Robert Tumarkin

UNSW Business School · Sydney NSW 2052 · Australia ·

I am a Senior Lecturer (Assistant Professor) with the School of Banking and Finance at UNSW Business School in Sydney, Australia. I teach Corporate Finance to undergraduates and Research Methods to postgraduates. This site provides information on my publications, access to proprietary data sets on CEO optimism and corporate board and committee meetings, and links to tools I have written to aid with research.


Getting Away with Merger: Proximity and Acquisition Activity

Working Paper

Mergers and acquisitions (M&A) exhibit a pronounced proximity effect, with half of all announced transactions in the continental U.S. involving firms that are less than 300 miles apart. Using the introduction of new flights between urban centers as a quasi-natural shock, we find that proximity induces M&A. The results suggest that information flow, particularly during the due diligence stage, is improved, helping firms complete deals and penetrate organizational opacity. While historical factors that influence M&A may relate to geographic proximity, our results indicate that travel is a more important determinant and that communication and travel infrastructure can create significant business value.

Death by Committee: An Analysis of Corporate Board (Sub-) Committees

Journal of Financial Economics, 2021, Vol. 141, 1119-1146

Theoretical models on group decision-making suggest that sub-group usage can affect communication among members and decision quality. To examine the trade-offs from forming sub-groups, we assemble a detailed dataset on corporate boards (groups) and committees (sub-groups). Boards have increasingly used committees staffed entirely by outside directors. Twenty percent of all director meetings occurred in such committees in 1996; this increased to over forty percent by 2010. We find evidence that such committee usage can erect barriers to communication and impair decision-making. Sub-groups are relatively understudied, but our results suggest that they play an important role in group functioning.

Does the Political Power of Non-Financial Stakeholders Affect Firm Values? Evidence from Labor Unions

Journal of Financial and Quantitative Analysis, 2018, Vol. 53, 1101-1133

While corporate political connections are known to enhance firm values, we demonstrate that union political activity can have the opposite effect. We examine the consequences of a recent state law in Australia that restricts union political activity, but does not change collective bargaining rights. In the wake of this law, the values of affected unionized firms significantly increase and, consistent with this market reaction, these firms are able to negotiate more favorable labor contracts than their unionized peers in other states. The evidence strongly suggests that unions use political activism to extract rents from corporations and benefit their members.

Stocking Up: Executive Optimism, Option Exercise, and Share Retention

Journal of Financial Economics, 2015, Vol. 118, 399-430

We show that an executive is optimistic about her company's prospects if and only if she retains some of the shares received whenever she exercises company stock options. Empirically, an indicator of optimism based on this idea matches the expected relations between optimism and corporate decision-making better than commonly used indicators based on the timing of option exercise. This makes sense, as our model of an executive's optimal option exercise and portfolio choice demonstrates that the timing of option exercise depends just as much on stock and other executive characteristics as it does on optimism.

News or Noise? Internet Postings and Stock Prices

Financial Analysts Journal, 2001, Vol. 57, 41-51

The anecdotal evidence is growing that postings in Internet financial forums affect stock prices, either because the postings contain new information or because they represent successful attempts to manipulate stock prices. From an investment perspective, knowing whether this phenomenon is pervasive is important. We examined the relationship between Internet message board activity and abnormal stock returns and trading volume in the period from mid-April 1999 to mid-February 2000. Our study focused on the discussion forum, an extremely popular site whose format permits the construction of an objective measure of investor opinions. For stocks in the Internet service sector, we found that on days with abnormally high message activity, changes in investor opinion correlated with abnormal industry-adjusted returns. These event days also coincided with abnormally high trading volume, which persisted for a second day. However, we found that message board activity did not predict industry-adjusted returns or abnormal trading volume, which is consistent with market efficiency.


Corporate Board and Committees

Annual meetings and described responsibilities

The dataset provides information about the meetings and allocation of information gathering and decision-making responsibilities of corporate boards and committees. The source is taken from definitive proxy statements (DEF14A) filed with the U.S. Securities and Exchange Commission.

The dataset is indexed by corporate Central Index Key (CIK) and proxy statement filing date. For each CIK-proxy statement filing date combination, the data provides the number of annual meetings and the Latent Dirichlet Allocation (LDA) weight of information gathering and decision-making responsibilities for the corporate board and its committees. Details on the natural language processing techniques used to generate this data is available in the paper and its online appendix.

Citation: Adams, R., Ragunathan, V., Tumarkin R., Death by Committee: An Analysis of Corporate Board (Sub-) Committees, Journal of Financial Economics, Forthcoming.

Updated 2020 October

Share Retainer

An indicator of executive optimism

Share Retainer is an indicator variable that identifies optimistic CEOs. It is defined as 1 (optimistic) if the cumulative shares retained by a CEO on days with option exercise during a fiscal year exceeds 1% and 0 (not optimistic) otherwise. When Share Retainer is not available for a CEO during a fiscal year, it is set to the most recent computed value.

The dataset covers CEOs that are found in both the Thomson Reuters Insider Filings (Thomson) and Compustat Executive Compensation (ExecuComp) datasets through 2021. Expanded datasets will be updated periodically. Please sign up for email notifications if you would like to be alerted when updated data is available.

Citation: Sen, R., Tumarkin R., Stocking up: Executive optimism, option exercise, and share retention, Journal of Financial Economics, 2015, Vol. 118, 399-430.

    Check here to confirm that your institution has active subscriptions to both the Thomson Reuters Insider Filings (Thomson) and Compustat Executive Compensation (ExecuComp) datasets.
Updated 2022 November


I primarily work in Haskell, a statically-typed, pure, functional programming language, with the occasional foray into imperative programming through Rust when computational speed is needed. The tools provided here are of general use for research in the social sciences. Please contact me should you think of any features that would be helpful in your work.


A fast multi-core fuzzy name matcher for CSV files. yenta allows users to dynamically combine numerous text matching algorithms easily and output multiple possible matches. Written in Rust, yenta emphasizes throughput and is capable of processing about 1 million fuzzy matches per minute on a 2019 8-core desktop. Documentation is available on the Github wiki.


A command line utility to locally index and download filings from the SEC Edgar database. edgar is multithreaded, allowing for non-blocking simultaneous downloads. Form downloading is query based allowing researchers to request specific CIKs, company names, form-types, and date ranges (or combinations thereof).


Export empirical specification results to LaTeX. Designed to be quick, easy, and powerful. Never spend days formatting tables again (in principle). Full documentation is forthcoming.


A NLP tool for grammatical extraction of data from text. GrammE implements an embedded domain specific language that allows researchers to quickly and flexibly implement exhaustive sub-graph searches of NLP dependencies data. Please email me for details and access.