OPINION
Holmes, Judge:
When a company winds up an employee-benefit plan and distributes its assets, section 402(b)
says an employee receiving his share of those assets has to pay tax on “the amount actually distributed.” Michael Schwab and his wife Kathryn Kleinman both received life-insurance policies as their share of an employee-benefit plan that was ending. They argue that surrender charges on both the policies made them worth nothing at the time of their receipt. The Commissioner argues that we must consider only what the insurance company calculated to be the policies’ “stated values” in figuring out what the “amount actually received” by Schwab and Kleinman was. The dispute is a novel one.
Background
Schwab and Kleinman are the sole shareholders of Angels & Cowboys, Inc.
They are also employees of the corporation; Schwab works as a graphic designer and Kleinman as a photographer. Schwab has created award-winning logos and posters for clients that include Major League Baseball, the Muhammad Ali Center, Nike, Pebble Beach, Polo Ralph Lauren, Robert Redford, and the San Francisco Opera. One collector described Schwab’s work: “Like a clearing in a dark, unfathomable forest, or an island in a turbulent sea, the graphic art of Michael Schwab is a welcome sight, a safe harbor, amidst the enigmatic and increasingly illegible pool of contemporary art and design.” Merrill C. Berman, Michael Schwab Studio — About Michael, Michael Schwab Studio, http:// www.michaelschwab. com/studio/studio — about, html (last visited Jan. 1, 2011). Kleinman is also highly talented and has done photography for such high-profile clients as Apple Computer, the GAP, Microsoft, and Wolfgang Puck Foods. Studio-Clients, Kathryn Kleinman Studio, http: //www.kathrynkleinman.com/html / studio — clients.html (last visited Jan. 1, 2011).
Accountants follow success, and George Stameroff, a Marin County CPA, was the couple’s accountant until 2001. He prepared Schwab and Kleinman’s tax returns throughout the ‘90s and also gave them financial-planning advice. In 2000, he recommended that the couple buy life-insurance policies through a multiple-employer welfare-benefit trust administered by Benistar. The trust was an employee-benefit plan known as the “Advantage 419 Trust,” because it was designed to conform with section 419A(f)(6).
Benistar was aimed at small-business owners and was the nation’s largest administrator of such plans. Stameroff gave Schwab and Kleinman the Benistar marketing brochures that claimed the plan allowed “qualified professionals, entrepreneurs, and closely-held business owners to obtain life insurance for themselves and for key employees on a tax-deductible basis.” These promotional materials emphasized that the plan assets (invested, in Schwab and Kleinman’s case, in an S&P 500 stock-index fund) would grow tax-free and that the death benefits would be income-tax free. According to Stameroff and the Benistar marketing materials, if the plan were terminated, the policies would be distributed to the participants and their value net of surrender charges would be taxable.
Schwab and Kleinman liked what Stameroff had to say about the Advantage 419 Trust and decided to adopt it. But their relationship with Stameroff would soon come to an end. They found out that he was an authorized agent of Benistar and decided to look for another accountant because Schwab felt they “didn’t have a clear rapport with him.” In 2001, Sander Stadtler replaced Stameroff as the couple’s CPA. Stadtler consulted with the couple regarding tax-return preparation and other financial matters. Part of his consultation included an extensive review of the Advantage 419 Trust. He asked Stameroff several questions about the plan to “better understand the various costs and charges in the plan, required contributions, and projected results.” On Stadtler’s recommendation, in 2002 Schwab and Kleinman reduced their death benefits from $5.5 million to $2.4 million. Stadtler also opined that if the couple terminated the plan they would be taxed on the net cash-surrender value of the life-insurance policies.
All seemed well. But roiling in the background was the IRS’s view, which it had held since at least 1995, that most trust arrangements promoted as multiple-employer welfare-benefit funds “do not satisfy the requirements of the section 419A(f)(6) exemption.” Notice 95-34, 1995-1 C.B. 309, 310. In Booth v. Commissioner, 108 T.C. 524 (1997), the Commissioner successfully challenged a plan’s reliance on section 419A(f)(6) by showing that it was really a series of single-employer plans rather than a true multiple-employer plan. In spite of Notice 95-34 and the Commissioner’s litigation success, most taxpayers continued to take the position that their 419 plans were allowable under the Code. The Commissioner raised the stakes in 2000 by designating 419 plans described in Notice 95-34 as “listed transactions.” Notice 2000-15, 2000-1 C.B. 826.
