Albrecht - How to Spot a Monopoly (2025)
Publication: Works in Progress, Issue 21 (December 2025) Author: Brian Albrecht URL: https://worksinprogress.co/issue/how-to-spot-a-monopoly/
Summary
Albrecht argues that the standard proxies regulators use to measure competition — market concentration and markups — are fundamentally misleading because they cannot distinguish between markets that are dysfunctional and markets where the best firms are simply winning. He proposes the Olley-Pakes decomposition as a superior alternative: it measures the covariance between firm-level productivity and market share, capturing whether productive firms are actually gaining share.
He supports this with four case studies: the AT&T breakup (where all metrics agree), Colombia’s 1990s trade liberalisation (where Olley-Pakes reveals that reallocation, not individual improvement, drove productivity gains), post-2008 Britain (where the decomposition reveals a stalled reallocation engine that concentration metrics miss), and Novo Nordisk’s Ozempic profits (where high markups reflect genuine innovation returns).
Key Claims
- Concentration is an outcome of market processes, not a cause → static-vs-dynamic-efficiency
- The Olley-Pakes covariance between productivity and market share is a better competition diagnostic → olley-pakes-decomposition
- Post-2008 UK suffered from zombie firms blocking reallocation → uk-productivity-stagnation
- The biggest gains from trade liberalisation come from reallocation, not individual firm improvement
Methodology / Approach
Uses the Olley-Pakes productivity decomposition across four international case studies. Compares it with conventional concentration ratios and markup measurements. Plant-level data in the Colombian case.
Limitations
Does not address natural monopoly or network goods where competition is structurally unviable. Conflates static allocative efficiency with overall market health. The Bell Labs counterexample challenges the AT&T case. The framework cannot detect erosion of dynamic efficiency over time or distinguish productivity gains from demand manipulation.