A large behavioral corporate finance literature links overconfident chief executives to overpaying for acquisition targets, but the supporting estimates come from qualitatively different proxy families: stock-option-based measures (delayed option exercise, Holder67-type), press-based hubris indices, and a newer direct measure built from the synergies forecast error (the deviation of forecasted from realized operating synergies). These families have never been tested against each other on a common sample of acquisitions, and psychometric evidence indicates that extant executive-overconfidence measures fail convergent validity. The problem is to determine, within one harmonized empirical design, whether the proxy families identify the same underlying overconfidence construct and whether they yield consistent estimates of the effect of CEO overconfidence on the takeover premium paid.

Background

The takeover premium in mergers and acquisitions (M&A) is the percentage by which an offer price exceeds the target firm's pre-announcement market price; it is the canonical measure of how much a bidder pays above the target's standalone value. Roll's hubris hypothesis (Journal of Business, 1986) proposed that systematically overconfident bidders — executives who overestimate the returns to combining two firms — win contests against equally or better-informed rivals and thereby overpay, a corporate version of the winner's curse.

Empirical work had to make 'CEO overconfidence' observable from archival data, and three proxy families emerged. First, stock-option-based measures: Malmendier and Tate ('CEO Overconfidence and Corporate Investment', Journal of Finance, 2005) introduced the Holder67 indicator, which classifies a CEO as overconfident when the CEO, despite large undiversified personal exposure, repeatedly fails to exercise company stock options that are far 'in the money' (above a 67% moneyness threshold) — behavior hard to reconcile with rational exercise and interpreted as revealed belief in the firm's future outperformance. In 'Who Makes Acquisitions? CEO Overconfidence and the Market's Reaction' (Journal of Financial Economics, 2008), Malmendier and Tate used this and a press-based measure to show that overconfident CEOs are more acquisitive (especially via diversifying deals) and that the market reacts more negatively to their bids; notably, that paper estimates acquisitiveness and announcement returns rather than a premium equation. The option-based evidence on premiums itself comes from Croci, Petmezas and Vagenas-Nanos (International Review of Financial Analysis, 2010), who study UK acquisitions and report that overconfident (option-based) managers overpay, and from Pavićević and Keil (Strategic Management Journal, 2021), who find a positive relation between the option-based overconfidence indicator and acquisition premiums in 349 S&P 500 deals from 2001–2018. Second, press-based measures: Hayward and Hambrick (Administrative Science Quarterly, 1997) constructed a CEO hubris index from media portrayals and linked it to higher premiums in large acquisitions, and Malmendier and Tate (European Financial Management, 2005) showed that a 'confident/optimistic' press characterization corroborates option-based results for corporate investment — a cross-proxy corroboration that has, however, no analogue for the premium relation. Third, direct forecast-error measures: Ismail and Mavis (International Review of Financial Analysis, 2022) propose the synergies forecast error (SFE), the deviation between the operating synergies management forecast at deal announcement and the synergies actually realized after integration; CEOs with high SFE pay higher premiums and their acquirers earn lower returns.

The measurement foundation is contested. Hill, Kern and White (Journal of Business Research, 2014) analyzed the convergent validity — the degree to which different instruments capturing the same construct correlate — of seven unobtrusive executive-overconfidence measures drawn from secondary data, and found they do not adequately converge, concluding that existing measures are not measuring the same construct; an expert sort task likewise rated none of the seven as sufficient. Each family also imposes its own sample restrictions: option-based proxies require detailed executive compensation data (available mainly for large listed firms), press-based proxies require substantial media coverage, and SFE requires disclosed synergy forecasts plus years of post-deal realized performance. New instruments keep being proposed — a Bayesian-network probability measure (Hatoum, Moussu and Gillet, International Review of Financial Analysis, 2022), and shock-based identification strategies for overconfidence in acquisitions (Wintoki and Xi, Journal of Empirical Finance, 2026) — but none compares the established families against each other in the premium setting, and the meta-analysis of Miao, Coombs, Qian and Oh (Journal of Management & Organization, 2023) examines firm-performance relations with cultural moderators, not measurement convergence. The source studies' own open questions repeatedly flag the proxies as noisy, indirect, or sample-restricting and call for better measurement of managerial beliefs.

Problem Statement

On a common (or explicitly harmonized) sample of acquisitions, estimate the effect of CEO overconfidence on the takeover premium separately under each of the established proxy families — the stock-option-based measures (delayed in-the-money option exercise, Holder67-type), the press-based hubris measures, and the forecast-error-based synergies forecast error (SFE) measure — and determine: (a) whether these measures exhibit convergent validity as indicators of a common latent overconfidence construct, assessed on units (CEO, or CEO–deal, with the choice stated and justified by the researcher) on which the measures being compared are jointly observed; and (b) whether the sign, magnitude (under an explicitly stated standardization that makes estimates comparable across the differently scaled proxies), and statistical significance of the estimated overconfidence–premium relation are consistent across the families under a common specification of deal, target, acquirer, and CEO controls. Constructing and validating a new instrument is not required by the question.

