#424 | The 4% Rule and Life Insurance
A Refresher on the 4% Rule
Fewer things in finance have been more unjustifiably maligned than the “4% Rule.” As interest rates fell in the mid-2010s, much was written about how the 4% Rule was too aggressive for the current environment. Then equities ripped and the 4% Rule was too conservative. Then inflation spiked and the 4% Rule wasn’t enough. Then interest rates rose and the 4% Rule became the 5% Rule.
All of these knocks misunderstand the concept. The 4% Rule isn’t a rule in a strict sense. It’s a rule of thumb for a baseline income level that will reliably avoid failure in retirement. If you just ran pure stochastic asset models through historical asset returns and took the average result, you’d get something more on the order of 6.5%. But that’s not the question. The 4% Rule is about avoiding failure in retirement, not pushing the limits on potential upside. And it’s supposed to flex with broader capital markets and each person’s retirement income situation.
The Rule itself is calibrated to produce a 90% success rate over a 30 year time horizon on a portfolio that is half stocks and half bonds. Decrease the success rate or the time horizon and the income rate will increase. Pushing the portfolio into more stocks can also increase the potential income, but there are limits because of the higher historical volatility of equities. After the first year, the withdrawals are modeled to increase with inflation. Assuming an inflation rate of 2%, the initial 4% withdrawal grows to nearly 7% by the end of the 30th year. And income is quoted based on the initial portfolio amount in dollars, not as a percentage of the current account balance.
In the rubric of the 4% Rule, income is any distribution from the portfolio – including investment expenses and advisory fees. Technically, an advisor charging a wrap fee of 1% on an investment portfolio is taking 25% of the initial income capacity under the 4% Rule. Whether the advisor and asset manager are adding value in excess of their fees is a question that needs to be factored into the calculation for determining income. And, of course, taxes are also included as a distribution.
The key input into the 4% Rule is long-term capital markets assumptions. What will US equities and bonds deliver over the next 10 years? JP Morgan says stocks will do 6.7% and Vanguard says 2.8%. Research Affiliates puts total bond returns at 5.1% and BlackRock has them at 3.7% over the next 10 years. These are meaningful differences that directly affect potential safe withdrawal rates. Charles Schwab, for example, thinks that the right safe withdrawal number these days is closer to 4.5% for a 90% confidence and 4.8% for a 75% confidence. Morningstar pegged it at 3.7% for 2025.
For all of publicity and mountains of analysis that have gone into the concept of a safe withdrawal rate in broader personal finance literature, it seems as though effectively none of that gets applied to life insurance. It’s not even a part of the conversation. I can count on one hand the number of times that I’ve heard someone talk about illustrated withdrawals from a life insurance policy in terms of an initial withdrawal percentage on the cash value. Instead, illustrated results serve as the “reasonable” expectation. We trade the broadly accepted rule of thumb for the siren song of a specific number spat out in the illustration.
The 4% Rule and Variable UL
Consider the fact that Variable UL uses underlying mutual funds directly holding the same assets that underpin the 4% Rule. All of the logic that applies to the 4% Rule should also apply to Variable UL. And yet, you can run a VUL illustration at a 12% gross rate and kick out an illustrated initial income stream equal to 11% of the Account Value prior to distributions. Take a look at the relationship below between the net illustrated rate and the illustrated income stream:
The numbers above don’t even capture the full picture. Recall that the 4% Rule references all income from the policy, including investment expenses and advisory fees. Obviously there are other fees with life insurance as well, particularly COIs in these later durations. The chart below shows the same scenario as the one above but this time includes the average cost of COIs through the 30 year income period quoted as an increase to the initial withdrawal rate.
The reason that the illustrated income (both with and without COIs) exceeds the assumed net investment return is because there is zero asset volatility and complete knowledge of future returns. In that scenario, there is no need for conservatism. The illustration can run the solve with precision and leave no margin for error. Illustrations are set up, in other words, to do exactly the opposite of the 4% Rule.
That is by design. Illustrations have always been intended to be used as explanations of how life insurance policies work, not projections of returns. And yet, of course, life insurers, brokers and agents often use them in exactly the opposite way, as projections but not explanations. The loose and inherently conservative nature of the 4% Rule reflects the messiness and uncertainty of the real world. The precise and zero-margin nature of the illustration doesn’t reflect a superior strategy – it reflects the fact that illustrations are not projections. They are simply mathematical exercises with no connection to actual performance and should never, under any circumstance, be interpreted as anything other than that.