Taxpayers are required to disclose listed transactions on their returns, and promoters of such transactions have to register them with the IRS. But many taxpayers took the position that their particular plans weren’t described in Notice 95-34 and so were not “listed transactions.” The IRS then issued proposed regulations on section 419A(f)(6) plans in 2002, sec. 1.419A(f)(6)-l, Proposed Income Tax Regs., 67 Fed. Reg. 45938 (July 11, 2002), that more or less tracked its litigation position.
The proposed regulations caught the attention of BISYS, the plan’s new administrator, who hired outside counsel in 2002 to assess the situation. BISYS eventually concluded that the Advantage 419 Trust would not be able to comply with the proposed regulations. By 2003 it became clear that the Treasury Department would adopt the regulations substantially as proposed; BISYS terminated the plan for all employers, including Angels & Cowboys. The plan then distributed the life-insurance policies to Schwab and EJeinman in October 2003. At the time of distribution, Schwab’s policy had a “policy value” of $48,667 and Kleinman’s had one of $32,576. “Policy value” is an important term in this case, and it’s defined in the plan documents as “premiums less policy loads, plus net investment return, less policy charges, partial surrenders, and any indebtedness.” Schwab and KLeinman had two options upon distribution — continue paying premiums to keep their life-insurance coverage, or surrender the policies for their value less any surrender charges.
But there was a catch. The policies were of a type called variable universal life, a relatively new type of contract for this old industry. A key characteristic of universal life-insurance policies is that they disconnect to some degree a life-insurance feature (i.e., payment of money upon death) from an investment feature (i.e., the use of premiums to acquire assets that fund the insurance payment). The insurer selling a universal-life policy typically segregates payments from its customers in separate investment accounts from which it makes deductions to pay for the insurance component of the policy. At death, the customer’s beneficiary gets what’s left in the separate account. Under a variable universal life-insurance contract, the customer typically can choose from a menu of different investments (often set up to closely resemble mutual funds) with varying returns and thus varying payouts upon death, though there is (as was true under the contracts here) a minimum death-benefit guaranty.
The expected premiums for Schwab and Kleinman on their variable universal-life policies were quite steep. For Schwab, the premium was originally more than $136,000 a year; for Kleinman, it was $120,000. In some of the illustrations that the insurance company used, Angels & Cowboys would be paying such premiums for decades; in some, the firm would pay premiums for only ten years. But we find that the firm paid the premiums only for the policies’ first year.
The policies nevertheless remained in effect pursuant to a “no-lapse provision.” This provision states:
During the first 3 policy years if the sum of all premiums paid on this policy * * * is greater than the no lapse premium multiplied by the number of months the policy has been in force, the policy is guaranteed not to lapse even if the net cash surrender value is zero or less. If less than the no-lapse premium is paid during the first 3 policy years, the policy will not necessarily lapse provided the net cash surrender value is greater than zero.
Schwab’s “no-lapse” premium was set at $3,548.77. This meant that the policy wouldn’t lapse for the first three years — even if no more premiums were paid — so long as the initial premium payment of $136,000 remained greater than $3,548.77 x N (where N = the number of months the policy had been in effect) or the net cash-surrender value remained greater than zero.
Kleinman’s “no-lapse” premium was $3,776.69. With the initial premium payment of only $120,000, the net cash-surrender value of her policy would have to exceed zero after only 31 months to avoid a lapse or incur an obligation to pay more premiums.
When in 2002 Schwab and Kleinman elected to reduce the coverage, those scheduled premiums shrank as well, but in both cases to more than $22,000 a year. The “no-lapse” premiums shrank as well: Schwab’s to $2,498.31; and Kleinman’s to $2,510.34.
One thing did not change: Both Schwab and Kleinman had directed that their premium payments be segregated into accounts whose value fluctuated with the S&P 500 stock index. The death benefit and cash-surrender value depend on those fluctuations in investment returns.
And the policies’ surrender charges were greater than their stated policy values in October 2003, meaning Schwab and Kleinman wouldn’t get any cash if they immediately surrendered their policies upon receipt. Here’s how the numbers looked at distribution:
Schwab Kleinman
$48,667 $32,576 Stated policy value
49,225 46,599 Surrender charges
(558) (14,023) Net cash-surrender value
The surrender charges lasted eleven years and would be reduced by 20 percent a year only in years 8-12 (starting in 2008). But if the S&P 500 were to go up or if further premiums were paid, the policy values would increase as well. Schwab’s policy seemed more worthwhile — it could potentially be in the black in a matter of weeks. And by December 2, 2003, when Schwab asked to surrender his policy, he was ahead by approximately $1,100. But he then changed his mind and contacted the insurance company in mid-December to reverse his termination request. The value of the policy net of surrender charges increased in the interim, reaching $1,630 by December 16. Schwab continued to hold the policy and pay his premiums at least until the trial.
