Market spoiling stemming from information asymmetry has slowed the adoption grid-independent technologies that replace fuel-based lighting in the developing world. End users typically first experience lighting technology innovations via flashlights. The rapid emergence of inexpensive LED flashlights is a potentially good advancement in this regard, as LED lighting can be longer-lived, have higher initial light output, and be more energy-efficient than incandescent. However, our laboratory tests and end user interviews indicate that these products often fall far short of advertised performance levels and typically fail after a few months of use. Our study of purchasing decisions by 23 Kenyan market traders given an opportunity to purchase warrantied LED lamps found that prior experience with inexpensive LED flashlights significantly reduced their probability of purchasing (p = 0.0028). As additional evidence of consumer skepticism, in a large statistical survey, we also find that willingness to pay increases significantly once an LED lighting product is directly handled and tested by the end user. If LED lighting is to achieve its potential as a superior substitute for fuel-based lighting, effective policy measures are needed to remove the information asymmetry between expected and actual performance. One such measure, independent testing and certification, has measurably increased the quality of products available in the off-grid lighting marketplace.