One lens that we can use to get a feel for how insurance policies can react in the real world is Life Insurance Sustainability Analytics (LISA), which I’ve referenced numerous times in previous articles. LISA runs stochastic return scenarios through life insurance policies to get a sense of real-world performance under the embedded capital markets assumptions. With a 50/50 stock/bond portfolio, the 4% Rule yields a 92% success rate in the LISA analysis for a VUL. If the 4% Rule works in the real world, then it should work for Variable UL too – and that’s exactly what LISA shows, even after taking COIs into account.
If you’re showing VUL, it’s probably smart just to go with the 4% Rule rather than what the illustration spits out for an income solve. The pitch for income out of a life insurance policy should never be that it beats the raw income in a traditional investment account. The pitch for life insurance is that it carries a death benefit, which itself has value, and that the income is tax controlled and potentially even tax free. Even with a lower income stream, the total package of benefits for a life insurance policy can be superior to a traditional investment account. That’s the pitch.
I would even go so far as to say that income should rarely be illustrated in life insurance sales. An illustration showing an overfunded policy without income is a valid demonstration of the overall efficiency of the policy. If you’re illustrating a 6% gross return with 0.25% in investment expenses and a long-term IRR of 5%, then it’s reasonable to say that the drag from policy charges is 0.75%. Of that 0.75%, something like 0.4% is likely from Cost of Insurance charges, which aren’t a charge at all – they’re simply pre-funding a future benefit. The question is whether the 0.35% of fees related to insurance costs, including commissions, is worth the tax advantages.
Taking income from the policy is more conceptual than mechanical. The client can take withdrawals that reduce the policy values or they can collateralize the cash value with policy loans. Withdrawals are a zero-cost option, assuming there’s no pro-rata surrender charge hanging around, and fixed policy loans have a stated cost. There is a little bit of complexity that needs to be explained, but life insurance is way simpler than a brokerage account where tax treatment can be all over the place. Clients should have an understanding of how to access their cash in either circumstance.
Illustrating income is unnecessary to make the sale, but it sets potentially dangerous expectations. The last thing a policyholder should do is get a number in their head for income that we know, with absolute certainty, is not the right number for what they’ll actually be able to take systematically from their policy in retirement. The beauty of the 4% Rule is that it works for the moment when you want to take income. Unlike an illustration, it doesn’t set up an expectation 20 years prior to taking income about what exactly that income will be. That’s when things get dangerous.
The 4% Rule and Indexed UL
How does the concept of the 4% Rule apply to fixed life insurance? That’s where things get interesting. Indexed UL is, by default, set up to be more aggressive than VUL despite lower illustrated rates. On a 45 year old Preferred Male, Symetra Ascent IUL 2.0 produces a 30 year withdrawal percentage starting at age 65 of 8.34%. F&G clocks in even higher at 8.66% – which is still not as high as Allianz at 8.92%. Nationwide and PacLife are more modest at 7.07% and 6.85%, respectively.
As I’ve written many times before, illustrated income from Indexed UL at the default maximum AG 49 rate is calibrated to a sub-50% probability of success with real-world return variability. The 4% Rule, by contrast, is calibrated for a 90% success rate. I scratched around on the internet and found a grid showing safe withdrawal rates by success rates and for a 60/40 equity/bond portfolio, a 50% safe withdrawal rate is 6.5% – which is still substantially lower than the withdrawal rate illustrated from an Indexed UL.
That’s only part of the issue. The other part is that Indexed UL illustrates such a high withdrawal percentage in part because of the “arbitrage” conditions on indexed loans. Symetra, F&G and Allianz illustrate enormous withdrawal percentages because all three companies tack fixed interest bonuses onto their engineered index allocations that flow through to the illustration as loan arbitrage to the tune of as much as 1.5%. If mere mortals could withdraw money from their brokerage account and be paid 1.5% for doing so, then, of course, the safe withdrawal rate would be much higher. But they can’t. Why not? Because they have to live in the real world, not the fabricated world of life insurance illustrations.
