Rankings / Candidate
Steven T. Berry
Yale University · Industrial organization
Currently #4, tier Darkhorse, score 19.7, named by 4 sources.Score history: 100.0, 25.0, 14.3, 14.3, 14.3, 14.3, 14.3, 19.7, 19.7, 19.7, 19.7
The work. The B in BLP. The 1995 paper with Levinsohn and Pakes showed how to estimate demand for differentiated products from market share data alone, with consumers who differ in taste and firms that set prices strategically.
The case for. BLP is the workhorse of empirical industrial organization. Every merger review runs some version of it.
The case against. Tirole took the IO prize in 2014 and Card, Angrist and Imbens took the empirical methods prize in 2021. A three-way split with Levinsohn and Pakes leaves no room.
The KmikeyM angle. BLP prices products against their rivals. KmikeyM is a market with one product and no substitutes, which is either the model's edge case or its refutation.
What the sources say
- Clarivate alumni (prior years) named Steven T. Berry: named in 2020: for their BLP random coefficients logit model for demand estimation. (heat 100) Preseason: the Clarivate alumni pool
- Scott Cunningham ranked Steven T. Berry #2 of 2: the backup pick, with Levinsohn and Pakes; as reported by Tol (heat 50) Polymarket opens with seventeen names, and six new sources join the board
- Nicholas Decker named Steven T. Berry: Oct 4, on X: "Confident enough in Bresnahan-Berry-Pakes that I think I'm going to write the article in advance" (heat 100) Polymarket opens with seventeen names, and six new sources join the board
- Maia Mindel named Steven T. Berry: secondhand, from a roundup of economists' threads (heat 100) Polymarket opens with seventeen names, and six new sources join the board