A Surprise Tax From Life Insurance Loans

 

For some reason, the taxability of life insurance seems to be an old reliable in tax controversy.

Granted, there are areas involving life insurance that are not intuitive. The taxation of a split-dollar life insurance policy to an employee can be a bit puzzling until you have studied it one or ten times. There is also the tax history of “janitors insurance,” which resulted in yet another tax acronym (“EOLI”), the creation of Form 8925, and the recurring question “what is the purpose of this form” from young tax accountants ever since.

 

No, what we are talking about is the income taxation of vanilla-ice cream-on-a-regular-cone life insurance. Life insurance is normally nontaxable. You can change that answer by not ordering vanilla.

David and Cindy Fugler bought permanent (that is, cash value) life insurance on their two children in 1987. There was the initial year payment, plus additional yearly premiums, some of which were paid by borrowing against the policies. After many years, they cashed-in the policies. The life insurance company sent Forms 1099, which the Fuglers did not report on their joint tax return.

COMMENT: As we have discussed before, the IRS loves to trace Forms 1099 to tax returns, as the process can be computerized and requires no IRS manpower. You, on the other hand, have no such luck and will likely contact your tax preparer/advisor – and incur a fee - to make sense of the notice. There you have current tax administration in a nutshell: increasingly shift compliance to taxpayers by requiring almost everything to be reported on a 1099. It is a brilliant if not cynical way to increase taxes without – you know – actually increasing taxes.

Here is a recap of the relevant Fugler numbers:      

		Policy #1		Policy #2
Cumulative premiums paid		6,850		6,850
Accumulated cash value		22,878		23,428
Outstanding loan & interest		(19,845)		(20,699)
Settlement check		3,033		2,729

 

On first impression, it might seem odd that the Fuglers did not report the two distribution checks: the $3,033 and the $2,729. This is the amount they received upon policy cancelation – and after repaying policy loans and related interest and whatnot charges. Then again, one does not normally expect to have taxable income from life insurance. One should still report the 1099 amount (so the IRS computers have something to latch onto) and thereafter adjust the numbers to what one considers correct. Without that latch, these IRS matching notices are automatic.

So, what do you think:

·      Do the Fuglers have income?

·      If so, what is the income amount?

To reason through this, think of the life insurance policies as savings accounts. Granted, inefficient savings accounts, but the tax reasoning is similar. If you put in $6,850 and years later receive $22,878, the difference is likely (some type of) income. The same reasoning applies to the second policy.

So, you have income. Is there some way to not have income? Sure, if the cumulative premiums you paid exceed any cash value. In that case any refund would be a return of your own money.

But what is the income amount: is it the checks they received: $3,033 (for policy #1) and $2,729 (for policy #2)?

Normally, this would be correct, but the Fuglers borrowed against the polices. The loan did not create income at the time (because of the obligation to repay). That obligation has now been repaid with cash that would otherwise have been included in those distribution checks. You cannot avoid income by having a check go directly to your lender. Tax advisors would have a field day if only that were possible.

I would say that the income amount is the cash received plus the loan forgiven: $16,028 (policy #1) and $16,578 (policy #2).

Before thinking the result unfair, remember that the Fuglers did receive the underlying cash. The timing for the taxation of the loan was delayed, but even that result was pro-taxpayer. This is not phantom income that we sometimes see in other areas of the Code.

There is some chop in the numbers for the loan forgiven. As you can imagine, there are all kinds of fees and charges in there, as well as possibly accrued interest on the loan.  The Fuglers thought of that also, arguing that the accrued interest should not be taxable – or at least should be deductible.

The “should not be taxable” is a losing argument, as all income is taxable unless the Code says otherwise. It does not, in this case.

That leaves a possible interest deduction.

The problem here is that Congress limited the type of nonbusiness loans whose interest is deductible: loans on a principal residence; loans used to buy or carry investments, college loans; loans (starting in 2026) on a new car with final assembly in the United States. Any other nonbusiness loans are considered personal, meaning the interest thereon is also personal and thus nondeductible.

The Fuglers could not fit into any of those deductible categories. There was no subtraction for interest, no matter what the insurance company called it.

The Fuglers had taxable income. They reported none of it on their return. The IRS – as usual – wanted interest and penalties and whatever else they could get.

The Tax Court agreed.

