In 1931, Paul Dirac predicted a particle with the same mass as the electron and the opposite charge. This particle, the "positron", was observed the following year, in a cloud-chamber photograph taken by Carl Anderson. Of course, neither envisaged any medical applications of the discovery. But, today, PET scanners identify tumours by capturing the gamma rays that are emitted when positrons annihilate inside a patient's body. GPS satellites correct their clocks for the bending of spacetime, and the World Wide Web was born so that physicists could share information. Far-reaching as they are, such consequences of fundamental research cannot be easily anticipated, nor quantified in conventional economic terms.
The High-Luminosity upgrade of CERN's Large Hadron Collider (HiLumi LHC) now poses a similar challenge. No economist can attach a probability, let alone a monetary value, to physics results that have not yet been obtained. The project has nonetheless undergone a thorough cost-benefit analysis. Deliberately cautious, the analysis leaves out discoveries altogether and counts only benefits that can be reasonably estimated. Even on this conservative basis, the returns are still expected to be nearly twice the investment costs, with any new knowledge representing an additional gain.
Everything is measured against a future without the HiLumi LHC. In this counterfactual scenario, the LHC keeps running as it is until around 2030 and is then switched off. Costs and benefits are worked out for both scenarios, and only the difference is credited to the upgrade. Because many of the parameters cannot be evaluated in advance, the study relied on fifteen critical variables whose values were randomly sampled from plausible ranges. A total of 50 000 simulations were performed using this approach. In 94% of them, the outcome was a net positive.
The biggest share of the benefits, about 40.3%, comes from the training of people. Thousands of students and young researchers are contributing to the realisation of the machine and experiments and carrying their skills into the wider economy. The upgrade will maintain that knowledge flow well into the 2030s, and the careers that it seeds will continue for another half a century. A further 37.7% of the benefits comes from industrial suppliers, who pick up new techniques and open new markets by building equipment for the collider, and from software developed for the programme and made available free of charge. The remaining benefits come from visitors, scientific publications and the value that people place on fundamental research.
The approach developed for the LHC and the HiLumi LHC has also been applied to synchrotron light sources, hadron therapy and Earth-observation satellites. More recently, it was used to assess the proposed Future Circular Collider, which is likewise expected to repay society more than it costs. With each study feeding and calibrating the next, the economics of Big Science is advancing much like the instruments it studies.
