هزینه ی اثرات سرمایه و تغییرات در انتظارات رشد در سراسر لیست متقابل آمریکا
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|11678||2009||27 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Financial Economics, Volume 93, Issue 3, September 2009, Pages 428–454
This paper examines whether cross-listing in the U.S. reduces firms’ costs of capital. We estimate cost of capital effects implied by market prices and analyst forecasts, which accounts for changes in growth expectations around cross-listings. Firms with cross-listings on U.S. exchanges experience a decrease in their cost of capital between 70 and 120 basis points. These effects are sustained and exist after the Sarbanes-Oxley Act. We find smaller reductions for cross-listings in the over-the-counter market and for exchange-listings from countries with stronger legal institutions. For exchange-traded cross-listings, the cost of capital reduction accounts for over half of the increase in firm value, whereas for other types of cross-listings the valuation effects are primarily attributable to contemporaneous revisions in growth expectations.
There is mounting evidence that countries’ institutional frameworks play an important role for access to finance and equity valuations (e.g., La Porta et al., 1997 and La Porta et al., 2002). In light of this evidence, cross-listing in the U.S. has been suggested as a way for firms from countries with poor institutions to overcome these shortcomings (Coffee, 1999; Stulz, 1999). Consistent with this notion, several studies show that cross-listings have significant effects on firms’ market values, using either event study returns (e.g., Foerster and Karolyi, 1999; Miller, 1999; Lee, 2004) or comparisons with firms that are not cross-listed (e.g., Doidge, 2004; Doidge et al., 2004 and Doidge et al., 2009). This evidence suggests that U.S. cross-listings offer substantial benefits. However, the sources of these benefits are not yet well understood (e.g., Leuz, 2003; Doidge, Karolyi, and Stulz, 2004). One important question is whether and to what extent cross-listing in the U.S. affects firms’ costs of capital. The bonding argument suggests that a U.S. cross-listing strengthens the protection of outside investors (e.g., Coffee, 1999; Stulz, 1999), which in turn makes it easier for the firm to raise external finance (e.g., Reese and Weisbach, 2002; Benos and Weisbach, 2004; Doidge, Karolyi, and Stulz, 2004). Moreover, listings on Nasdaq, NYSE, or Amex require foreign firms to comply with U.S. Securities and Exchange Commission (SEC) disclosure rules, which typically imply a substantial increase in disclosure and could manifest in a lower cost of capital (e.g., Verrecchia, 2001; Lambert, Leuz, and Verrecchia, 2007). In addition, cross-listing can improve investor recognition and enlarge a firm's investor base, increase liquidity, and overcome market segmentation, all of which could reduce the cost of capital (e.g., Merton, 1987; Karolyi, 1998; Foerster and Karolyi, 1999; Karolyi and Stulz, 2003). A potential concern about the documented valuation effects of U.S. cross-listings is that they merely reflect concurrent changes in firms’ growth opportunities that do not stem from cross-listing per se. That is, firms can seek cross-listings when they experience an expansion in their growth opportunities, but the decision is unrelated to bonding and the growth expansion does not reflect a reduction in the cost of capital due to cross-listing. Moreover, Foerster and Karolyi, 1999 and Foerster and Karolyi, 2000, Miller (1999), and Sarkissian and Schill (2009) provide evidence of return underperformance after cross-listing in the U.S., which raises the question of whether the documented valuation benefits are in fact sustained in the long run. Similarly, the debate about delistings from U.S. exchanges and the costs of the Sarbanes-Oxley Act (SOX) questions the existence of sizeable cross-listing benefits, such as a reduction in the cost of capital (Hostak, Karaoglu, Lys, and Yang, 2007; Zingales, 2007). Thus, it is still an open and topical question whether U.S. cross-listings persistently reduce the cost of capital. To shed light on these issues and the mechanism by which cross-listings affect firms’ valuations, we analyze ex ante estimates of firms’ costs of equity capital implied by market prices and analyst forecasts. This approach explicitly accounts for changes in the market's growth expectations around cross-listings. It also allows us to gauge the magnitude of both cash flow (or growth) effects and cost of capital effects on firms’ valuations. Our analysis is based on a large panel of more than 40,000 firm-year observations from 45 countries over the period from 1990 to 2005. We collect a comprehensive sample of 1,097 U.S. cross-listings and classify them into exchange listings, over-the-counter (OTC) listings, and private placements, accounting for the different regulatory consequences the firms face. For an exchange listing, firms have