When the records don’t match reality

david murphy_NHH_ Trude Brun Wilhelmsen
When institutions rely on records, do those records match the activity they describe? David Murphy will defend his thesis on 15 September. Photo: Trude Brun Wilhelmsen
PhD Defense

2 September 2026 07:40

When the records don’t match reality

On Tuesday 15 September David Murphy will hold a trial lecture on a prescribed topic and defend his thesis for the PhD degree at NHH.

Title of the thesis:  

«Institutional Recycling» 

SUmmary

David Murphy´s dissertation asks: when institutions rely on records, do those records match the activity they describe? He calls the problem “institutional recycling”: ideas, ownership claims or transactions can appear more than once, making the underlying activity harder to see. 

In the first study, Murphy compares thousands of economics working papers from a large US research archive. He finds a steady decline in similarity across most of the period, followed by a recent spike in textual similarity: papers using more similar wording. The increase is concentrated in shared language rather than a comparable rise in similarity of meaning. The long-running trend has therefore changed direction, although the study does not determine why. 

In the second, coauthored study, they examine US-traded certificates representing European shares. Using lending-market data, they find sharp increases in the amount of stock on loan around ex-dividend dates, when entitlement to the next dividend changes. The increases are largest where dividend taxes create the strongest incentive, much smaller for UK shares, where that incentive is near zero, and sharply weaker in the post-2014 period of greater US regulatory scrutiny.

Together, the results strongly suggest a link to certificates issued before underlying shares were secured, a practice documented by regulators. The data do not identify individual traders or refund claims. 

In the third, coauthored study, the researchers examine Denmark’s 2016 requirement for better proof of share ownership. After the reform, unusual lending around dividend dates falls by about 95 per cent relative to neighbouring markets. Their model estimates that annual net dividend-tax revenue was about US$991 million higher than the path predicted without the reform. At the same time, they find no measurable negative effects on share prices, company investment or dividends over the period studied. 

Prescribed topic for the trial lecture: 

"How do dividends respond to tax rates?" 

Trial lecture:  

Karl Borch Aud, 10:15 – 11:00

Defense: 

Karl Borch Aud, 12:15 – 14:00  

Supervisors:  

Professor Floris Tobias Zoutman (main supervisor), Department of Business and Management Science, NHH  

Evelina Gavrilova-Zoutman, Department of Business and Management Science, NHH 

Members of the evaluation committee: 

Professor Øyvind Thomassen (chair of the committee), Department of Business and Management Science, NHH 

Professor Arun Advani, University of Warwick 

Associate Professor Sarah Clifford, University of Oxford