The Real Cost of Benchmarking
Working paper. First version April 2024; this draft August 2026.
Abstract
Benchmark-linked capital flows increase firms' CAPM βs, thereby raising managers' perceived cost of equity and reducing investment. Using exogenous variation from Russell and S&P 500 reconstitutions, we show that inclusion in a benchmark stock index increases a stock's CAPM β. Managers interpret the higher β as a higher cost of equity and reduce investment. Consistent with this mechanism, benchmark inclusion also raises the perceived cost of equity among stock analysts and regulators. Industries with larger increases in βs due to benchmarking have accumulated less capital over the past two decades. Benchmark-induced changes in the cross-section of CAPM βs do not cancel out but affect aggregate investment because higher βs fall on many firms with high investment elasticities, while lower βs benefit a few large but inelastic firms.
FESE De La Vega Prize '26 FIRS PhD Paper Prize '25 SFS Cavalcade PhD Student Award '25
In plain language
When a stock joins a major index like the S&P 500, index funds buy and sell it together with every other member. That makes its price move more with the market, so its measured beta rises. Many companies use beta to set their hurdle rate, so the higher beta makes investment look more expensive and they invest less.
Key findings
- Joining a benchmark index raises a stock's CAPM β. The design uses Russell and S&P 500 reconstitutions.
- Managers treat the higher β as a higher cost of equity and reduce investment.
- Stock analysts and regulators also raise their cost-of-equity estimates after inclusion.
- The effects do not cancel out: higher βs land on many investment-sensitive firms, lowering aggregate investment.
Key figure
Cite this paper
Kontz, C. and Hanson, S. (2024). The Real Cost of Benchmarking. Working paper. SSRN 4884602.
Presentations and awards
- Seminars: Federal Reserve Bank of New York, Notre Dame Mendoza, NYU Stern, UT Austin McCombs, Boston College Carroll, Ohio State University Fisher, Bocconi University, University of Virginia Darden, National University of Singapore, Einaudi Institute, Copenhagen Business School, University of Maryland Smith, Singapore Management University
- Conferences: EUROFIDAI Paris December Meeting 2026*c, FiFi 2026, EFA Annual Meeting 2026, AFA Annual Meeting 2026, SFS Cavalcade 2025, FIRS 2025, Johns Hopkins Carey Finance Conference, Four Corners/FMRC Conference on Indexing, Investment Company Institute, Northeastern University Finance Conference 2025, UIC Finance Conference 2025c
- Awards: FESE De La Vega Prize 2026, SFS Cavalcade PhD Student Award, FIRS Conference 2025 Prize for PhD Students, Shortlisted for Brandes Center Prize
* scheduled, c co-author
In the media
- Bloomberg, Matt Levine's Money Stuff: All Public Companies Are Alike
- Frankfurter Allgemeine, Der Sonntagsökonom: Die ETF-Falle
- UC San Diego Rady School of Management, The Brandes Center: Roundtable Summary
Questions about this paper
What does this paper find, in plain terms?
Joining an index like the S&P 500 makes a company's stock move more with the market, which raises the cost of capital its managers believe they face. As a result, index inclusion leads firms to invest less, and the growth of index investing has held back investment across the economy.
What is the main finding?
Inclusion in a benchmark index raises a stock's CAPM β. A 10 percentage point increase in benchmarking intensity raises β by 0.25 on average. At a 6% equity risk premium that is about 150 basis points on the CAPM-implied cost of equity, and the cost of equity is still more than 50 basis points higher after six years. Firms respond by investing less: a benchmarking-induced 100 basis point increase in the CAPM-implied cost of capital lowers capital expenditures by a cumulative 10.2% over six years.
What data does the paper use?
The benchmarking intensity measure of Pavlova and Sikorskaya (2023), which is the share of a stock's market capitalization held by funds benchmarked to an index that includes the stock and covers around 90% of mutual fund and ETF assets. It is combined with CAPM β estimates and firm-level data for 1998 to 2018 and with the perceived cost of capital and hurdle rate data of Gormsen and Huber. Further evidence comes from Morningstar, Value Line, and I/B/E/S analysts, the authorized returns of regulated utilities and railroads, ISS Incentive Lab data on executive pay, and NBER-CES manufacturing industries.
How is the effect identified?
The main design is a difference-in-differences around the annual Russell index reconstitution, using stocks within 300 ranks of the Russell cutoffs. Changes in benchmarking intensity around the reconstitution then serve as an instrument for changes in β. A second natural experiment, additions to the S&P 500, shows a similar increase in β and a decline in investment.
How do I cite this paper?
Kontz, C. and Hanson, S. (2024). The Real Cost of Benchmarking. Working paper. SSRN 4884602. A BibTeX entry is in the Cite this paper section.
Where can I get the latest draft?
The current draft (August 2026) is available as a PDF on this site and on SSRN. The online appendix is a separate PDF.