This article gives you a simple system for solving two problems that break most financial plans: using targets that don’t keep up with inflation, and expecting yourself to suddenly switch from saving to spending at some arbitrary point in the future.
It does two things. First, it replaces fixed dollar targets with a unit that scales with inflation. Second, it uses that unit to define thresholds that guide gradual changes in your behavior over time.
The result is a system that keeps your numbers meaningful and makes the transition from saving to living off your money something you can actually follow through on.
TL;DR — The System
- Look up CPI-U (Consumer Price Index) on bls.gov
- Find your maximum annual retirement contributions (401k/403b + IRA) and call this $M$
- Calculate the ratio of $M$ to CPI-U and call this $N$:
$N = \frac{M}{\text{CPI-U}}$ - Once $N$ is calculated, never update it. Freeze that value permanently
-
Choose an expected nominal return on your investments and call this $r$ (for example, $r = 0.10$)
-
Calculate the amount of assets you need to generate $M$ in yearly returns. This is your Contribution Replacement Unit (CRU):
$\text{CRU} = \frac{\text{CPI-U} \times N}{r}$ -
Use CRU as your base unit (“Savings Drone”)
- Define net worth thresholds as multiples of CRU, such as $0.5 \times \text{CRU}$, $1.0 \times \text{CRU}$, and $2.0 \times \text{CRU}$
- As your CRU multiple increases, gradually reduce how aggressively you save
That’s the entire system:
- One base unit
- One frozen ratio
- And a set of thresholds that scale automatically with inflation
How the System Works
The system has two parts. First, you calculate a base unit — the Contribution Replacement Unit (CRU), or “Savings Drone” — which converts your annual contributions into an equivalent amount of capital. Then, you use that unit to define thresholds that guide how your behavior changes over time. The first step gives you the measurement. The second step tells you what to do with it.
Calculating the Contribution Replacement Unit
Before you can set thresholds, you need a base unit. This system uses the Contribution Replacement Unit (CRU), which represents the amount of capital required to replace your annual retirement contributions. I like to call this a savings drone, because it’s like a copy of me working to increase my savings.
The calculation is simple, but each piece has a specific role.
Here are the inputs and derived values:
| Variable | Meaning | How to Get It |
|---|---|---|
| CPI-U | Consumer Price Index (urban) | Look it up on the bls.gov website |
| M | Maximum annual retirement contributions (401k/403b + IRA) | Look up the current contribution limits |
| N | Number of CPI units represented by M | Calculate: N = M / CPI-U (set once, then fixed) |
| r | Expected nominal rate of return | Choose an assumption (e.g., 0.10) |
| CRU | Contribution Replacement Unit (Savings Drone) | Calculate: CRU = (CPI-U × N) / r |
Note: CPI-U can be found on the Bureau of Labor Statistics website at bls.gov.
The 3 Steps
The process works in three steps.
First, determine M, your maximum annual retirement contribution. This is based on the contribution limits for your workplace plan (401k or 403b) and your IRA.
Second, convert that into a ratio. Divide M by CPI-U:
N = M / CPI-U
This gives you the number of “CPI units” represented by your contribution. And this is the key step: once you calculate N, you freeze it. You do not update it over time.
That fixed ratio is what allows the system to scale with inflation. Even though actual contribution limits tend to lag behind inflation, your system doesn’t. By holding N constant and updating CPI-U, your effective contribution target adjusts automatically.
Third, calculate the Contribution Replacement Unit:
CRU = (CPI-U × N) / r
This answers a very specific question: how much capital do you need so that, at your expected rate of return, your investments produce the same annual contribution as you?
One important note: the rate of return r is nominal, not inflation-adjusted. In this system, inflation is already accounted for through CPI-U. Using a nominal return keeps the calculation internally consistent.
In my case, I use r = 0.10, which approximates the long-term CAGR of the S&P 500.
Once you’ve calculated CRU — or “Savings Drone” — you now have a base unit that scales with inflation and represents your own contribution in capital form.
If you’re actually going to use this, put it in a spreadsheet. Set it up once, and then all you need to do is update CPI-U periodically. Everything else will adjust automatically.
And that’s what makes the next step possible.
Creating Thresholds
Once you have your Contribution Replacement Unit (CRU) — or Savings Drone — the rest of the system is straightforward.
