BEYOND THE NUMBERS: MINING THE ANNUAL REPORTS FOR HIDDEN CUES INDICATIVE OF FINANCIAL STATEMENT FRAUD

Unlike previous fraud detection research, a vast majority of which has focused primarily on the use of quantitative financial information to predict fraud, in this study we examine qualitative textual content in annual reports to predict fraud and see whether there are discernible differences in the writing and presentation style between companies that committed fraud and those that did not. We believe that while numeric financial information in the annual reports can hide details of fraud, textual information relating to writing and presentation styles in such reports provides valuable clues pertaining to the existence of fraud. In this study we use the chi‐square test to analyse our data and test hypotheses about predictors of fraud that may explain linguistic feature variations in fraudulent and nonfraudulent annual reports. We provide new results on the usefulness of the qualitative content of annual reports in detecting fraud. Copyright © 2012 John Wiley & Sons, Ltd.     

Comparing numerical data and text information from annual reports using self-organizing maps

More and more companies provide their accounting information in electronic form today. The accounting information in electronic form can be found in large commercial databases or on the web. This information is of great interest for different stakeholders, i.e., stockholders, creditors, auditors, financial analysts, and management. For the stakeholders it is important to be able to extract both quantitative and qualitative information concerning the companies they are interested in. The annual reports contain information both in numerical and symbolic form. So far, only the numerical information has been analyzed with help of computers. However, technology has evolved and in particular neural networks in the form of self-organizing maps (SOMs) provide a new tool for analyzing also text information. In this paper, we compare results on quantitative data with results on qualitative data from annual reports. We use smart encoding, SOMs, and document histograms for comparing the performance of forest companies worldwide. Firstly, we cluster the companies according to, on the one hand, quantitative information, and on the other hand, qualitative information. Secondly, we compare the results produced by the clustering methods. Our results of the comparison show that there is a difference between the results.     

The application of text mining in accounting

By facilitating the derivation of knowledge and qualitative measures from textual data, text mining techniques have come into vogue in various domains and industries. Namely in accounting, text mining outputs can elucidate, complement, and validate the customary quantitative data. This study creates an up-to-date view of text mining applications in accounting practice. Through a critical review of text mining literature, insight is given into the stages of a typical text mining process, contemporary text mining techniques that have been named valuable in an accounting context, and the information that can be obtained by applying these techniques.     

The annual report tone and return Comovement—Evidence from China's stock market

Our paper uses A-share firms listed on Shanghai Stock Exchange (SSE) and Shenzhen Stock Exchange (SZSE) to explore the role of the textual tone of Chinese listed companies' annual reports on Chinese equities' return comovement. Our paper finds that the tone embedded in A-share companies' annual reports significantly increases Chinese stocks' return comovement as the tone exacerbates information asymmetry and reduces the quality of information. In addition, the annual report tone has a prominent effect on Chinese stocks' return comovement for stocks with low institutional ownership, private ownership stocks, and stocks with low market shares. Further investigation shows that the impact of the tone of the annual report on Chinese stocks' return comovement is strong during economic expansions, and when investor sentiment is high. An additional examination on hidden information in the textual tone suggests that published financial data do not explain a substantial part of the textual tone that is associated with the increase in the comovement, suggesting that private information should enhance informational opaqueness in the stock market.     

Discovering bank risk factors from financial statements based on a new semi‐supervised text mining algorithm

This paper aims to comprehensively uncover bank risk factors from qualitative textual risk disclosures reported in financial statements, which contain a huge amount of information on bank risks. We propose a new semi‐supervised text mining approach named naive collision algorithm to analyse the textual risk disclosures, which can more accurately identify bank risk factors compared with the typical unsupervised text mining approach. We identified 21 bank risk factors in total, which is far more than identified in previous studies. We further analyse the importance of each bank risk factor and how the importance of each risk factor changes over time.     

央行货币政策报告文本信息、宏观经济与股票市场

本文利用金融情感词典和文本分析技术,分析中国人民银行货币政策执行报告的文本情绪、文本相似度和文本可读性等多维文本信息,刻画央行货币政策执行报告的文本特征,探究货币政策报告的文本信息与宏观经济和股票市场的关系。实证研究发现,货币政策报告的文本情绪的改善会引起显著为正的股票市场价格反应,报告文本相似度的增加会引起股票市场波动性的显著降低,报告可读性对公布后股票市场的波动性影响不显著。货币政策报告文本情绪还与诸多宏观经济指标显著相关。进一步研究发现,引起股票市场显著反应的是报告文本情绪中反映货币政策指引的部分,而反映宏观经济历史状态的部分对股票市场的影响不显著。本文从文本大数据分析角度证明了我国央行沟通的有效性,对国内央行沟通相关研究形成了有益补充。     

Some effects on information load on search patterns used to analyze performance reports

This paper contains a framework within which to describe how the amount of information in a performance report impacts on a manager's report analysis process. Generally, the data collected from experienced managers as they analyzed performance reports were not completely supportive of the framework based hypotheses. An explanation is presented to account for the discrepancies. Future research could use this explanation to describe further how managers analyze performance reports.     

