Are contributions involuntary public good experiments inflated because subjects are given money for their initial endowments? There is evidence that people receiving small, one time "windfall gains" have a high marginal propensity to consume them, and when doing so, exhibit greater riskseeking behaviour. Similar effects may be present in voluntary contribution experiments, causing subjects to contribute more to public goods than they would if using their own money. The effect ofwindfall money is tested by comparing VCM contribution rates when subjects supply their own endowments with those when endowments are provided, while holding constant the distribution of total promised earnings.