In Simple Words
Imagine paying a painter by the room without checking their work. If the painter feels well-compensated by a generous rate, they take extra care not to spill paint on the floor, reciprocating fair pay with high-quality effort even when nobody is watching.
Malawian sorters without quality penalties
- 5 MWK per unit
- Comparison rate
- 25 MWK per unit
- No clear change in sorting errors
Selected lowest and highest rates. Estimated trend adjusts for minimum accepted pay. Uncertain errors without penalties do not prove reciprocity.
See source →Research question
Standard principal-agent models predict that unmonitored piece-rate workers will reduce product quality to maximize speed and earnings, ignoring employer quality preferences.
Participants
N = 612 contracted workers across 1,461 worker-day observations in rural Malawi.
Study Design
Randomized field trial comparing error rates across piece rates (5 to 25 MWK) in the absence of explicit quality penalties.
Results
In the unmonitored condition, increasing piece rates has a negligible effect on error rates (+0.006 to +0.009 errors per unit, p > 0.05), indicating workers do not excessively compromise quality for speed when piece rates rise, consistent with gift-exchange reciprocity or anti-shirking norms.
Takeaway
Test for non-monetary reciprocity in piece-rate contracts before implementing costly monitoring systems that risk crowding out intrinsic worker goodwill.