Does short sales deregulation affect qualitative information disclosure?

We examine the impact of short sales deregulation on firms' disclosure of non-financial qualitative information. Our simple analytical model predicts that, after short sales deregulation and when the cost of disclosing proprietary qualitative information to the firm and its executives is sufficiently high, shortable firms respond by disclosing less proprietary and more non-proprietary qualitative information than non-shortable firms. Using a textual analysis of qualitative information about the supply chain, available in the management disclosure and analysis sections of the annual reports of a sample of Chinese firms, and applying a staggered difference-in-differences research design, we find evidence consistent with the model's prediction.     

研发文本信息:真的多说无益吗?——基于分析师预测的文本分析

本文采用文本分析技术提取年报研发文本信息,并从分析师预测角度解读其信息价值.研究发现:(1)R&D文本信息显著降低了分析师预测偏差及分歧度,而且信息透明度发挥了中介效应.(2)R&D文本信息披露越多,未来三年的创新产出以及政府研发补助越多,说明企业倾向于选择性披露有利于企业的研发信息.(3)企业披露动机越强,分析师获取私有信息的能力越弱,R&D文本信息对分析师预测偏差和分歧度的降低作用越显著.以上结论说明我国年报R&D文本信息具有一定信息含量,对分析师来讲“多说有益”.本文从研发文本视角以及分析师预测角度验证了我国年报文本信息的积极效应,有助于丰富R&D以及分析师预测文献.     

Market reaction to quantitative and qualitative order backlog disclosures

Analysts and practitioners have long sought information on order backlog (OB) as indicators of future sales, and in turn, of future earnings and stock returns. OB disclosures, though mandatory for annual reports, are voluntarily included in some quarterly reports and are sometimes presented only in textual narration. Given that the required annual OB data may be partially preempted by voluntary quarterly disclosures, we test whether quarterly OB disclosures are used by market participants, especially the qualitative OB disclosures, which were not tested before. We show that OB growth is helpful in forecasting future sales and thus assign a positive tone to qualitative OB disclosures that indicate OB growth. Both quarterly quantitative OB increases and positive qualitative tone are associated with immediate and drift returns, after controlling for other disclosures during the quarterly earnings announcements and variables that affect voluntary disclosure. Our results indicate that regulators may need to consider requiring OB disclosures in quarterly intervals when OB is sufficiently material.     

Impact of internal control quality on the information content of social responsibility reports: A study based on text similarity—Evidence from China

To date, the impact of internal control on text information social responsibility reports has been rarely explored. This study investigates the correlation between internal control and information content of social responsibility reports based on the similarity of text information of reports by listed enterprises in 2006–2017. Overall, high-quality internal control results in low text similarity of social responsibility and high information content of social responsibility reports. The internal control quality could enhance the information content of social responsibility reports by decreasing the enterprise’s agency costs. For large-scale corporations, non-state enterprises and those with intensive public attention, effective internal control can enhance the information content of social responsibility reports. The agency cost can exert a partial mediating effect for large-scale corporations and enterprises with intensive public attention and complete mediating effect for non-state enterprises. Moreover, this study extends the scope of internal control to the quantitative dimension of text information and has crucial implications for listed enterprises, investors, and governments regarding internal control and disclosure of social responsibility information.     

UNDERSTAND THE GLOBAL ECONOMIC CRISIS: A TEXT SUMMARIZATION APPROACH

Economic crises are significant threats to macroeconomic stability. They can incur large costs and bring devastating effects on economies, with the effects often spilling over into other economies. Since 2007 we have witnessed the most severe and widely spread economic crisis since the Great Depression of the 1930s. In the meantime, a huge amount of ongoing media coverage, reporting, analysis and debate concerning the global economic crisis has been generated. In this study we explore the possibilities of applying text summarization tools to learn from text documents the various discussions surrounding the global economic crisis. Included in our analysis are blog posts and articles of highly influential economists, as well as official speeches and publications of government organizations. The ICSI‐ILP extractive summarizer is applied in a large number of experiments, and the summary outputs are manually examined and evaluated. The results provide us with insights into the potential and limitations of state‐of‐the‐art summarization systems when used to help us quickly learn and digest large amounts of textual information. The results also suggest different ways to break the limitations of text summarization technology. Copyright © 2013 John Wiley & Sons, Ltd.     