Our case this time was Fugler v Commissioner, T.C. Summary Opinion 2025-10.

Recurring Proposal For Estate Beneficiary’s Basis In An Asset

There is an ongoing proposal in estate taxation to require the use of carryover basis by an inheriting beneficiary.

I am not a fan.

There is no need to go into the grand cosmology of the proposal. My retort is simple: it will fail often enough to be an unviable substitute for the current system.

You might be surprised how difficult it can be sometimes to obtain routine tax reports. I have backed into a social security 1099 more times than I care to count.

And that 1099 is at best a few months old.

Let’s talk stocks.

Question: what should you do if you do not know your basis in a stock?

In the old days – when tax CPAs used to carve numbers into rock with a chisel – the rule of thumb was to use 50% of selling price as cost. There was some elegance to it: you and the IRS shared equally in any gain.

This issue lost much of its steam when Congress required brokers to track stock basis for their customers in 2011. Mutual funds came under the same rule the following year.

There is still some steam, though. One client comes immediately to mind.

How did it happen?

Easy: someone gifted him stock years ago.

So?  Find out when the stock was gifted and do a historical price search.

The family member who gifted the stock is deceased.

So? Does your client remember - approximately - when the gift happened?

When he was a boy.

All right, already. How much difference can it make?

The stock was Apple.

Then you have the following vapid observation:

Someone should have provided him with that information years ago.

The planet is crammed with should haves. Take a number and sit down, pal.

Do you know the default IRS position when you cannot prove your basis in a stock?

The IRS assumes zero basis. Your proceeds are 100% gain.

I can see the IRS position (it is not their responsibility to track your cost or basis), but that number is no better than the 50% many of us learned when we entered the profession.

You have something similar with real estate.

 Let’s look at the Smith case.

Sherman Darrell Smith (Smith) recently went before the Tax Court on a pro se basis.

COMMENT: We have spoken of pro se many times. It is commonly described as going to Tax Court without an attorney, but that is incorrect. It means going to Tax Court represented by someone not recognized to practice before the Tax Court. How does one become recognized? By passing an exam. Why would someone not take the exam? Perhaps Tax Court is but a fragment of their practice and the effort and cost to be expended thereon is inordinate for the benefits to be received. The practitioner can still represent you, but you would nonetheless be considered pro se.

Smith’s brother bought real property in 2002. There appears to have been a mortgage. His brother may or may not have lived there.

Apparently, this family follows an oral history tradition.

In 2011 Smith took over the mortgage.

The brother may or may not have continued to live there.

Several years later Smith’s brother conveyed an ownership interest to Smith.

The brother transferred a tenancy in common.

So?

A tenancy in common is when two or more people own a single property.

Thanks, Mr. Obvious. Again: so?

Ownership does not need to be equal.

Explain, Mr. O.

One cannot assume that the real estate was owned 50:50. It probably was but saying that there was a tenancy in common does not automatically mean the brothers owned the property equally.

Shouldn’t there be something in writing about this?

You now see the problem with an oral history tradition.

Can this get any worse?

Puhleeeze.

The property was first rented in 2017.

COMMENT: I suspect every accountant that has been through at least one tax course has heard the following:

The basis for depreciation when an asset is placed in service (meaning used for business or at least in a for-profit activity) is the lower of the property’s adjusted basis or fair market value at the time of conversion.

One could go on Zillow or similar websites and obtain an estimate of what the property is worth. One would compare that to basis and use the lower number for purposes of depreciation.

Here is the Court:

Petitioner used real estate valuation sources available in 2024 to estimate the rental property’s fair market value at the time of conversion.”

Sounds like the Court did not like Smith researching Zillow in 2024 for a number from 2017. Smith should have done this in 2017.

If only he had used someone who prepares taxes routinely: an accountant, maybe.

Let’s continue:

But even if we were to accept his estimate …, his claim to the deduction would fail because of the lack of proof on the rental property’s basis.”

The tenancy in common kneecapped the basis issue.

Zillow from 2024 kneecapped the fair market value issue.

Here is the Court:

Petitioner has failed to establish that the depreciation deduction here in dispute was calculated by taking into account the lesser of (1) the rental property’s fair market value or (2) his basis in the rental property.”

And …

That being so, he is not entitled to the depreciation deduction shown on his untimely 2018 federal tax return.”

