Campbell R. Harvey
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Campbell R. Harvey has a total of 146 co-authors. Here are the top co-authors by number of shared publications:
John R. Graham
Duke University Durham North Carolina USA
32 shared publicationsGeert Bekaert
23 shared publicationsYan Liu
13 shared publicationsChristian T. Lundblad
10 shared publicationsClaude B. Erb
9 shared publications- Otto Van Hemert
Is the director of core strategies at Man AHL in London, UK
9 shared publications
Latest Publications
- Open Access
- 24 October 2026
- Published by Elsevier in SSRN Electronic Journal
Stagflation and Asset Management
We analyze asset performance across inflation and growth states using 107 years of monthly US data. We construct a systematic three-regime taxonomy classifying each month as Overheating, Stagflation, or Contained (our baseline) using real-time inflation and growth signals, unlike prior work relying on fixed inflation thresholds or ex-post NBER recession dating. Both high-inflation states produce poor returns: the 60/40 portfolio earns-1.3% p.a. in real terms in Overheating and-0.2% in Stagflation, versus 9.0% in Contained periods. We also condition on whether inflation is accelerating or decelerating. Under our six-month specification, accelerating inflation in Overheating is associated with materially weaker equity and bond returns than decelerating inflation, though this is sensitive to the measurement horizon. In Stagflation, these differences are generally less pronounced. This contrast may partly reflect timing: by the time Stagflation is confirmed, some deterioration may already be priced in and markets may be anticipating recovery.
- 24 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessStagflation and Asset Management
We analyze asset performance across inflation and growth states using 107 years of monthly US data. We construct a systematic three-regime taxonomy classifying each month as Overheating, Stagflation, or Contained (our baseline) using real-time inflation and growth signals, unlike prior work relying on fixed inflation thresholds or ex-post NBER recession dating. Both high-inflation states produce poor returns: the 60/40 portfolio earns-1.3% p.a. in real terms in Overheating and-0.2% in Stagflation, versus 9.0% in Contained periods. We also condition on whether inflation is accelerating or decelerating. Under our six-month specification, accelerating inflation in Overheating is associated with materially weaker equity and bond returns than decelerating inflation, though this is sensitive to the measurement horizon. In Stagflation, these differences are generally less pronounced. This contrast may partly reflect timing: by the time Stagflation is confirmed, some deterioration may already be priced in and markets may be anticipating recovery.
- 14 October 2026
- Published by Elsevier in Journal of Financial Economics
- Vol. 185
Disagreement of disagreement
- 14 October 2026
- Published by Elsevier in Journal of Financial Economics
- Vol. 185
Disagreement of disagreement
- Open Access
- 7 October 2026
- Published by Elsevier in SSRN Electronic Journal
The Price of Conviction
Disagreement is usually measured as the dispersion of investors' mean beliefs. We argue such an approach is incomplete because it ignores the conviction of beliefs: how tightly investors hold their views. Theoretically, we show that when investors disagree about both the mean and the variance of payoffs, prices load on conviction-weighted beliefs, and investors with high conviction exert outsized demand influence. This creates a mispricing channel governed by the correlation between directional views (optimism or pessimism) and conviction. Empirically, we employ a machine-learning ensemble of hundreds of diverse model-investors, measuring within each one a return forecast paired with its conviction. Consistent with the theoretical mispricing implications, our long-short optimism-conviction portfolio, formed where mean disagreement is high, earns significant excess returns (13.2% annualized six-factor alpha; 16.0% CAPM). The effect survives among easy-to-short stocks and a battery of robustness tests.
- 7 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe Price of Conviction
Disagreement is usually measured as the dispersion of investors' mean beliefs. We argue such an approach is incomplete because it ignores the conviction of beliefs: how tightly investors hold their views. Theoretically, we show that when investors disagree about both the mean and the variance of payoffs, prices load on conviction-weighted beliefs, and investors with high conviction exert outsized demand influence. This creates a mispricing channel governed by the correlation between directional views (optimism or pessimism) and conviction. Empirically, we employ a machine-learning ensemble of hundreds of diverse model-investors, measuring within each one a return forecast paired with its conviction. Consistent with the theoretical mispricing implications, our long-short optimism-conviction portfolio, formed where mean disagreement is high, earns significant excess returns (13.2% annualized six-factor alpha; 16.0% CAPM). The effect survives among easy-to-short stocks and a battery of robustness tests.
- Open Access
- 29 September 2026
- Published by Elsevier in SSRN Electronic Journal
AI and Scholarship in Financial Economics
I detail the 45-minute process of drafting a submission to the UCLA's Human x AI Finance conference that was held in April 2026. There was no human involved in evaluating the papers. My paper was one of four papers selected from 159 submissions. I focus on the implications for our profession including changes that journals need to make and how we evaluate Ph.D. students as well as the new challenges in faculty promotion cases. I also make recommendations with respect to institutional research integrity rules.
- 29 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAI and Scholarship in Financial Economics
I detail the 45-minute process of drafting a submission to the UCLA's Human x AI Finance conference that was held in April 2026. There was no human involved in evaluating the papers. My paper was one of four papers selected from 159 submissions. I focus on the implications for our profession including changes that journals need to make and how we evaluate Ph.D. students as well as the new challenges in faculty promotion cases. I also make recommendations with respect to institutional research integrity rules.