Integrating XBRL data with textual information in Chinese: A semantic web approach

Due to formatting differences, the difficulties of processing the textual disclosures and integrating them with quantitative financial data are well documented in the literature. Using a design science methodology, this paper describes a method that automatically extracts relevant textual data from annual reports published in Chinese. These extracted words are then mapped to a knowledge framework we proposed. This paper shows that it is technologically feasible to reorganize the MD&A contents into any given knowledge structure to improve the search capability, readability, and cohesiveness of the MD&A contents. Finally, we demonstrate a prototype system that uses semantic web technology to achieve information integration that presents XBRL formatted accounting data with relevant textual disclosures together to assist user decision making.     

A LONGITUDINAL STUDY OF CHANGE IN THE ENGLISH NATIONAL HEALTH SERVICE

This paper reports the findings of a longitudinal comparative case study of three National Health Service (NHS) hospital Trusts in England, investigating the perceptions of clinical, managerial and accounting professionals towards changing cost accounting and performance measurement practices. It incorporates both qualitative and quantitative data analysis, and is based on a contextualist understanding of change management, utilising the content‐process‐context approach (Pettigrew and Lapsley, 1994) to investigate the influence of receptive versus non‐receptive contexts on change. The analysis reveals limited success in improving performance measurement practices (the content of change) in Trusts. Nevertheless the specific context within which change was operationalised was found to be very important, with central mangers playing a key role in influencing change. The process of change indicated slow shifts in clinical‐accountant‐managerial relations, partly driven by changes in financial flows within the organisations.     

Annual report readability and stock liquidity

We examine the effect of annual report textual complexity on firms’ stock liquidity. Using techniques from computational linguistics, we predict and find that less readable filings are associated with lower stock liquidity. Our study provides evidence that difficult‐to‐read annual reports hinder investors’ ability to process and analyze information contained in corporate annual reports, reducing thereby their willingness to trade which decreases stock liquidity. Our findings are robust to a battery of sensitivity tests, including endogeneity, use of alternative estimation techniques, and use of alternative liquidity and readability proxies.     

Integrating Sustainability Into Capital Markets: Bloomberg LP And ESG's Quantitative Legitimacy

The rise in prominence of environmental, social, and governance data has been driven in large part by a growing interest among investors who seek to gain an edge through the incorporation of such data in their investment decision‐making. There are, however, several significant obstacles to the integration of ESG data into mainstream investing analysis. Perhaps most important, while finance today is a fundamentally quantitative discipline, ESG is often qualitative. Moreover, the ESG data that is available is incomplete and inconsistent, due largely to a reliance on voluntary reporting by individual companies. In short, ESG has not yet earned its quantitative legitimacy in the eyes of the investor community. Nevertheless, recent work in the area of stranded asset values has provided Bloomberg LP, a leading provider of financial data and analytics, an opportunity to “bridge theory and practice” by translating the stranded assets framework into a first‐cut valuation tool designed for mainstream financial analysts. The tool offers a quantitative introduction to an ESG issue that the authors believe will eventually become an important focus of many investment decision‐makers' analysis. While the tool continues to evolve in analytical sophistication, the authors “preview” it here in its early form as one step towards Bloomberg's broader vision of “sustainable finance,” and the company's role in supporting the quantitative maturation of ESG through the twin engines of standardization and disclosure.     

地方政府财政治理特征研究:来自中国政府工作报告文本识别的经验证据

地方政府财政治理是国家治理的微观基础和重要组成部分,但是学界依然缺乏针对其特征的定量识别分析。本研究基于1997—2017年国务院-省-市政府工作报告文本,借助计算机文本分析技术,定量识别了中国地方政府财政治理特征及其动态趋势。研究结果显示:(1)治理目标上,地方政府强调发展目标,在公平目标上的关注有提升空间;治理方式上,地方政府强调要处理好政府与市场的关系。(2)治理方向上,“五位一体”词频的变化反映了地方政府财政治理顺应时代需求。产业结构上,第一产业仍然是关注重点,第二产业词频略有下降而第三产业词频有上升趋势。(3)市政府与国务院、省政府工作报告均有较高的文本相似度,表明地方政府积极贯彻上级政府的政策安排。这一研究对于廓清地方政府财政治理状况提供了一种全新视角,并对进一步提高地方财政治理能力具有借鉴意义。     

Predicting financial distress of Chinese listed companies using machine learning: To what extent does textual disclosure matter?

Using machine learning to predict the financial distress of Chinese listed companies, this study shows that the incremental value of textual disclosure in financial distress prediction diminishes in the presence of detailed financial data. Detailed financial data itself has the capacity to accurately predict financial distress, and its prediction performance is not improved when combined with predictors extracted from textual disclosure. The model using combined predictors attaches more importance to financial-data-based predictors than textual-data-based ones. Our results provide evidence about the overstated value of textual disclosure and the understated information value of detailed financial data in financial distress prediction.