Do the Math: What SBP Is Actually Worth to Your Household
“Math for Marines” is a popular and sarcastic saying that pokes fun at the average Marine's academic ability. The better joke is how hard math actually is under pressure.
I scoffed at the notion as a second lieutenant, right up until our first armoury turn-in after a field exercise. Thirty-two Marines and one Corpsman went to the field. Thirty-three personnel, thirty-three rifles, thirty-three sets of NVDs, thirty-three K-Bars. I stood at the armoury window and watched every one of them turn in every piece of issued gear. With my own eyes, and in my Rite-in-the-Rain notebook, it was all there.
Two hours later we were still missing a K-Bar. I am not exaggerating: we were there all night, until somebody finally realised a K-Bar had been dropped in the bayonet cage, and nobody had counted the cage to discover there was an extra one.
Math is genuinely hard in the military. The cost of getting that one wrong was a few hours of a Friday night. The cost of getting your pay and benefits wrong runs to thousands — and with SBP you are making a half-million-dollar decision. Worth getting right rather than leaving to chance.
This is Part 2 of three. Part 1 covered the mechanics and the one deadline that matters. Here we build the math.
What Net Present Value Is, and Why Professionals Use It
You cannot compare a payment you make today to a payment your spouse receives in 2051 without a translation step. A dollar thirty years from now does not buy what a dollar buys this afternoon. You have felt that at the pump and in the grocery aisle over a matter of months. Over decades it is far larger: since roughly 1924, the U.S. dollar has lost about half its value every 24 years.
SBP asks you to give up part of your pension now, in 2026 dollars, so your spouse can keep receiving part of that pension after you die. If that happens in 2050, those are 2050 dollars — different by a factor of two. To judge the value of the future benefit honestly, every figure has to be converted to the same year. Net present value is what does that.
The only real trick is consistency: pick your assumptions and use the same ones everywhere. Throughout this series — and inside the free SBP Defeater calculator we built — those assumptions are:
- A 3% discount rate, because that is roughly what inflation has averaged in the U.S. economy since 1924.
- 10.51% when we model investing the premium instead, the S&P 500's historical average annual return since inception.
Two further choices need stating, because they cut against my own conclusion on purpose. When I model a financial decision I try hard not to rig it in my favour. I do not believe SBP is the best value for my household, so I have deliberately avoided understating it:
- We do not apply COLA to the SBP annuity. SBP payments rise with inflation every year, and leaving that out understates SBP by roughly 20% over a 15-year survivorship.
- We ignore the tax your spouse pays on the annuity. SBP income is taxable to them, which would reduce it by 10–37% depending on their other income.
Those two roughly cancel. The point is that if the argument survives with a thumb pressing on SBP's side of the scale, it survives.
My Household, Three Ways
This is not a hypothetical. We modelled our own decision nearly four years ago, and we are living the controlled experiment now — one that runs thirty years. So these are real numbers, and they are ours.
I retired as an O-4. My starting pension was $3,090.00 a month, which makes my SBP premium $200.85 and the annuity to my wife $1,699.50, all in 2026 dollars. Here is what SBP is worth depending on when I die:
| Scenario | NPV of premiums I pay | NPV of benefit she receives | Net |
|---|---|---|---|
| I die 1 year after retiring, she lives 39 more | $2,371.49 | $454,674.90 | +$452,303.41 |
| I die 25 years after retiring, she lives 15 more | $42,354.54 | $116,356.80 | +$74,002.26 |
| I outlive her (premiums stop at paid-up, 30 yrs) | $47,639.49 | $0.00 | −$47,639.49 |
Read those three rows and you have the whole program.
- Die immediately and SBP returns roughly 192 times what you paid. In any investing context that is an extraordinary outcome. Terrible for me personally, but extraordinary.
- Die at a statistically ordinary age and it returns 2.7x. Seventy-four thousand dollars of net value. That is not nothing, but it is a long way from the 192x headline.
- Outlive your spouse and you have handed the government roughly $47,600 in today's dollars for a benefit nobody ever collects.
First planning principle: every year you stay alive, the premiums you have paid go up and the value of the benefit goes down. SBP is worth the most the moment you unlace your boots for the last time, and it is worth less every day after. If you have no intention of dying soon, the headline value is not for you.
Rank Changes the Answer
The second variable is what you are retiring on, because both sides of the ledger scale with your pension. I wanted to compare roughly the low end of the scale against a reasonable high end — no admirals or generals were harmed in the building of this scenario. Same profile for three service members, each modelled as dying 25 years after retirement with a spouse surviving 15 more:
| Rank | Monthly premium | NPV premiums | NPV benefit | Net |
|---|---|---|---|---|
| E-7 @ 20 years ($2,736.60) | $177.88 | $37,510.49 | $103,049.19 | +$65,538.70 |
| O-4 @ 20 years ($3,090.00) | $200.85 | $42,354.54 | $116,356.80 | +$74,002.26 |
| O-6 @ 25 years ($7,930.50) | $515.48 | $108,703.12 | $298,630.28 | +$189,927.16 |
The O-6 nets nearly three times what the E-7 nets on identical life-expectancy assumptions. The reason is that premiums and benefits are percentages, not fixed dollar amounts.
