Where Capital Comes From: Labor, Knowledge, Agency, Wisdom
In the money tree I made the case for time: plant early, water, never chop it down. But I left out the other half, and it’s just as important. A tree grows from a seed — and seed size matters.
If your capital is tiny, compounding still works, but it asks for brutal patience. A $1,000 seed and a $100,000 seed both double every seven years; one just starts the race decades behind. Small capital means you need more time, and needing more time is exactly what makes the plan hard to stick to. So the question this post answers is: where does the seed money come from?
It comes from work. And there are really only four kinds of work — sorted not by title, but by what you’re actually selling.
The four jobs, and what each one sells
1. The McDonald’s job — you sell labor. No barrier to entry. You can start tomorrow. That’s the gift and the trap: because anyone can do it, it pays the least, and you’re selling your hours cheap. Worse, it’s first in line to be automated. The thing with no barrier to entry is also the thing with no barrier to replacement — a robot or a kiosk does it next year and never asks for a raise.
2. The engineer or doctor job — you sell knowledge. A high barrier to entry: years of school, exams, training. That barrier is the point — it’s why the hours sell for more. This is the path I took. But notice what you’re still doing: selling time, just at a higher rate. And the barrier that protected you is getting lower. For the first time, AI is coming for knowledge work, not just muscle work. Expertise is safer than labor. It is no longer safe.
3. The business owner — you sell agency. Start small. A one-person company is enough. What you’re selling here isn’t time or even knowledge — it’s agency: accountability and resourcefulness. You own the outcome. When something breaks, no boss fixes it; you do, or it stays broken. That ownership is the hardest thing for an AI to take, because agency isn’t a task you can hand off — it’s the willingness to be the one holding the bag. The catch: a good one-person business still needs knowledge underneath it. Agency without skill is just busyness.
4. The investor — you sell wisdom. At the top, you stop selling your hours entirely and let capital do the work. What the investor sells is wisdom — judgment about what to own and the patience to hold it. And wisdom is the slowest thing to buy: it comes from experience, from thinking, from learning, and most of all from your own failures. There’s no school for it. But here’s the honest knot at the center of this path: investing takes capital — which is the very thing we set out to get. You can try to invest other people’s money instead, but attracting it is neither easy nor quick. So for most of us, investing isn’t the first job. It’s the one the first three jobs eventually pay for.
The ladder underneath them
Stack the four and a pattern shows up. Each rung sells something harder to copy than the one below:
- Labor — cheapest, replaceable by a robot.
- Knowledge — pricier, now replaceable by AI.
- Agency — hard to replace, but only works with knowledge under it.
- Wisdom — hardest of all, but needs capital to start.
It reads like a clean climb. It isn’t, quite. You don’t graduate from one to the next and abandon the last — the investor still needs knowledge; the business owner still does labor at weekends. The rungs stack; they don’t replace each other. And no rung is shameful. The McDonald’s job is honest work and a real seed. The point isn’t to sneer at the bottom — it’s to not get stuck there by accident, selling cheap hours to a machine that’s about to do them for free.
Where I have to be humble
I’ve really only lived one of these four from the inside — the knowledge rung. So take the rest as a map drawn from the edge of a town I haven’t fully walked:
- My replacement predictions are guesses. I don’t actually know how fast AI takes knowledge work, or which jobs it skips. I’m planning around the risk, not a forecast I trust.
- I’m only now stepping toward agency myself — building small apps and this blog, mostly through vibe coding. I’m a beginner at it, writing this as much to figure it out as to teach it.
- And I’m not a great investor. I have wisdom enough to know how little wisdom I have, which is most of what I’ve got. The failures that buy the rest, I’m still collecting.
So I won’t tell you which path is right for you. I don’t know your barriers, your talents, or your luck.
The part that’s the same on every path
Here’s what I am sure of, because it’s the one thing all four jobs share: none of them make you free by themselves. A high salary that’s fully spent is just a fancier McDonald’s job. The investor with no surplus has nothing to invest. The job only sets how fast you can fill the watering can. What you do with the water is the same on every path.
And it’s the boring thing I keep coming back to: spend under your means, and put money into the tree — consistently, every single month. Labor, knowledge, agency, or wisdom — pick the rung you can climb. But the climbing isn’t the point. The point is the surplus you carry up it, and the tree you feed with it, month after month, until the fruit feeds you back.