Pereira Pundrich, the Engineer Who Analyzes Corporate Communication's LanguageTHE ASSISTANT PROFESSOR OF ACCOUNTING, WHO HAS BEEN WORKING IN BUSINESS INTELLIGENCE IN THE PAST, STUDIES NON FINANCIAL INFORMATION THAT AFFECTS INVESTORS' DECISIONS
In a world where big data is expanding in all directions, Computer Engineering can be an appropriate background for a researcher of Finance and Accounting. At least, that is the path taken by Gabriel Pereira Pundrich, who studies merger and acquisitions, corporate governance and big data. Born in Brazil from a family of German and Italian immigrants, he started his career working in Brazil developing solutions for the banking industry and later as a consultant in Italy in the business intelligence sector. “I really enjoyed the idea that an engineer could design solutions for issues that affect the daily lives of people. I started seeing engineering as a powerful tool to understand business questions”. He turned down offers for a doctorate in Engineering and applied for a PhD in Accounting at the University of Technology in Sydney, Australia. “It was a risky transition. I have never regretted it”. He is now Assistant Professor at Bocconi’s Department of Accounting. In the next academic year, he will teach a new course on “Big data for business decisions”.
In 2016 Pundrich was awarded with a Young Researcher Grant from Bocconi for his works on merger and acquisitions. “More than 50% of all acquisitions fail”, he says. “WhatsApp was sold to Facebook for 21.8 billion dollars: how much will the company return to its investors? Microsoft acquired Nokia for 7.9 billion dollars: in the following years, 96% of the payment was impaired. There is a great inefficiency”. As a financial accounting researcher, he aims to find a way to mitigate the investors’ risk, to increase trust between them and managers, to propose, if needed, the disclosure of new information. In the working paper Recognizing intangibles following a business combination, Pundrich and Miles Gietzmann analyze the ways intangibles are defined and evaluated. “Most of the assets that explain the amount of money spent in the aforementioned acquisitions are intangibles such as trademark, customer relationship or technology. According to our first results, the less is known about the company that is being purchased, the less managers are able to recognize intangibles and thus they allocate them to ‘goodwill’. So it could be that managers do not really know what they are buying”.
Beyond the balance sheet
Pundrich ascribes to his engineering background the use of unstructured data that is public but not commercially available. “To open a data provider website and to download a dataset would be much easier, but I would get the same information that everyone else has”. For instance, he is currently involved with Miles Gietzmann and Francesco Grossetti in a textual analysis of documents signed by managers before bankruptcy. “Nowadays we have information about companies flowing 24 hours a day. They provide the chance to evaluate a company in real-time fashion. So, in the first part of the project, we use natural language processing to look through the information that managers voluntarily disclose to the market before a bankruptcy event. Are they providing over-positive, negative or neutral signals to investors?”. The second part of the project will be devoted to the analysis of information disclosed in press releases, news and social media. The researchers are going to compare what the company says about itself to what the market says about the company. “In most of the research done nowadays, they count the number of words used in a text. This is a useful tool, but it does not take into consideration double negatives or the order of the words. We are using more sophisticated techniques that allow us to spot where a negative information is provided in a document: is it at the beginning of the text or hidden at the bottom of the file? The balance sheet is no more the only source of relevant information”.
Mining for data, mining for gold
Pundrich has a story of looking for non-financial information. During his tenure in Sydney, he turned his attention to Australian mining firms. In Does industry specialist assurance of non-financial information matter to investors? (with Andrew Ferguson, 2015), he examined market reactions to specialist non-financial assurance. In this setting, the disclosure and the assurance of non-financial information, such as the amount of gold a company has found, are mandatory. The role is undertaken by a competent person that has some similarities to an auditor. “I collected the name of the geologists than sign those reports and measured their reputation. I compared the market reaction when the same amount of gold is discovered by two companies and assured by a high-reputation and a low-reputation geologist. I found that specialist assurance matters little in a low litigation environment like the Australian mining industry, yet it matters. This is one of the many forms of non-financial information that are not audited and not featured in the balance sheet, yet they drive investors’ decisions. The way information is disclosed to the market in real time is changing accounting”.
by Claudio Todesco