Second planning principle: the higher your pension, the more SBP delivers in absolute dollars. A senior officer with a spouse likely to survive them has a much harder time justifying opting out than an enlisted retiree does.
The Variable Nobody Briefs: The Gap Between You
Here is where the honest analysis lives, and where the standard briefing goes completely silent.
SBP costs 6.5% for everyone. Same rate regardless of your age, your spouse's age, either person's health, or the gap between you. That flat pricing is a large cross-subsidy, and it means the program is underpriced for some households and badly overpriced for others. The question that decides which side you are on is simple: who is statistically likely to die first, and by how much?
In my case I am a man and my wife is two years younger. American women outlive American men by several years on average. Stack a two-year age gap on top of a multi-year longevity gap and the odds are reasonably good that I die first and she collects. That is the standard situation in the military, and it is why my numbers show positive value in the ordinary case.
Now flip it. If you are the service member and your spouse is older than you, or the longevity odds run in your favour rather than theirs, you are probably buying an annuity nobody will ever collect.
The only certainty in SBP is that the premiums cost something; the benefit is unknown. If your spouse predeceases you, premiums stop at their death and the damage depends entirely on how long they live. Here are total premium payments in 2026 dollars, by spouse longevity, assuming they die before you:
| Rank | Spouse survives 15 yrs | 20 yrs | 30 yrs (max, paid-up) |
|---|---|---|---|
| E-7 | $25,757.85 | $32,073.53 | $42,191.01 |
| O-6 | $74,644.69 | $92,947.12 | $122,266.98 |
An O-6 in that position pays somewhere between seventy-five thousand and a hundred and twenty thousand dollars, in today's money, for a benefit that never pays out.
Nobody wants to outlive their best friend, and age is not the only factor. Health belongs in this calculation too, and it cuts both ways: a retiree with a significant disability rating has a materially shorter life expectancy, which makes SBP more likely to pay. A spouse with a chronic condition makes it less likely. Nobody at outprocessing is going to ask you either question.
Third planning principle: the age and health gap between you and your spouse matters more than the premium rate does. Run it honestly before you sign, and if the odds say your spouse predeceases you, look very hard for a reason to stay enrolled.
The Thirty-Second Version
You do not need a spreadsheet for a first read. I have walls of them behind the calculator, but the screening math fits on a napkin.
You buy 55% of a pension for 6.5% of a pension. Divide: 55 ÷ 6.5 = 8.46. Now see how many times 8.46 fits into 12: 1.42. For every twelve months of premium you pay, your spouse collects about 1.42 months of benefit — the same ratio as 0.118 years. That gives you a simple linear screen:
(Your years left to live × 0.118) + your spouse's age when you die = the age they must reach to break even
Run it on my case, from retirement at 42, assuming I live to 79 — 37 more years:
- 37 × 0.118 = 4.37 years of benefit needed to pay the premiums back
- 4.37 + 77 (my wife's age when I would die) = 81.37
She would need to reach roughly eighty-one and a half to get our premiums back. Her life expectancy is 82.64, so she likely clears it — with about fifteen months of actual profit at the far end. You can find your own figures in the Social Security actuarial tables.
Two honest caveats on the shortcut: it ignores discounting, and it ignores the paid-up provision. It is a screening tool, not a decision. But if the number it produces sits well beyond your spouse's life expectancy, you have your answer without opening a spreadsheet.
The Objection Worth Taking Seriously
When my wife and I ran these numbers, the Admin Chief pushed back, and the pushback was good. His point: SBP is not an investment, it is insurance. You do not buy insurance for the expected case. You buy it for the disaster. And the disaster here is that I die early and she spends thirty-five years without the income we built our retirement around. In that scenario SBP pays nearly half a million dollars in today's money.
He was right, and the argument deserves its full weight rather than a wave of the hand. In fact, if you weighted every possible outcome by its actual probability instead of running a single median path, SBP would likely look better than the table above, not worse. That is not something I am hiding. It is simply that probability-weighting concentrates on the case SBP was designed for — an enormous payoff in an early death. This series is about what the majority will actually experience.
There is a shortcut for the other side of it too. If you are a man with a wife ten or fifteen years younger, and you have had a long, hard career with the injuries, diagnoses and exposures to match, you are the candidate on the SBP recruiting poster. Take it.
And one more thing before Part 3. This entire model assumes you would rather have more money than less. If your first priority is guaranteed income protection for a surviving spouse, and you are willing to accept a smaller total estate to get it, SBP is a solid choice — no matter what you read in Part 3. That is a large part of why you run the numbers and then talk it through with someone: to check that the answer still matches your values.
Part 3 asks whether there is a better option.
Run your own numbers rather than mine. The SBP Defeater is free and needs no signup. Open the SBP Defeater.
This content is educational in nature and does not constitute financial, tax, or legal advice. Life expectancy, tax treatment, and benefit eligibility vary by individual circumstance — confirm your own numbers with a qualified professional before acting. You are responsible for your own financial decisions.