- Open Access
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
Measuring Investor Disagreement: Proxies, Pitfalls, and a Path Forward
Disagreement among investors is central to modern theories of trading and asset prices, yet it is fundamentally unobservable. We review the empirical literature that measures it, which now spans dozens of proxies drawn from analyst forecasts, return volatility, trading and short interest, institutional holdings, option prices, and, more recently, text and machine learning methods. We organize this literature around the constructs these proxies are meant to capture, their data requirements, and the alternative interpretations that limit each one. A recurring theme is that the leading proxies are only weakly correlated with one another in the cross section of firms, so empirical conclusions about disagreement can depend heavily on which proxy is used. We discuss why this pattern arises, survey recent efforts to combine proxies and to discipline the choice among them and offer practical guidance for researchers who measure disagreement or who must assess findings that rest on a particular proxy.
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMeasuring Investor Disagreement: Proxies, Pitfalls, and a Path Forward
Disagreement among investors is central to modern theories of trading and asset prices, yet it is fundamentally unobservable. We review the empirical literature that measures it, which now spans dozens of proxies drawn from analyst forecasts, return volatility, trading and short interest, institutional holdings, option prices, and, more recently, text and machine learning methods. We organize this literature around the constructs these proxies are meant to capture, their data requirements, and the alternative interpretations that limit each one. A recurring theme is that the leading proxies are only weakly correlated with one another in the cross section of firms, so empirical conclusions about disagreement can depend heavily on which proxy is used. We discuss why this pattern arises, survey recent efforts to combine proxies and to discipline the choice among them and offer practical guidance for researchers who measure disagreement or who must assess findings that rest on a particular proxy.
- Open Access
- 12 June 2026
- Published by Elsevier in SSRN Electronic Journal
CFOs Meet LLMs
Business sentiment is a closely watched economic signal, but measuring it is slow and costly: surveys reach only a few hundred firms, arrive periodically, and take time to compile. We show that large language models hold the potential to address these shortcomings. We prompt an LLM to role-play as the CFO of a specific company at a specific date and focus on the economic-optimism question on the Duke-Federal Reserve CFO Survey over 2002-2025. We find that the LLM reproduces individual human responses: the predicted optimism score significantly forecasts the CFO's actual answer, surviving firm and year-quarter fixed effects and a control for the most recent prior response. Predictive accuracy increases with the amount of information supplied, as both respondent history and firm characteristics improve fit, and the relationship persists under quarterly aggregation. With appropriate conditioning, LLMs may be able to serve as credible digital twins of executives, offering scalable, high-frequency expectations data for financial research and policy.
- 12 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessCFOs Meet LLMs
Business sentiment is a closely watched economic signal, but measuring it is slow and costly: surveys reach only a few hundred firms, arrive periodically, and take time to compile. We show that large language models hold the potential to address these shortcomings. We prompt an LLM to role-play as the CFO of a specific company at a specific date and focus on the economic-optimism question on the Duke-Federal Reserve CFO Survey over 2002-2025. We find that the LLM reproduces individual human responses: the predicted optimism score significantly forecasts the CFO's actual answer, surviving firm and year-quarter fixed effects and a control for the most recent prior response. Predictive accuracy increases with the amount of information supplied, as both respondent history and firm characteristics improve fit, and the relationship persists under quarterly aggregation. With appropriate conditioning, LLMs may be able to serve as credible digital twins of executives, offering scalable, high-frequency expectations data for financial research and policy.
- 8 June 2026
- Published by Elsevier in Research Policy
- Vol. 55 (7)
Reprint of: The evolution of decentralized exchange: Risks, benefits, and oversight
- 8 June 2026
- Published by Elsevier in Research Policy
- Vol. 55 (7)
Reprint of: The evolution of decentralized exchange: Risks, benefits, and oversight
- Open Access
- 13 March 2026
- Published by Taylor & Francis in CFA Magazine
- Vol. 82 (2) , 46-68
Fundamental Growth
- 13 March 2026
- Published by Taylor & Francis in CFA Magazine
- Vol. 82 (2) , 46-68
Open AccessFundamental Growth
- Top Cited Papers
- 10 February 2026
- Published by Elsevier in Research Policy
- Vol. 55 (3)
The evolution of decentralized exchange: Risks, benefits, and oversight
- 10 February 2026
- Published by Elsevier in Research Policy
- Vol. 55 (3)
Top Cited PapersThe evolution of decentralized exchange: Risks, benefits, and oversight
- Open Access
- 1 January 2026
- Published by Elsevier in SSRN Electronic Journal
Disagreement of Disagreement
We show that major investor disagreement proxies have near zero correlation – making disagreement pricing inferences challenging. We develop a frictionless model that generates overpricing from latent disagreement, jointly rationalizes major proxies, and generates new testable predictions. Equilibrium interrelationships motivate a new disagreement measure, DIS, which is more predictive of returns cross-sectionally than major proxies and more consistently predictive across macroeconomic regimes. We provide evidence that the friction-based channel of Miller (1977) and a frictionless mechanism coexist and are empirically separable. DIS also provides a micro-founded, disagreement-based account of the IVOL puzzle, absorbing IVOL's predictability while revealing its regime and friction dependence.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.
- 1 January 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessDisagreement of Disagreement
We show that major investor disagreement proxies have near zero correlation – making disagreement pricing inferences challenging. We develop a frictionless model that generates overpricing from latent disagreement, jointly rationalizes major proxies, and generates new testable predictions. Equilibrium interrelationships motivate a new disagreement measure, DIS, which is more predictive of returns cross-sectionally than major proxies and more consistently predictive across macroeconomic regimes. We provide evidence that the friction-based channel of Miller (1977) and a frictionless mechanism coexist and are empirically separable. DIS also provides a micro-founded, disagreement-based account of the IVOL puzzle, absorbing IVOL's predictability while revealing its regime and friction dependence.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.