Kleinman’s policy was so deeply under water that she let it lapse shortly after distribution by not paying the required $108,031 premium.
She didn’t get any money from the insurance company because her policy’s net cash-surrender value was negative.
BISYS did not issue any 1099 forms after it distributed the policies, so when Stadtler prepared the couple’s 2003 return, he did not report any income from their distribution. We find that in taking this position the couple was also relying on Stammeroff’s, the plan administrator’s, and Stadtler’s 2001 conclusion that they would be taxed only to the extent of the net cash-surrender value. Schwab provided all the materials that Stadtler asked for and answered all of his questions in the course of preparing the return. (Though we do find that he did not specifically ask Stadtler about the tax consequences of the distributions.)
The Commissioner issued a notice of deficiency asserting increases in tax and penalties for Schwab and Kleinman’s failure to include the stated policy values as income. They timely petitioned the Tax Court as residents of California. We tried the case in San Francisco.
Discussion
This case is about the rules for taxability of property distributed after the termination of employee-benefit plans. Such plans come in “qualified” and “nonqualified” varieties. Qualified plans must meet the requirements of section 401, and all the complicated regulations governing their funding, nondiscriminatory terms, employee coverage, distribution, and other requirements. Meeting such requirements allows for favorable tax treatment of qualified plans, but not all plans can comply; hence the existence of nonqualified plans. Nonqualified plans are generally subject to fewer statutory and regulatory requirements, but they also receive less favorable tax treatment. The rules for taxing distributions from qualified and nonqualified plans differ as well. Section 402(a) governs distributions from qualified plans, and section 402(b) governs distributions from nonqualified plans.
I. Determining the Amount Actually Distributed
The Advantage 419 Trust was a nonqualified plan, so we apply section 402(b)(2). That section reads:
The amount actually distributed or made available to any distributee by any trust described in paragraph (1) shall be taxable to the distributee, in the taxable year in which so distributed or made available, under section 72 (relating to annuities) * * * [Emphasis added.]
But what amount was “actually distributed” when BISYS transferred the life-insurance policies to Schwab and Kleinman? The Commissioner claims $81,243 — their total stated policy value. Schwab and Kleinman see things differently and claim that nothing of value was “actually distributed.” They rely first and most insistently on the plain language of the Code. The words “amount actually distributed” appeared in the Code as far back as 1921 in section 219(f), Revenue Act of 1921, ch. 136, sec. 219(f), 42 Stat. 227, 247, which became section 165 in 1928, Revenue Act of 1928, ch. 852, sec. 165, 45 Stat. 791, 839.
Schwab and Kleinman point out that the committee reports for both the 1928 and 1932 Acts don’t address taxing the stated value of an insurance policy.
See S. Rep. No. 72-665, sec. 165 (1932), reprinted in 1939 — 1 C.B. (Part 2) 496, 520; H.R. Rep. No. 70— 2, sec. 165 (1927), reprinted in 1939-1 C.B. (Part 2) 384, 398-99.
Finding no esoteric meaning in the legislative history, Schwab and Kleinman point us to the dictionary, which defines “actually” as “in fact; in reality.” American Heritage Dictionary 18 (4th ed. 2000). Because they are cash-basis taxpayers, Schwab and Kleinman argue they would have to actually or constructively receive income before they would incur any tax liability. See sec. 1.451-l(a), Income Tax Regs.; see also United States v. George, 420 F.3d 991, 996 (9th Cir. 2005). There’s no actual receipt here, and no constructive receipt because no income was credited and made available to them without restriction: In the words of the regulation, Schwab and Kleinman could not “draw upon it at any time,” and their control of the policy’s value (if “control” is the right word) was “subject to substantial limitations or restrictions.” Sec. 1.451-2(a), Income Tax Regs. The stated policy values, they argue, had no cash equivalence or economic value upon distribution. Schwab and Kleinman admit that the insurance policies that they got showed “policy values,” but reasonably point out that that “value” is not what they could actually have gotten in hand at the time of distribution. They conclude, therefore, that they received something that had “no economic, monetary or cash surrender value.”
