginning after December 22, 2010, to all net capital losses, including those incurred in tax years beginning on or before December 22, 2010. If so, the RIC in the example could eliminate the $800x deficit in its accumulated E&P for its first tax year beginning after December 22, 2010.

Even if this interpretation of the effective date language is correct, it does not eliminate the permanent deficit problem that RIC NOLs create. Under section 852(b)(2)(B) RICs cannot deduct NOLs. Yet RICs continue to have to reduce their accumulated E&P (but not their current E&P) for NOLs.¹⁰⁵ This situation is even worse than the one that existed for CLCs before the RIC Modernization Act, when RICs at least had eight years to use the CLCs; for NOLs, there is no opportunity whatsoever to use them. Unfortunately, it appears that only an amendment to the code can fix this problem. That amendment should provide that NOLs, like CLCs, reduce neither current nor accumulated E&P.¹⁰⁶

VIII. Conclusion

RICs face several potential tax issues when they experience losses. The solutions offered in this report would go a long way in alleviating or eliminating them.

IX. Exhibits

A. Exhibit 1: Proposed Amendment

Reg. section 1.382-3(k) Special rules for certain regulated investment companies.

[Renumber reg. section 1.382-3(k)(2) as reg. section 1.382-3(k)(3).]

(2) Ownership by certain participant-directed plans. — (i) General rule. — Stock ownership of a regulated investment company by a qualified participant-directed plan shall be disregarded and such plan shall not be treated as an entity for purposes of sections 1.382-2T(h)(2) and 1.382-3(a). The stock owned by such qualified participant-directed plan shall be treated as having been acquired directly by the participants of such plan on the date and in the manner acquired by the plan.

(ii) Qualified participant-directed plan defined. — Qualified participant-directed plan means a qualified trust described in section 401(a) or an eligible deferred compensation plan (within the meaning of section 457) maintained by an employer described in section 457(e)(1)(A), in each case to the extent that a participant in such plan directs the investment of assets allocated to an individual account attributable to the participant under the plan. The fact that the available investments may be limited or that available investments may change from time to time shall not prevent a plan from being a qualified participant-directed plan. However, a plan is not a qualified participant-directed plan to the extent a participant selects only the investment objective (e.g. growth, income, balanced), but not the actual investments.

. . .

(4) Effective date. — (i) General rule. — Paragraph (k)(2) applies to testing periods with testing dates after the date of publication of a Treasury decision adopting these rules as final regulations in the Federal Register. However, taxpayers may rely on testing periods with testing dates prior to the date of publication.

B. Exhibit 2: Accumulated E&P Deficit

If Current-Year Investment Company Taxable Income (ICTI) Is . . .	And Current-Year Capital Gain/Lossa Is . . .	And	Is an Accumulated E&P Deficit Created?b
Negative (NOL)	Zero	N/A	Yes; = NOL
Zero	Loss (CLC)a	N/A	Yes; = CLC
Negative (NOL)	Loss (CLC)a	N/A	Yes; = NOL + CLC
Zero	Zero	N/A	N/A
Positive	Loss (CLC)a	ICTI exceeds absolute value of loss	No; decrease paid-in capital (PIC) by CLC
Positive	Loss (CLC)a	ICTI does not exceed absolute value of loss	Yes; = ICTI minus absolute value of CLC. Also, decrease PIC by ICTI
Negative (NOL)	Gain	Absolute value of NOL exceeds gain	Yes; = gain minus absolute value of NOL. Also, decrease PIC by gain
Negative (NOL)	Gain	Absolute value of NOL does not exceed gain	No; decrease PIC for NOL
Positive	Zero	N/A	N/A
Zero	Gain	N/A	N/A
Positive	Gain	N/A	N/A
Zero	Zero	N/A	N/A
aCapital losses are attributable to tax years before the effective date of the RIC Modernization Act. bFund distributes exact amount required to avoid entity-level tax.

C. Examples

Current-Year ICTI	Current-Year Capital Gain/Lossa	Distributionb	Accumulated E&P Deficit?
(500)	0	0	(500)
0	(500)a	0	(500)
(500)	(500)a	0	(1,000)
0	0	0	N/A
500	(100)a	500	No; decrease PIC by 100
500	(900)a	500	(400); decrease PIC by 500
(900)	500	500	(400); decrease PIC by 500
(100)	500	500	No; decrease PIC by 100
500	0	500	N/A
0	500	500	N/A
500	500	1,000	N/A
0	0	0	N/A
aCapital losses are attributable to tax years before the effective date of the RIC Modernization Act. bFund distributes exact amount required to avoid entity-level tax and does not use any pre-RIC Modernization Act CLCs to reduce the amount it must distribute.