You create thresholds as multiples of that unit.
That’s it.
Each multiple tells you how your investments compare to your own ongoing contributions.
At 0.5 CRU, you’re still doing most of the work. Your contributions dominate, and your investments are playing a smaller role.
At 1 CRU, your assets are roughly matching you. Your investments are contributing about as much to your future as you are.
At 2 CRU, your assets are doing twice the work. Your investments are contributing about twice as much as you are.
This gives you a simple way to measure where you are on the spectrum from labor-driven progress to asset-driven progress.
And that’s the entire point.
The higher the multiple, the less dependent you are on your own continued effort to move things forward. Your system starts shifting from something you actively power to something that increasingly runs on its own.
What you do with that information is up to you.
This article isn’t going to prescribe specific thresholds or tell you exactly when to change your behavior. That’s a much broader question, and it depends on your preferences, your risk tolerance, and how you want to live.
But the general principle is straightforward.
As your CRU multiple increases, the importance of aggressive saving decreases.
Not all at once. Not as a switch. But gradually.
And that gives you a framework for making those adjustments in a controlled way, instead of guessing or relying on arbitrary numbers.
What Problem Does This Solve?
This system solves two related problems that quietly break most long-term financial plans.
The first problem is behavioral.
Most people are very good at learning how to save money. They build habits around spending less, optimizing everything, and pushing as much as possible into savings. The issue is that this behavior doesn’t automatically turn off when it’s no longer needed.
A lot of financial planning assumes that it will. It assumes that once you reach retirement, or some predefined milestone, you’ll just switch modes — from saving to spending — without much difficulty.
In practice, that doesn’t happen.
People don’t flip a switch. They carry their habits with them. So instead of gradually shifting into using their money, they often stay in aggressive savings mode far longer than they should.
This system solves that by replacing the “big switch” with a series of smaller, pre-defined thresholds. Instead of asking you to suddenly behave like a completely different person, it guides you through gradual changes in behavior over time.
The second problem is measurement.
Even if you could define the perfect set of thresholds, there’s still the question of whether those numbers will mean anything in the future. A fixed number might make sense today, but over time inflation erodes its usefulness. What looks like a solid target now can become too low to support the same lifestyle later.
So the target itself becomes a moving problem.
This system solves that by tying its thresholds to inflation. Instead of using fixed numbers, it anchors everything to a reference that adjusts over time, so the thresholds continue to represent the same level of purchasing power.
Put together, these two pieces do something important.
They give you a way to change your behavior gradually, and they make sure the numbers driving those changes stay meaningful over time.
Background
Changing your financial habits over time is not optional. It’s required. The way you behave when you’re building wealth is not the way you should behave when that wealth is supposed to support your life.
The problem is that most people don’t treat aggressive saving as a phase with an endpoint. They treat it as the goal itself. Save more. Spend less. Repeat forever. And if you do that long enough, you can end up in a pretty strange place — surrounded by wealth, but still acting like you don’t have any. Basically Ebenezer Scrooge, sitting on a pile of money and still trying to get by with one lump of coal.
A lot of retirement planning quietly assumes that this won’t be a problem. The idea is that you’ll just flip a switch. One day you’re saving aggressively, the next day you’re retired, and now you’re comfortably spending down your assets. Clean transition. New behavior. No friction.
That’s not how people work.
Behavior doesn’t change overnight just because your spreadsheet says it should. The habits you build over decades don’t disappear on command. If anything, they get stronger. So when people finally reach retirement, they often don’t spend the way the plan assumed they would.
Financial planners run into this all the time. People struggle to shift out of saving mode. They hesitate to spend. They keep optimizing, keep holding back, keep acting like they’re still in the accumulation phase. And as a result, many of them end up with more money at the end of their lives than they had when they first retired.
Which is a strange outcome if the whole point of saving was to eventually use the money.
So the real problem isn’t just building enough wealth. It’s making the transition from saving to living off of that wealth — and doing it in a way that actually sticks.
And then there’s a second problem layered on top of that: inflation.
Inflation is what makes long-term planning messy. Over time, prices go up. Not because your money is becoming more powerful, but because it’s becoming less powerful. It takes more dollars to buy the same things.
It’s easy to get turned around here, because the numbers go up. Your income might go up. Prices go up. Account balances go up. But what matters isn’t the number — it’s what that number can buy.