The Commissioner’s argument is not nearly as straightforward. He begins with the incontestably true observation that there is no regulation or caselaw directly on point. He then argues that:
• The insurance policies should be treated as if they were annuities;
• treating them as if they were annuities means including in Schwab and Kleinman’s income their “entire value;” and
• “entire value” does not include any surrender charges.
The first problem is that section 402(b)(2) says that the “amount actually distributed * * * shall be taxable to the distributee * * * under section 72 (relating to annuities).” This does not mean that any “amount actually distributed” is an annuity, but only that the taxability of whatever amount was “actually distributed” has to be computed by using section 72’s rules on recovery of the taxpayer’s investment in the contract. The Commissioner nevertheless points to section 1.402(b)-l(c)(l), Income Tax Regs. Like the Code, that regulation includes the phrase “actually distributed,” but it continues with an example of the distribution of an annuity contract:
If, for example, the distribution from such a trust consists of an annuity contract, the amount of the distribution shall be considered to be the entire value of the contract at the time of distribution. * * * [7ci.]
This tells us something: When an annuity contract is distributed, it’s the “entire value” of the contract that is taxable under the rules governing the taxation of annuities.
And the Commissioner’s argument rests in a subtle way on extending the regulation’s command that an annuity contract’s “entire value” is the “amount actually distributed” to the valuation of life-insurance policies like Schwab’s and Kleinman’s.
The Commissioner’s shifting of our focus to the meaning of the phrase “entire value” — remember, a term taken from an example in the regulation about the valuation of an annuity contract — and away from the phrase “amount actually distributed” aims to take advantage of a regulation that defines “entire value” in a somewhat unusual way. That regulation, section 1.402(b)-l(b)(2)(i), Income Tax Regs., provides that the “entire value” does not take into account what the regulations call a “lapse restriction.” The Commissioner then continues his argument by asserting that surrender charges on a life-insurance policy are a type of lapse restriction, an argument he recently won in Cadwell v. Commissioner, 136 T.C. 38 (2011). In Cadwell, we disregarded surrender charges after looking to Revenue Procedure 2005-25’s safe-harbor definition of fair market value for a formula to apply in default of the taxpayer’s failure to offer any reason we shouldn’t. Id. at 58-59.
That revenue procedure became effective only after the distributions here. And looking to the regulation the Commissioner points us toward, section 1.402(b) — l(b)(2)(i), Income Tax Regs., we see that it uses the word “value” and provides that
The net fair market value of all the assets in the trust is the total amount of the fair market values (determined without regard to any lapse restriction, as defined in § 1.83 — 3(h)) of all the assets in the trust, less the amount of all the liabilities (including taxes) to which such assets are subject or which the trust has assumed * * * as of the date on which some or all of the employee’s interest in the trust becomes substantially vested. [Emphasis added.]
The thing to notice about the Commissioner’s argument on this point is that it is based on language in the part of the regulation governing the valuation of an employee’s rights to assets still held in trust at the time those rights become vested. But Schwab’s and Kleinman’s policies were distributed — they were not still held in trust. The relevant regulation for this situation is not section 1.402(b)-l(b), but section 1.402(b)-l(c), Income Tax Regs., which doesn’t even mention “lapse restrictions.”
That leaves us back where we started — trying to find the meaning of the phrase “amount actually distributed.” Schwab and Kleinman point us to regulatory language for qualified plans, telling us that the taxable value of an insurance contract “actually distributed” to a plan’s participant is its “policy cash value.” Sec. 1.402(a) — l(a)(l)(iii), Income Tax Regs. (Remember that the Schwab-Kleinman distribution was from a nonqualified plan — like the Commissioner they’re also pointing to a facially inapplicable regulation and arguing by analogy.)
That’s not quite right — it’s not just the “policy cash value” but “all other rights under such contract” that count toward fair market value. They do argue by analogy, however, that we should take surrender charges into account here because on this point there is no reason the distribution of a life-insurance contract from a qualified plan should be treated differently from a nonqualified distribution. But there’s a problem with the cited regulation — it’s effective as of August 29, 2005. The prior version of the regulation — which would have applied to distributions from qualified plans at the time Schwab and Kleinman received their distribution — makes no reference to “policy cash value.” Instead, it provided that the “entire cash value” of a life-insurance contract distributed from a qualified plan is taxable. Sec. 1.402(a)-l(a)(2), Income Tax Regs. And in a development that Schwab and Kleinman couldn’t foresee, we recently construed that language to mean something different from “fair market value.” See Matthies v. Commissioner, 134 T.C. 141, 150-51 (2010) (construing pre-2005 regulations under section 402(a) as requiring that the “entire cash value” of life insurance policies be determined without regard to surrender charges).
