آیا خرده XBRL باعث افزایش کارآیی اطلاعاتی می شوند؟ شواهد اولیه از دریفت اعلان درآمد

In 2009, the Securities Exchange Commission (SEC) mandated public firms to file their financial statements using eXtensible Business Reporting Language (XBRL). The SEC's main motive behind this mandate is that XBRL filings would enhance the informational efficiency in the stock markets by making financial data easier to use and analyze for a broad range of investors. Using a sample from the first wave of mandated XBRL filers, we find a decline in post earnings announcement drift for the good news portfolio in the post-XBRL adoption period. Instead of a drift associated with underreaction, we find that markets overreact to negative earnings surprises for the bad news portfolio during our observation period, which coincides with the financial crisis. We detect limited evidence that XBRL adoption mitigates overreaction, which is another form of market inefficiency. We also find limited evidence that XBRL particularly benefits small investors.

On January 30, 2009, the SEC released a final rule requiring public companies to provide their financial statements to the Commission and on their corporate websites in interactive data format using the eXtensible Business Reporting Language (henceforth, XBRL). XBRL is a new Internet based document language for business and financial reporting purposes that will soon replace the Hyper Text Markup Language (henceforth, HTML) that has been broadly used in EDGAR filings of 10K or 10Q for almost 20 years. The main advantage of XBRL data over HTML data is that the former is computer readable. The SEC and proponents of XBRL claim that XBRL will eliminate costly manual collection, facilitate processing of financial information, and make the information easier to analyze by investors (AICPA (American Institute of CPAs), 2012, April 25, Efendi et al., 2011a, Efendi et al., 2011b and Gray and Miller, 2009). A working group of the American Accounting Association's Information System and Artificial Intelligence/Emerging Technology Sections examined the prospects of XBRL and concluded that XBRL would be vital to market democratization (Debreceny et al., 2005). Mandatory XBRL reporting would benefit a broad range of investors because it provides everyone an access to computer readable financial information in a timely manner at no cost. These proposed benefits suggest that XBRL would enhance the usefulness of financial data and thus improve informational efficiency in the capital markets. This study seeks to answer the research question: Do XBRL reports increase the information efficiency in the capital markets? Specifically, we measure the degree of informational efficiency in the capital markets with the tendency of post-earnings announcement drift (henceforth, PEAD). PEAD is described as a market anomaly that stock return drifts in a positive (negative) direction subsequent to a positive (negative) earnings surprise. Earnings surprises arise when the reported earnings are above or below benchmarks such as analyst forecast amount or previous period amount. PEAD has been one of the longest surviving market anomalies since it was first documented by Ball and Brown (1968).

This study investigates use of XBRL reporting and its potential impact on the capital markets, specifically, regarding informational efficiency. XBRL has been touted as a way to eliminate costly manual collection, aid processing of financial information, and facilitate analysis of information by investors. To proxy for information inefficiency, we use post earnings announcement drift (PEAD), which is one of the longest surviving market anomalies. The research question was stated as: Do XBRL reports increase the information efficiency in the capital markets? Using a sample of 7619 firm-quarter observations from 474 firms, we find that PEAD declines in the post-XBRL period for good news portfolio. The result is less certain for the bad news portfolio. Instead of market underreaction to earnings announcement, we find market overreact to bad news during our sample period which overlaps with the financial crises. We document slight evidence that adoption of XBRL mitigates overreaction to bad earnings news, which is another form of informational inefficiency. However, the latter results should be interpreted cautiously pending further analysis. Future studies, for example, can examine other cases of bad news events such as a stock market crash, dividend omission announcement, going concern audit disclosures, etc. (Harris and Spivey, 1990, Liang et al., 2011 and Taffler et al., 2004). Overall, our results are consistent with the notion that XBRL improves information efficiency in the capital markets.