The Entrepreneurial Option Has a Boundary
A side job becomes a meaningful route toward full-time self-employment only when two questions are considered together: what has the person learned from the side work, and what would they give up by making it their main work? This final judgment brings the evidence on realized transitions together with the evidence on planned career stages. It does not produce a universal rule for when to quit. It produces a way to tell an information-rich opportunity from a financially necessary side job, and a deliberate delay from a failure to act.
Every transition has an information side and a protection side
The option-test perspective explains why people may work independently while retaining a wage job. The side activity can expose a venture to customers, reveal whether the person can deliver it, and clarify whether the life of self-employment is desirable. The employment relationship supplies something different: income, health insurance, predictable routines, and a fallback if the venture performs badly. A transition is therefore not a one-dimensional comparison between side-work revenue and salary. It is an exchange between two bundles of value.
The two research designs illuminate different parts of that exchange. The longitudinal moonlighting study observes employment states and asks which conditions are associated with later full-time self-employment. The staged-entrepreneurship study asks people at the beginning of their professional careers whether they intend to combine paid work with self-employment or enter paid work before becoming self-employed later. One provides evidence about movement that occurred during an observed period; the other provides evidence about how people anticipate timing and sequence. They cannot be connected person by person, but they can be placed in the same model of a career decision.
On the information side, a high share of income from moonlighting is the strongest empirical marker in the panel study. Moonlighters whose side work supplied at least half of their income were more likely to enter full-time self-employment, with an association of about eight percentage points in the panel-long outcome. The relationship remained positive when the share was measured continuously, when Logit and Probit models were used, and when the outcome definition changed. That repeated pattern is consistent with a side venture generating a signal strong enough to support commitment.
The signal remains ambiguous. A high income share may indicate demand, but it may also reflect low primary-job earnings or an already heavy dependence on unstable side work. The measure does not show margins, repeat customers, duration, scalability, or entrepreneurial intent. It cannot tell us whether the person learned that the venture can support full-time work or simply experienced a favorable period of contract work. The evidence is strong for an association and weaker for the precise psychological story behind it.
Primary-job attachment supplies a different signal. In the panel study, an additional 10 hours of work in the primary job was associated with a 2.7-percentage-point lower probability of transitioning during the sample period. Fewer hours may indicate that wage employment is insufficient or unstable, making self-employment a fallback. They may also leave more time for building a side venture. The study measures hours, not the reason for those hours, so the association cannot distinguish pressure from opportunity.
Primary-job income makes the boundary sharper. Lower income was expected to push moonlighters toward entrepreneurship, but primary-job income was positive and significant in five of six specifications. That pattern is consistent with resources—savings, credit access, or other capacity to absorb risk—or with unobserved ability and opportunity. A weak attachment to a job and a stronger financial position can coexist. The person may have less reason to remain in the current role and more ability to carry the consequences of leaving.
The protection side can block an otherwise attractive option. Employer-provided health insurance was negatively associated with entry in both transition horizons. In the full sample, the estimate corresponds to about a 3.7-percentage-point lower probability in the longer-run model; the negative relationship was clearer in the men-only estimates. The paper interprets this as job lock: leaving the primary job can mean losing a benefit whose value is not captured by the venture’s current earnings. The person may not be rejecting entrepreneurship.