This difference in the regulations may seem odd — both section 402(a) and 402(b) contain the phrase “amount actually distributed,” yet the regulations interpreting each subsection differed before 2005 and continue to differ today. We must apply them as written. In the absence of regulatory guidance, we hold that the “amount actually distributed” means the fair market value of what was actually distributed. One textual clue in the regulation itself that supports this is in the illustration of an annuity contract that is distributed. Section 1.402(a)-l(a)(2) used to say that, in such a case, it’s the “entire cash value of such contract at the time of distribution” that is included in income; section 1.402(b)-l(c) says that, if a nonqualified plan distributes an annuity contract, the value of the distribution is the contract’s “entire value.” In Matthies, we suggested that this latter phrase — “entire value” — “might plausibly be construed as synonymous with ‘fair market value’ ” and represented “a generalized valuation standard.” Matthies, 134 T.C. at 150-51.
But the fair market value of insurance contracts can be a slippery concept, and is not necessarily synonymous with net cash-surrender value. Consider, for instance, the case of a taxpayer who buys a single-premium life-insurance contract and immediately gives it to her children. The purchase price obviously represents one good measure of its value but, as is true of many life-insurance contracts, the surrender charges that would apply for a number of years would make the net cash-surrender value less than the purchase price. The Supreme Court analyzed the problem:
Surrender of a policy represents only one of the rights of the insured or beneficiary. Plainly that right is one of the substantial legal incidents of ownership. But the owner of a fully paid life insurance policy has more than the mere right to surrender it; he has the right to retain it for its investment virtues and to receive the face amount of the policy upon the insured’s death. That these latter rights are deemed by purchasers of insurance to have substantial value is clear from the difference between the cost of a single-premium policy and its immediate or early cash-surrender value * * * [Guggenheim v. Rasquin, 312 U.S. 254, 257 (1941); citations omitted.]
In Guggenheim, the Court held that the time between purchase and gift was short enough that “cost is cogent evidence of value” and so it was the purchase price of the insurance policy that was the best measure of its value. Id. at 257-58.
But in another case that same term, the Court also held that a paid-up policy that had been in effect for a much longer time — five years — had a value best measured by the cost of a replacement policy on the then-current age of the insured. United States v. Ryerson, 312 U.S. 260, 261 (1941). In the case of single-premium policies, we summarized the state of the law sixty years ago:
The cash surrender value is the market value only of a surrendered policy and to maintain that it represents the true value of the policy is to confuse its forced liquidation value at an arbitrary figure with the amount realizable in an assumed market where such policies are frequently bought and sold. * * *
The rule is, then, that the fair market value of a single premium life insurance policy for the purpose of determining taxable gain derived from exchange of insurance policies is the same price that any person of the same age, sex, and condition of health as the insured, would have to pay for a life policy with the same insurance company on the date the exchange took place. * * *
[Parsons v. Commissioner, 16 T.C. 256, 262 (1951).]
But this caselaw all involves paid-up policies. The Schwab-Kleinman policies were not paid up, but instead required years more of steep premium payments. And substantial parts of their values were tied to the fluctuations of a broad stock-market index. How should a court measure their fair market values? At least in the context of the gift tax,
the regulations offer some guidance. For life-insurance policies that have “been in force for some time and on which further premium payments are to be made,” the insurer’s policy reserves are used to approximate the value.
Sec. 25.2512-6(a), Gift Tax Regs. But this method is not permitted when “the unusual nature of the contract” results in a valuation “not reasonably close to the full value.” Id. And the IRS, in Notice 89-25, 1989-1 C.B. 662, modified by Notice 98-49, 1998-2 C.B. 365, and Rev. Rui. 2002-62, 2002-2 C.B. 710 took the position that in the case of distributions from a qualified plan (which, remember, the distribution here was not), taxpayers could use the “stated cash surrender value” unless total policy reserves were “a much more accurate approximation of the fair market value of the policy.” Id. Q&A-10, 1989-1 C.B. at 665.
Lacking evidence of the insurer’s policy reserves,
we begin our look for the fair market value of each policy with what the insurer called its stated policy value, which was $48,667 for Schwab and $32,576 for Kleinman. But, unlike a traditional life-insurance policy’s value (which only grows over time), these policy values fluctuated with the stock market. They look more like the net asset value of a mutual fund (with the obvious difference that a mutual fund investment does not provide a death benefit), and the surrender charges look very much like a back-end load. Just as we wouldn’t ignore such charges in calculating the value of shares,
we don’t ignore them here.
