The biggest bet any founder makes is themselves, and equity is the only instrument that lets it compound.
My parents asked me recently, at a family dinner, why I am wired the way I am: why keep launching companies, when a normal desk job was available the entire time. I gave a short answer, but it made me pause, and now this is the longer answer. No career path avoids suffering, so the only real decision is which kind, for how long, and whether you picked it or it picked you. The only language I trust to be honest about that is return and risk.
It boils down to risk and returns, where outcomes are produced through alpha (the part of a return not explained by the market). If you own the S&P 500, you get market returns. If you want more, you have to be doing non-consensus, and be right about it, or you’re just holding beta with a story attached.
The reality is most people apply that framework to only their portfolio and never to what generates money: their own time. The largest position you will ever take is the decade of execution you commit somewhere, and for most people that position is unpriced, undiversified and mis-directed, and sitting inside a structure built to capture the upside on your behalf.
The allocation problem
Time and capital deployed inside an existing structure return roughly what that structure returns. A large company has already found its product, priced its risk, and built the machine that converts effort into revenue. The marginal contribution of any one operator is compressed by design, which is what makes the system durable, and why the return on your judgment inside it is priced to the mean.
Compensation makes this explicit. Most people believe their salary is a measurement of what they are worth. However, a salary is a market clearing price for a role, benchmarked against everyone else doing approximately that role. You can outperform that estimate substantially and the number mostly will not move, because the pricing mechanism is the role, not you.
Equity prices judgment instead of hours, with a lag long enough that the market cannot correct you in real time. Decisions you make in year one still pay you in year ten, and the compounding accrues to your balance sheet rather than to the entity that employed you while you learned. That is the whole structural argument for building something, and it has little to do with autonomy or the aesthetics of founding.
The decade filter
In my experience, starting a company takes at least ten years to produce an impactful outcome from a material event. That’s not a hard fast rule, but it’s the commitment horizon I use to decide whether a problem is worth the allocation, built from watching my own ventures take that long to resolve.
Once I internalized that duration of time, the selection criteria changed completely: an idea compelling for a quarter is a different object than a problem I’m willing to be wrong about in public for a decade. An innovator’s refuge is to become an entrepreneur, and I think the refuge part is the whole test. You start a company because you spent months trying to convince yourself to do something else, and the argument kept failing against a problem you couldn’t stop working on anyway. A clever idea doesn’t clear that bar on its own, since the cost of producing a plausible one has collapsed alongside the AI models that generate it.
The failed argument is the signal. Enthusiasm decays on contact with the first enterprise procurement cycle, and conviction in the business model rarely survives contact with customers. What survives is the observation that you were going to spend your attention on this problem regardless of who capitalized it, so you may as well own the instrument that captures the result. Underneath the ten years is the thing that makes them hard: suffering itself.
Dukkha
The Buddhist concept of dukkha is usually translated as suffering, the persistent gap between how things are and how you want them to be. That gap is life’s baseline; the only condition without suffering is death.
The question assumes a choice between suffering and its absence, and that choice doesn’t exist. The desk job carries its own suffering: the coordination tax, the political alignment work, the quarter spent building consensus for a decision that took an afternoon to reach and the accumulated weight of executing someone else’s thesis with your one non-renewable input. That is real suffering, and its defining feature is that you didn’t select it; it arrived with the seat.
The founder version is existential in a way employment structurally cannot be, since there’s no organizational buffer between a bad decision and its consequence, and the accountability doesn’t distribute. I’ve picked that version repeatedly, because you chose the problem, the constraint, and the specific way the next decade will be hard. That’s the actual luxury of agency, narrower than most people assume, because the suffering is fixed and all you get to pick is which suffering you’re metabolizing. Picking it is one thing. Getting paid for having picked it correctly is a separate mechanism, and it’s the one my parents’ question skips past entirely.
Equity as the conviction instrument
Standard financial advice pushes toward diversification, and that advice is correct for capital you cannot influence, since spreading across uncorrelated positions is the only free improvement available when you have no edge on any individual one. Your own execution is the exception: it’s the single position where you set the inputs, and concentration is what makes it valuable.
The math is asymmetric in a specific way. Downside on a concentrated decade is bounded and mostly legible in advance: forgone compensation, some years of lower liquidity, reputational uncertainty that resolves faster than anyone expects. Upside is unbounded and convex, since the same decade of accumulated context, relationships, and judgment keeps producing returns across whatever comes after, whether or not this particular company clears. The capital outcome is one branch. Compounded judgment is the other, and it doesn’t get marked down when a market turns.
Equity is the vehicle that lets the first branch pay out proportional to the second. Without it you still do the work, absorb the suffering, accumulate the judgment, and the financial compounding lands somewhere other than your account. With it, being right early gets paid late and at scale, the only structure where picking the suffering and owning the return are the same decision.
If you’re going to spend the decade anyway, the ownership structure is the variable most people leave unexamined while agonizing over the idea.
Where I could be wrong
The strongest argument against concentration is that most concentrated bets fail, and survivorship makes the returns look better than the base rate supports. I take that seriously. The distribution of outcomes for founders is brutal, and the median is well below what the median operator earns inside a functioning company. I’m describing what alpha requires, and alpha is by construction what most participants don’t earn.
The second argument is that the compounding-judgment claim is convenient and hard to falsify. A decade inside a well-run company at scale may produce judgment equally durable and better calibrated, developed against real distribution and real constraints instead of a market that may not exist. I’ve seen both paths produce excellent operators, and I don’t think the evidence cleanly favors one.
The diagnostic
So the honest answer to why I’m wired this way: I’m not wired to avoid suffering any better than anyone else. I just want to be the one who picked it, and to own the instrument that pays out if I picked correctly. The useful question, the one I should have given my parents instead of the short version, is what problem you’d still be working on in year ten, after the original thesis has been rewritten twice, after the market read you were most confident about turned out to be partly wrong, and after nobody is paying attention to whether you show up.