Removing illustrated loan arbitrage right-sizes the withdrawal rates. Allianz doesn’t even illustrate fixed loans, which is an inexcusable oversight. Using standard fixed loans, Symetra’s illustrated withdrawal rate drops to just 6.55% – almost dead-on the 6.5% safe withdrawal for the 60/40 portfolio with a 50% success rate. Nationwide, by contrast, uses very little illustrated loan leverage and the illustrated withdrawal rate barely changes.
The solution for Indexed UL is to completely ignore what the illustration says – just like VUL. I would argue that the 4% Rule still broadly applies to Indexed UL. I ran an Indexed UL product through LISA and, surprise surprise, the 4% Rule generates a 90% probability of success. However, there is an additional complexity with Indexed UL because unlike long-term capital markets returns, which have a historical basis, Indexed UL performance is dependent on the Cap provided by the life insurer. The chart below shows how the success probability changes with the Cap assumption using the original income stream.
However, this isn’t really the issue. This graph shows the probability of success of the illustrated income stream based on the 4% Rule for the illustrated Account Value in the 20th year. If we were to apply the 4% Rule at the time of withdrawal to whatever Account Value is available then, theoretically, the success percentages should be the same for all of the Cap rates. But that’s not what happens. Instead, the probability of success drops with the Cap even though all of the scenarios use the exact same 4% Rule income structure.
The reason is because lower Caps don’t just mean lower returns. A lower Cap is, essentially, a less efficient strategy for getting exposure to the S&P 500. Think about it this way – the 4% Rule is an acknowledgement that the future is unknown and that, as Morningstar and Charles Schwab have done, it should be updated every now and then to reflect capital markets assumptions.
The same is true for IUL. If equity return assumptions drop, then the withdrawal percentage should drop too. But the Cap is another factor. If the Cap drops, then the assumed income should drop because of lower policy performance but the illustrated withdrawal rate should drop as well. It’s a two-factor effect, especially as the Cap drops further. Gauging withdrawal rates on Indexed UL is actually more difficult than in Variable UL, despite the fact that Indexed UL ostensibly is a simpler product with downside protection. With Indexed UL, you’re dealing with both capital markets assumptions and non-guaranteed participation. With Variable UL, only capital markets assumptions are at play.
The 4% Rule and Whole Life
Whole Life, however, is a different beast entirely. The 4% Rule is a way to create conservatism out of chaos. Whole Life already does that within the product structure. The guaranteed premium and cash values create an underlying floor return. Only the excess is passed through in the form of the Dividend Interest Rate, which reflects the net yield of a diversified portfolio. Every year, the DIR gets updated to a new rate.
As a result, the math for income on Whole Life looks something like the amortization schedule on a mortgage. The guaranteed cash values provide the baseline “principal” that can be withdrawn systematically through the income period. The future cash value growth, which comes in the form of both guaranteed cash value growth and dividends, can be collateralized on a dollar-for-dollar basis. Something like this:
This is, of course, just a conceptual framework. Actual policy loans or bank loans have interest costs that will diminish the amount that can come out. Taking withdrawals from the policy will reduce guaranteed cash value growth because of surrendering paid-up additions. The importance is not coming to a specific number – it’s the fact that taking income from Whole Life is a gentle glideslope with very few and known moving parts.
There is a particular role for each type of permanent insurance in retirement income planning. Variable UL offers exposure to traditional assets but with tax control. Indexed UL offers downside protection with equity-linked upside potential. But when it comes to actually taking income, both of those products should get the same treatment as other assets.
The 4% Rule is a simple heuristic for dealing with uncertainty while maintaining ample margin for error. Life insurance illustrations, by contrast, pretend that certainty exists and therefore reserve no margin for error for illustrated income. That is completely inappropriate for Variable UL and Indexed UL. The 4% Rule should apply to those products just like it does for other assets regardless of what the illustration shows.
Whole Life, by contrast, plays by different rules. Whole Life has such a stable return profile that it doesn’t need the 4% Rule. Conservatism is built into the product itself. The margin can run thinner because there is less variability. Whole Life serves as the zero-risk part of the income stream, which allows for more flexibility in taking income from risk-on assets such as equities. Usually, this level of stability requires a huge return tradeoff. But for Whole Life, the cost of stability is simply time. You have to wait for a Whole Life policy to really deliver the goods. And isn’t that the point of long-term planning anyway?
