Introducing payment changes the nature of contribution and the composition of contributors

What it is

Introducing payment into a previously unpaid community is often read as plain added incentive — pay money, get more and better contribution. What actually happens is two more fundamental things. The nature changes: contribution shifts from "my expression, my gift" to "my labor, my deliverable," and participants start viewing the platform through an employment lens — piece rates, refusals, corner-cutting all become legitimate. The composition changes: payment is a filter; a crowd selected by hourly wage (needs money, indifferent to the task) replaces a crowd selected by interest (enjoys the work, identifies with the community), and the two differ completely in quality distribution, retention, and task preference. Titmuss's classic comparison of blood supply set the prototype: paying for blood attracts a different donor population, and its quality composition changes with it. Money is not a layer stacked on the existing system — it rewrites the system.

Why it happens

Three mechanisms operate in sequence. Frame switching is the most immediate: Heyman and Ariely's experiments show that the form of reward (cash / gift / nothing) determines which mental account the task lands in — under market norms people invest "worth the price," under social norms "worth the relationship," and the two effort curves differ fundamentally; once cash appears, the social frame rarely recovers. Motivational crowding out cuts deeper: since Deci's puzzle experiments, a large body of work confirms that extrinsic payment for inherently interesting tasks undermines intrinsic motivation — not because people turn bad, but because causal attribution migrates ("why do I do this? for the money") and the task's meaning gets repriced. Compositional drift is the most durable: payment changes who enters the system. Open source lived through the whole arc — a majority of Linux kernel commits now come from salaried developers, with the volunteer share steadily shrinking; not the death of volunteering but the occupation of the entry channel by the professional one. Together: introducing payment is a one-shot, hard-to-reverse system rewrite, not an adjustable incentive knob. The dose question — why small payments are worse than none — is unfolded by the next entry.

Studying it

• Paradigm: field experiments (introduce payment for a randomized subgroup within a real volunteer community, tracking volume, quality, and retention against controls); cohort comparisons (behavioral trajectories of paid vs. volunteer contributors in the same system); before-after policy designs (interrupted time series on composition metrics when a platform introduces or withdraws incentives).
• Variables: presence of payment, reward form (cash / gift / acknowledgment), and timing (at community founding vs. after maturity) as independent variables; intrinsic-motivation scale scores, contribution quality, retention, and entrant composition as dependent variables.
• Use in interface research: incentive framing — how the same amount is presented as "payment" versus "gift / reimbursement" in wording and visual design.
• Methodological caveat: meta-analytic crowding effects vary with how interesting the task is and whether pay is performance-contingent — check both moderators before extrapolating; compositional drift needs years of observation window to surface; short lab tasks barely register identity and belonging effects at all.

Where it stops holding

Crowding out has a clear scope: it holds only for tasks and people with pre-existing intrinsic motivation. For tedious chores with no appeal, payment only adds supply (crowding in, not out) — there it is the right tool. Already-professionalized subsystems cannot be crowded out (they are market relations to begin with). Real large systems are hybrids: salaried and volunteer developers coexist in open source, bots and humans coexist on wikis — so the question shifts from "should we pay" to "how do the two kinds of contributors coordinate": priority lanes for paid work, attribution norms, and respect for volunteer time are the new hybrid-governance problems.

Applying it

• Partition, don't blend: keep paid and volunteer tasks visibly separate in interface and naming, so the market frame does not seep into the volunteer zone.
• Probe with non-cash forms first: gifts, reimbursements, and acknowledgments erode intrinsic motivation less than equivalent cash; compensation for core volunteers especially avoids wage-like framing.
• Track composition metrics, not just totals: entrant sources, volunteer retention curves, task-choice distributions — totals rising while volunteer retention collapses is the early signal of the system being rewritten.
• Before paying, ask what problem is being solved: insufficient supply on joyless tasks fits payment; quality and depth of engagement usually get worse with it.
• Verification: in a small pilot, track intrinsic motivation and retention for both the paid group and adjacent volunteer groups, over at least one full contribution cycle, before any rollout.

Related

• Same group: V9.06.1 Unpaid contribution runs on interest, reputation, and belonging rather than money · V9.06.3 Too-low pay reduces participation more than no pay · V9.06.4 Seeing one's contribution adopted by others is itself a reward · V9.06.5 Crowdsourcing pay must cover real time spent, or it becomes hidden low-wage work
• Nearby: V8.01 Unequal Participation · V9.05.5 Paying per completed item induces fast, low-quality work
• Search terms: motivation crowding · market norms · paid contributors