How Foreign Firms Communicate with US Investors

The following post comes to us from Russell Lundholm, Rafael Rogo, and Jenny Li Zhang, all of the Accounting Division at the University of British Columbia.

Foreign companies that trade their equity in the US face serious obstacles. They must navigate a complex set of SEC disclosure requirements, while at the same time satisfying US investor expectations about the frequency and content of voluntary disclosures. Their home country may be far from the US, speak a different language, use different accounting rules, and offer different types of investor protection than the US, and each of these differences presents a friction that must be mitigated in order to attract US investors. Given these cultural, procedural, and linguistic differences, one might expect that the disclosures of foreign firms would be of lower quality than their US firm counter-parts. Nonetheless, in our paper, Restoring the Tower of Babel: How Foreign Firms Communicate with US Investors, forthcoming in The Accounting Review, we find that foreign firms traded in the US present more numerical data and write more readable text in the Management Discussion and Analysis (MD&A) section of their 10-K, and write more readable text in their earnings press releases, than comparable US firms. More importantly, we find that the readability of text and amount of numerical data in both the MD&A and earnings press releases increase with the foreign firm’s distance from the US. Finally, we find that within a country, firms with relatively more readable disclosures attract relatively more US institutional investment.

We study two sources of disclosures, the mandatory MD&A section of the 10-K and voluntary earnings press releases. The SEC has repeatedly expressed concern about the readability of mandatory filings. In several releases after the Securities and Exchange Act of 1934, the SEC has encouraged clearer communication (Firtel 1999). The Wheat Report in 1969 complained about needlessly complex language and called on companies to improve the clarity of their disclosures. In 1998 the SEC issued the Plain English disclosure guidelines for all prospectuses in registered public offerings by domestic and foreign issuers, and produced a Plain English handbook on how to write clear SEC documents. More recently, SEC chairman Christopher Cox suggested that the SEC may someday use a readability measure such as the FOG index to assess compliance with the Plain English rules (Cox 2007).

There is academic evidence to back up the regulatory concern about the clarity of MD&A disclosures. Li (2008) finds that firms with difficult-to-read MD&A appear to be covering up unfavorable information about their future earnings. Miller (2010) finds that less readable 10-K reports are associated with lower trading volume around the 10-K filing date. Lehavy, Li and Merkley (2011) find that the amount of time analysts spend preparing their reports is inversely related to the readability of the firm’s 10-K filing. And finally, You and Zhang (2009) find that the market underreacts to news in the 10-K when the text is less readable.

In contrast to mandatory MD&A disclosures, press releases are voluntary, and the form and content are largely unregulated. These disclosures are typically more visible than regulatory filings, and earnings press releases often produce large price changes (Beyer, Cohen, Lys and Walther 2010). We collect a large sample of earnings press releases to complement our MD&A sample. While an exhaustive measure of disclosure activity would be very difficult to construct, the combination of MD&A and earnings press releases covers a significant amount of the available written disclosure for a firm.

Along with studying the readability of textual disclosures, we study a relatively new communication measure in this paper, the “number of numbers.” The number of numbers is different from readability; it represents a communication that is universally understood and often has more precision than a textual equivalent. Numbers are more likely to be auditable ex ante and verifiable ex post.3 Numbers are concrete statements and therefore construal level theory predicts that they can serve to lessen the psychological distance US investors might feel toward foreign firms (Elliott, Rennekamp and White 2012). Finally, neither text alone nor numbers alone constitute effective communication about a firm’s performance; rather, the combination of words and numbers make financial communications meaningful. By studying both textual readability and the degree of quantification with numbers, we triangulate on the firm’s overall clarity in communication.

There is extensive evidence that US investors greatly underweight foreign equity in their investment portfolio, a phenomenon known as the “home bias puzzle” (French and Poterba 1991). Ahearne, Griever and Warnock (2004) report that non-US equity comprises 50 percent of the global equity portfolio, yet US investors hold only 12 percent of their assets in non-US equity. In fact, the further the foreign firm is from the US, the greater the home bias is against them (Portes, Rey and Oh 2001). To better understand what aspect of geographic distance is contributing to the home bias, we examine three alternative notions of distance: linguistic distance, measured by whether or not the foreign firm is from an English-speaking country; accounting distance, measured by the similarity between the accounting rules of the foreign country and the US; and investor protection distance, measured by the similarity between the foreign country’s legal system and the US system. While each of these measures has its weaknesses, we generally find that as these alternative notions of distance increase, the readability of text and use of numbers in foreign firms’ disclosures increase.

In the presence of the home bias, our first set of results suggest that the further the foreign firm is from the US, the harder it works to inform US investors about its business and mitigate the home bias. It does so by providing more readable disclosures and by quantifying its disclosures with more numbers. Our second set of tests examine whether these disclosure efforts are successful. In particular, we examine whether a foreign firm’s US institutional ownership varies with the readability or numerical content of its MD&A and earnings press release disclosures. After controlling for a number of previously documented determinants of institutional ownership (Bradshaw, Bushee and Miller 2004), including country fixed effects, we find that firms with relatively more readable disclosures in a country have relatively more institutional ownership. Of course, the home bias is still present in the data and so the firm actions we document only partially resolve the information asymmetry. But by showing that the foreign firms’ response varies with the same distance variables that predict the home bias, and that more readable disclosures attract more US institutional investors, our results strongly suggest that management perceives a bias that must be overcome, and one of their responses is to use more numerical data and produce more readable text in their MD&A and earnings press releases.

The full paper is available for download here.

Both comments and trackbacks are currently closed.