Inflation is essentially the process of prices rising across the economy, driven by a mix of factors like increased demand, rising costs, and expansion of the money supply. You don’t need to model all of that to feel the effects of it. You just need to notice that the same lifestyle costs more over time.
And this creates a second failure mode in planning.
If you set a target number and leave it fixed, it doesn’t stay meaningful. It drifts. Over time, that number becomes too low to support the level of spending you originally had in mind. The plan doesn’t break all at once — it slowly becomes maladjusted.
So now you have two problems working together:
- People struggle to change their behavior
- And the numbers they’re aiming at don’t stay constant
Which makes the whole idea of “just pick a number and stick to it” a lot less reliable than it sounds.
That’s where this system comes in.
How I Got Here
This started on a beach in Oceanside while I was working through some financial planning with ChatGPT. Not the worst place to accidentally build a financial system.
I kept coming back to a pattern that shows up over and over again: people spend decades training themselves to save aggressively, and then when they finally retire, they don’t switch out of that mindset. They underspend. They die with more money than they started retirement with, not because that was the plan, but because behavior never caught up to the situation.
At the same time, I was thinking about ideas like coast FI — the notion that once you reach a certain level, your investments start doing most of the work and you don’t have to push as hard anymore. That idea made sense to me, but what really clicked was this: it shouldn’t be a binary switch. You don’t go from “saving machine” to “spender” overnight. That’s not how people work, no matter how many spreadsheets say otherwise.
What makes more sense is gradual change — step-by-step adjustments in behavior as your situation evolves.
The problem is, as soon as you try to define those steps, you run straight into inflation. If you set a number in, say, 1976 and then try to use that same number twenty years later, you’ve gone through a period of tremendous inflation. The number isn’t just a little off — it’s way off. It’s too low because it doesn’t account for inflation, and it no longer represents the same level of purchasing power.
So the question became: how do you build a system where the numbers stay meaningful over time?
Instead of trying to solve inflation myself, I looked for something that already exists. The government already does this work. They publish the Consumer Price Index for All Urban Consumers (CPI-U), and they update it continuously using real data and professional methodology. It’s not perfect, but it’s real, it’s maintained, and it’s free — which is a pretty good price.
That led to a simple idea: instead of fixing numbers in place, tie them to CPI-U so they automatically adjust over time.
When I started working through that idea, I was still working through it with ChatGPT — and this is where things went sideways.
The response I got was a wall of criticism. CPI isn’t perfect. It’s not personalized. It has limitations. And sure — all of that is technically true. But then it went a step further and suggested that I could build my own version of an inflation index instead. Because apparently I had some spare time to become a one-man Bureau of Labor Statistics.
That’s where it lost me completely.
I’m not an economist. I don’t have access to national datasets. I don’t have teams of researchers or decades of institutional knowledge. The idea that I should replicate that work myself — when professionals are already doing it and publishing it for free — is ridiculous.
But this wasn’t just a one-off bad suggestion. This is a pattern I’ve seen before when working with ChatGPT, and it shows up especially in financial topics.
When you propose something new — especially something that breaks from the usual way of doing things — ChatGPT becomes extremely conservative. It gets very critical, very quickly. It points out every flaw, every limitation, every edge case.
But at the same time, it often gives the status quo a free pass.
It doesn’t balance your idea against what you’re already doing. It doesn’t ask whether the current approach has flaws that are just as bad or worse. It just critiques the new idea in isolation.
In this case, every flaw in CPI was treated as a serious objection. But the default alternative — just picking a number and never adjusting it for inflation — wasn’t questioned at all. No scrutiny. No pushback.
That doesn’t make any sense.
If you apply the same level of criticism to the status quo, it completely falls apart. A fixed number in an inflationary world is not just imperfect — it’s guaranteed to become too low over time. It drifts further and further away from reality.
So the comparison isn’t “CPI vs perfect.” The real comparison is:
CPI (imperfect, but continuously updated) vs. Fixed numbers (guaranteed to become too low)
Once I framed it that way, the decision was obvious.
That was the turning point. I was so annoyed. ChatGPT be damned, I solemnly swore in that moment that I would incorporate CPI-U into my financial system, and ChatGPT would just have to put up with it.
Everything else followed from that, and to my pleasant surprise, the system ended up working out very well.