Methodology

Annual pension = multiplier × years of service × high-3 salary. The multiplier is 1.0% for most retirements and 1.1% if you retire at 62 or later with at least 20 years of service.

Years of service at retirement = years already served + (retirement age − current age). High-3 is the average of your three highest-paid consecutive years; we let you enter it directly so the calculations apply whether you're GS-13, GS-15, or SES.

Lifetime value = annual pension × years drawing it. Default is 25 years (retire at 57, draw to 82). We do not model COLA. We skip COLA. COLAs track inflation anyway, and you're comparing nominal dollars on both sides.

Cash = (base + bonus) × years. Equity = a fresh RSU grant every year, valued at annualGrant at grant date, then compounded at your chosen growth rate until the end of the horizon. This mirrors how Meta, Google, Amazon, and similar pay senior ICs and managers, with annual "refresh" grants on top of the initial sign-on grant.

Default 8%/yr is the long run S&P average. Individual Big Tech stocks have done much more. Or much less. Use the slider if you want to reality check a different rate.

One equity grant at join. At the exit year, the grant value multiplies by your exitMultiple. Proceeds then compound at the growth rate for the remaining years. After exit, we assume Big-Tech-style annual RSU refreshes (post-IPO companies start paying like Big Tech).

To model "the startup didn't IPO," set the exit multiple to 0. The cash side is unchanged - that's the point: even if the equity goes to zero, you've still earned real money for years.

• TSP balance. You'd have one in either path; tech 401(k) match is usually larger.
• Social Security. Same on both sides for most career-feds.
• Taxes. Federal pension is ordinary income; long-held RSUs/IPO proceeds get long-term capital gains treatment. Taxes favor the tech side.
• FEHB vs. tech health insurance. Not included in the math at all. FEHB in retirement is a genuine benefit; partially offsets the gap but rarely closes it.
• Survivor / disability benefits. Genuine, but small relative to the wealth gap.
• RSU dilution, market downside, unvested grants if you quit. Genuine risks. Stress-test with the growth slider and a 0× startup exit.

Use this for orders of magnitude. Most feds massively overestimate what their pension is worth and massively underestimate what equity does over a decade of compounding.