Here 'common (or explicitly harmonized) sample' means either one overlapping sample on which the compared measures are jointly observed, or family-specific samples estimated under matched specifications with reported sample overlap; the latter suffices for (b) but not for (a), because correlations among indicators cannot be computed from non-overlapping samples. Where the three-way intersection of the families is small, pairwise joint observation of the measures being compared is sufficient for (a).

References

  1. Richard Roll, 'The Hubris Hypothesis of Corporate Takeovers', The Journal of Business 59(2), 1986. DOI: 10.1086/296325, https://doi.org/10.1086/296325
  2. Mathew L. A. Hayward and Donald C. Hambrick, 'Explaining the Premiums Paid for Large Acquisitions: Evidence of CEO Hubris', Administrative Science Quarterly 42(1), 1997, pp. 103–127. DOI: 10.2307/2393810, https://doi.org/10.2307/2393810
  3. Ulrike Malmendier and Geoffrey Tate, 'CEO Overconfidence and Corporate Investment', The Journal of Finance 60, 2005. DOI: 10.1111/j.1540-6261.2005.00813.x, https://doi.org/10.1111/j.1540-6261.2005.00813.x
  4. Ulrike Malmendier and Geoffrey Tate, 'Does Overconfidence Affect Corporate Investment? CEO Overconfidence Measures Revisited', European Financial Management 11(5), 2005, pp. 649–659. DOI: 10.1111/j.1354-7798.2005.00302.x, https://doi.org/10.1111/j.1354-7798.2005.00302.x
  5. Ulrike Malmendier and Geoffrey Tate, 'Who makes acquisitions? CEO overconfidence and the market's reaction', Journal of Financial Economics 89(1), 2008, pp. 20–43. DOI: 10.1016/j.jfineco.2007.07.002, https://doi.org/10.1016/j.jfineco.2007.07.002
  6. Ulrike Malmendier and Geoffrey Tate, 'Who Makes Acquisitions? CEO Overconfidence and the Market's Reaction', NBER Working Paper 10813, 2004. DOI: 10.3386/w10813, https://doi.org/10.3386/w10813
  7. Ettore Croci, Dimitris Petmezas and Evangelos Vagenas-Nanos, 'Managerial overconfidence in high and low valuation markets and gains to acquisitions', International Review of Financial Analysis 19(5), 2010, pp. 368–378. DOI: 10.1016/j.irfa.2010.06.003, https://doi.org/10.1016/j.irfa.2010.06.003
  8. Aaron D. Hill, David A. Kern and Margaret A. White, 'Are we overconfident in executive overconfidence research? An examination of the convergent and content validity of extant unobtrusive measures', Journal of Business Research 67, 2014, pp. 1414–1420. DOI: 10.1016/j.jbusres.2013.08.011, https://doi.org/10.1016/j.jbusres.2013.08.011
  9. Stevo Pavićević and Thomas Keil, 'The role of procedural rationality in debiasing acquisition decisions of overconfident CEOs', Strategic Management Journal 42(9), 2021, pp. 1696–1715. DOI: 10.1002/smj.3319, https://doi.org/10.1002/smj.3319
  10. Ahmad Ismail and Christos P. Mavis, 'A new method for measuring CEO overconfidence: Evidence from acquisitions', International Review of Financial Analysis 79, 2022, article 101964. DOI: 10.1016/j.irfa.2021.101964, https://doi.org/10.1016/j.irfa.2021.101964
  11. Khalil Hatoum, Christophe Moussu and Roland Gillet, 'CEO overconfidence: Towards a new measure', International Review of Financial Analysis 84, 2022, article 102367. DOI: 10.1016/j.irfa.2022.102367, https://doi.org/10.1016/j.irfa.2022.102367
  12. Chao Miao, Joseph E. Coombs, Shanshan Qian and In-Sue Oh, 'CEO overconfidence and firm performance: A meta-analytic review and future research agenda', Journal of Management & Organization 31, 2023, pp. 1814–1836. DOI: 10.1017/jmo.2023.58, https://doi.org/10.1017/jmo.2023.58
  13. M. Babajide Wintoki and Yaoyi Xi, 'Shocks to CEO overconfidence and the deflation of hubris in acquisitions', Journal of Empirical Finance 88, 2026, article 101745. DOI: 10.1016/j.jempfin.2026.101745, https://doi.org/10.1016/j.jempfin.2026.101745