Again, we can agree that zero is inarguably wrong.

But such is tax law.

And yes, the Court mentioned that Smith failed to timely file his 2018 tax return, which is how this mess started.

Here is the Court:

Given the many items agreed to between the parties, we suspect that if the return had been timely filed, then this case would not have materialized.”

Let’s go back to my diatribe.

How many years from purchase to Tax Court?

Fifteen years.

Let’s return to the estate tax proposal.

Allow for:

• Years if not decades
• Deaths of relevant parties
• Failure to create or maintain records, either by the parties in interest or by municipalities tasked with such matters
• Soap opera fact patterns

And there is why I object to cost carryover to a beneficiary.

Because I have to work with this. My classroom days are over.

And because – sooner or later – the IRS will bring this number back to zero. You know they will. It is chiseled in stone.

And that zero is zero improvement over the system we have now.

Our case this time was Smith v Commissioner, T.C. Memo 2025-24.

Can You Have Income From Life Insurance?

 

I was looking at a recent case wondering: why did this even get to court?

Let’s talk about life insurance.

The tax consequences of life insurance are mostly straightforward:

(1) Receiving life insurance proceeds (that is, someone dies) is generally not an income-taxable event.

(2) Permanent insurance accumulates reserves (that is, cash value) inside the policy. The accumulation is generally not an income-taxable event.

(3) Borrowing against the cash value of a (permanent) insurance policy is generally not an income-taxable event.

Did you notice the word “generally?” This is tax, and almost everything has an exception, if not also an exception to the exception.

Let’s talk about an exception having to do with permanent life insurance.

Let’s time travel back to 1980. Believe it or not, the prime interest rate reached 21.5% late that year. It was one of the issues that brought Ronald Reagan into the White House.

Some clever people at life insurance companies thought they found a way to leverage those rates to help them market insurance:

(1)  Peg the accumulation of cash value to that interest rate somehow.

(2)  Hyperdrive the buildup of cash value by overfunding the policy, meaning that one pays in more than needed to cover the actual life insurance risk. The excess would spill over into cash value, which of course would earn that crazy interest rate.

(3)  Remind customers that they could borrow against the cash value. Money makes money, and they could borrow that money tax-free. Sweet.

(4)  Educate customers that – if one were to die with loans against the policy – there generally would be no income tax consequence. There may be a smaller insurance check (because the insurance is diverted to pay off the loan), but the customer had the use of the cash while alive. All in all, not a bad result – except for the dying thing, of course.

You know who also reads these ads?

The IRS.

And Congress.

Neither were amused by this. The insurance whiz kids were using insurance to mimic a tax shelter.

Congress introduced “modified endowment contracts” into the tax Code. The acronym is pronounced “meck.”

The definition of a MEC can be confusing, so let’s try an example:

(1)  You are age 48 and in good health.

(2)  You buy $4,000,000 of permanent life insurance.  

(3)  You anticipate working seven more years.

(4)  You ask the insurance company what your annual premiums would be to pay off the policy over your seven-year window.

(5)  The company gives you that number.

(6)  You put more than that into the policy over the first seven years.

I used seven years intentionally, as a MEC has something called a “7 pay test.” Congress did not want insurance to morph into an investment, which one could do by stuffing extra dollars into the policy. To combat that, Congress introduced a mathematical hurdle, and the number seven is baked into that hurdle.     

If you have a MEC, then the following bad things happen:

(1) Any distributions or loans on the policy will be immediately taxable to the extent of accumulated earnings in the policy.

(2) That taxable amount will also be subject to a 10% penalty if one is younger than age 59 ½.

Congress is not saying you cannot MEC. What it is saying is that you will have to pay income tax when you take monies (distribution, loan, whatever) out of that MEC.

Let’s get back to normal, vanilla life insurance.

Let’s talk about Robert Doggart.

Doggart had two life insurance contracts with Prudential Insurance. He took out loans against the two policies, using their cash value as collateral.

Yep. Happens every day.

In 2017 he stopped paying premiums.

This might work if the earnings on the cash value can cover the premiums, at least for a while. Most of the time that does not happen, and the policy soon burns out.

Doggart’s policies burned out.

But there was a tax problem. Doggart had borrowed against the policies. The insurance company now had loans with no collateral, and those loans were uncollectible.   