The policies had been in effect for three years before distribution, meaning that in eight years the surrender charges would expire. During that time, with the ups and down of the stock market, the stated values of the policies could rise or fall.
The relevant point in time for our analysis is the time of distribution. The policies here were flexible-premium policies — that’s one of the reasons why not paying any but the first year’s premium didn’t end the policy in year two. But by the end of year three — when Angels & Cowboys distributed the policies to Schwab and Kleinman — the “no-lapse” premium period had expired and the fall in the broad stock market meant that they had no positive net cash-surrender value. As the Commissioner admits in his brief, flexible-premium policies like these generally “will not lapse if premiums are paid such that the net cash surrender value remains greater than zero.”
But the net cash-surrender values here were less than zero. And because the parties fought mostly about whether surrender charges could be considered at all, they introduced little evidence specifically directed at establishing the fair market values for the policies. What we had was the policies’ stated values, the amount of premiums to be paid and the amount of any surrender charges, the terms of the contracts to the extent the parties introduced evidence about them, and the observed behavior of the taxpayers (e.g., the lapse of Kleinman’s policy and the greatly reduced coverage for Schwab for which he picked up the premiums himself). The variety of insurance policies is too great to adopt as a general rule either the Commissioner’s simple proposition that surrender charges should never count, or Schwab and Kleinman’s that such charges should always count, in determining a policy’s value. The particular facts of this case feature neither the dramatically springing cash value described in Notice 89-25, Q&A-10, 1989-1 C.B. at 665, nor the ability to use the distributed policy as consideration for a new policy without regard to surrender charges.
The Commissioner proposes that we find as fact that the Schwab-Kleinman policies had value in addition to surrender value because “accumulated cash value can be used to pay costs relating to maintaining the policies in force, can be borrowed against, or can be obtained in exchange for surrendering the policy, as the policy owner may choose.” But the evidence does not convince us that such options were available to Schwab and Kleinman under the policies so long as the policies had negative net cash-surrender value, as they did on the date of distribution. Cf. Matthies, 134 T.C. at 152 (holding that the insurer’s acceptance of the stated account value of a life insurance policy as payment in full of a single premium due on a replacement policy supported the conclusion that the entire cash value of the exchanged policy should be determined without regard to surrender charges).
On this record we are not persuaded that at the time of distribution to Schwab and Kleinman the policies had significant value apart from the small amount of the insurance coverage that was attributable to the single premium that Angels & Cowboys had paid on each policy some three years earlier. Though the value is small, the calculation is daunting because of ambiguity in the record, and we make only a tentative effort to ascertain exact figures.
After distribution, the premiums covered Schwab for up to 54 days
and Kleinman for 24 days
— in Schwab’s case, until he paid a premium to keep the policy going, and in Kleinman’s, until her policy lapsed. By applying the base rates for the guaranteed maximum monthly cost of insurance rates ($.446 for Schwab, $.4043 for Kleinman)
to the days covered, we attribute the following amounts: to Schwab, $1,900.33; to Kleinman, $765.62 — a total of $2,665.95.
Section 72 generally treats as taxable the amount distributed less any amount allocable to a taxpayer’s investment in the contract— and for Schwab and Kleinman, whose corporation had paid the premiums without including them in their income, the amounts invested in their contracts were zero. We therefore conclude that $2,665.95 is the “amount actually distributed” under section 402(b) and therefore included in taxable income under section 72.
II. Penalties
The Commissioner wants us to impose section 6662(a)’s 20-percent penalty on Schwab and KLeinman’s understatement of tax, either because it’s a substantial underpayment — in this case, more than $5,000
— or because it arose from their negligence or disregard of rules or regulations. See sec. 6662(b)(1) and (2).
But the Commissioner miscalculated. Our tentative calculations is that the understatement of income is less than $5,000. It’s clear that although the parties still must make Rule 155 computations, the understatement of tax will not exceed $5,000.
We also believe that Schwab and Kleinman made a reasonable attempt to comply with the provisions of the Code, and that they were not careless, reckless, or in intentional disregard of rules or regulations. See sec. 6662(c). Section 1.402(b)-l(c), Income Tax Regs., did not mandate that Schwab and Kleinman relinquish consideration of surrender charges in determining tax. And while they did not account for the lingering benefit of Angels & Cowboys’s premium payment, its effect on their income was minimal. We will not sustain the Commissioner’s determination of the penalty.
Decision will be entered under Rule 155.