This paper reports laboratory experiments that evaluate the performance of a flexible packagebidding format developed by the FCC, in comparison with other combinatorial formats. In general, the interest of policy makers in combinatorial auctions is justified by the laboratory data; when value complementarities are present, package bidding yields improved performance. We find clear differences among the combinatorial auction formats, however, both in terms of efficiency and seller revenue. Notably, the combinatorial clock provides the highest revenue. The FCC’s flexible package bidding format performed worse than the alternatives, which is one of the main reasons why it was not implemented.