You know there is a 1099 form for this.

Doggart did not report these 1099s in his 2017 income. The IRS easily caught this via computer matching.

Doggart argued that he did not have income. He had not received any cash, for example.

The Court reminded him that he received cash when he took out the loans.

Doggart then argued that income – if income there be - should have been reported in the year he took out the loans.

The Court reminded him that loans are not considered income, as one is obligated to repay. Good thing, too, as any other answer would immediately shut down the mortgage industry.  

The Court found that Doggart had income.

The outcome was never in doubt.

But why did Doggart allow the policies to lapse in 2017?

Because Doggart was in prison.

Our case this time was Doggart v Commissioner, T.C. Summary Opinion 2023-25.

New Life Expectancy Tables For Your Retirement Account

On November 7, 2019 the IRS issued Proposed Regulations revising life expectancy tables used to calculate minimum required distributions from retirement plans, such as IRAs.

That strikes me as a good thing. The tables have not been revised since 2002.

There are three tables that one might use, depending upon one’s situation. Let’s go over them:

The Uniform Lifetime Table

This is the old reliable and the one most of us are likely to use.

 Joint Life and Last Survivor Expectancy Table

This is more specialized. This table is for a married couple where the age difference between the spouses is greater than 10 years.

The Single Life Expectancy Table

Do not be confused: this table has nothing to do with someone being single. This is the table for inherited retirement accounts.

Let’s take a look at a five-year period for the Uniform Lifetime Table:

Age	Old	New	Difference
71	26.5	28.2	1.7
72	25.6	27.3	1.7
73	24.7	26.4	1.7
74	23.8	25.5	1.7
75	22.9	24.6	1.7

If you had a million dollars in the account, the difference in your required minimum distribution at age 71 would be $2,275.

It is not overwhelming, but let’s remember that the difference is for every remaining year of one’s life.

As an aside, I recently came across an interesting statistic. Did you know that 4 out of 5 Americans receiving retirement distributions are taking more than the minimum amount? For those – the vast majority of recipients – this revision to the life expectancy tables will have no impact.

Let’s spend a moment talking about the third table - the Single Life Expectancy Table. You may know this topic as a “stretch” IRA.

A stretch IRA is not a unique or different kind of IRA. All it means is that the owner died, and the account has passed to a beneficiary. Since minimum distributions are based on life expectancy, this raises an interesting question: whose life expectancy?

COMMENT: There is a difference on whether a spouse or a non-spouse inherits. It also matters whether the decedent reached age 70 ½ or not. It is a thicket of rules and exceptions. For the following discussion, let us presume a non-spouse inherits and the decedent was over age 70 ½.

An easy way to solve this issue would be to continue the same life expectancy table as the original owner of the account. The problem here is that – if the beneficiary is young enough – one would run out of table.

So let’s reset the table. We will use the beneficiary’s life expectancy.

And there you have the Single Life Expectancy Table.

As well as the opportunity for a stretch. How? By using someone much younger than the deceased. Grandkids, for example.

Say that a 35-year old inherits an account. What is the difference between the old and new life expectancy tables?

                                Old             48.5

                                New            50.5

Hey, it’s better than nothing and – again – it repeats every year.

There is an odd thing about using this table, if you have ever worked with a stretch IRA. For a regular IRA – e.g., you taking distributions from your own IRA – you look at the table to get a factor for your age in the distribution year. You then divide that factor into the December 31 IRA balance for the year preceding the distribution year to arrive at the required minimum amount.

Point is: you look at the table every year.

The stretch does not do that.

You look at the table one time. Say you inherit at age 34.  Your required minimum distribution begins the following year (I am making an assumption here, but let’s roll with it), when you are age 35. The factor is 48.5. When you are age 36, you subtract one from the factor (48.5 – 1.0 = 47.5) and use that new number for purposes of the calculation. The following year you again subtract one (47.5 – 1.0 = 46.5), and so on.

Under the Proposed Regulation you are to refer to the (new) Single Life Expectancy Table for that first year, take the new factor and then subtract as many “ones” as necessary to get to the beneficiary’s current age. It is confusing, methinks.

There are public comment procedures for Proposed Regulations, so there is a possibility the IRS will change something before the Regulations go final. Final will be year 2021.

So for 2020 we will use the existing tables, and for 2021 we will be using the new tables.