Originally estimated to cost $4.4 billion, the U.S. House of Representatives voted to kill the project in the summer of 1992, when costs had risen to $8.25 billion, but it was saved by the Senate, although a $100-million cut below requested funds put the project further behind schedule, increasing its costs even more. By the fall of 1993 the estimated cost had risen to a minimum of $11 billion (equivalent to $18 billion today), in part because administrative overhead proved larger than anticipated, and refined calculations of expected beam losses lead to a magnet redesign. (There were to be about 10,000 of them in the ring.) The latter’s increased cost, about $2 billion, could have been avoided by accepting a smaller ring and its resulting lower energy, but that idea was rejected by upper scientific and academic management.

But not all of the project’s costs were included in the initial estimates, according to a DoE report completed four years after the ax came down. About $500 million for detectors, $400 million for operations needed before the lab was finished, $60 million for land purchases and $118 million for DoE project management were excluded from cost estimates. Crucial to projects of such a size, a project cost and scheduling system was never fully implemented, concealing substantial cost overruns, according to the report.