to register with the SEC and file Form 20-F, which requires extensive disclosures and a reconciliation of foreign financial statements to U.S. generally accepted accounting principles (GAAP). In addition, firms are subject to SEC oversight and bear the threat of U.S. securities litigation. Cross-listings in the OTC market do not require a 20-F filing, but a registration statement using Form F-6 and home-country disclosures to the SEC. They are also subject to Rule 10b-5 and the Foreign Corrupt Practices Act, under which most SEC enforcement actions as well as private class action suits are brought (Karpoff, Lee, and Martin, 2008). Private placements under Rule 144A do not require SEC registration or any additional (public) disclosures. Given these regulatory consequences, we hypothesize that, if cross-listings reduce firms’ costs of capital, the effects are strongest for exchange listings, and it is not clear that private placements should experience any reduction. Consistent with this hypothesis, we find strong evidence that cross-listings on U.S. exchanges (Amex, Nasdaq, and NYSE) significantly reduce the cost of equity capital and that the effects are larger than for the other types of cross-listings. We obtain these results from cross-sectional regressions including firm-fixed effects as well as from difference-in-differences analyses of changes in the cost of capital, mitigating concerns about omitted variables, and selection on unobservable characteristics. Most regressions suggest an average reduction in the cost of capital between 70 and 120 basis points, which is economically significant, but not too large to be implausible. We also find evidence that cross-listings in the OTC markets reduce the cost of capital. The estimated effect is smaller—on average, between 30 and 70 basis points—and not as robust as the effects for exchange listings. U.S. private placements exhibit insignificant changes and, in some of our analyses, an increase in the cost of capital. This result is consistent with the findings in Miller (1999) and Doidge et al., 2004 and Doidge et al., 2009 as they also show opposite or insignificant valuation effects for private placements. One possible explanation for the elevated cost of capital is that private placements entail private communication with a small group of institutional investors, which could exacerbate information asymmetries among traders. The rank order of the cost of capital effects (from exchange listings to private placements) suggests that the regulatory consequences of U.S. cross-listings play an important role, which is consistent with the bonding hypothesis. Further corroborating this notion, we find that the reduction in the cost of capital for exchange listings is larger for firms from countries with weaker disclosure regulation and weaker protection against self-dealing by corporate insiders. We show that the cost of capital effects are sustained for many years after the cross-listing and that they are still present after the passage of SOX. In contrast, we do not find significant cost of capital effects for cross-listings on the London Stock Exchange. Both of these findings are consistent with recent evidence in Doidge, Karolyi, and Stulz (2009). We conduct extensive robustness checks to validate our findings. We use four different implied cost of capital models and obtain very similar results for each of them as well as aggregating (and weighting) the estimates from the four models. We also gauge the sensitivity of our findings with respect to key model assumptions, in particular those about long-run growth. One potential concern is that cross-listed firms have different long-run growth expectations than non-cross-listed firms, even prior to the cross-listing. To address this issue, we implement our models with long-run growth estimates that vary by firm, cross-listing status, country, and/or year, and obtain very similar results. Further, we control for differences in analyst forecast bias across firms and countries, and check that the results are not unduly affected by missing or negative earnings forecasts. Lastly, we explore the sources of the cost of capital effects and find little evidence that the results are driven by increases in the shareholder base or market liquidity after cross-listing, strengthening the conclusion that bonding plays an important role for our results.1 Our final set of analyses exploits the fact that financial analysts provide explicit forecasts for firm growth. Using these forecasts, we decompose the realized three-year return leading up to the cross-listing into two components, one due to changes in the cost of capital and one due to changes in expected future cash flows. Using this approach, we show substantial valuation effects from revisions in growth expectations for all three cross-listing types. However, for cross-listings on U.S. exchanges, the reduction in cost of capital accounts for more than half of the increase in value around cross-listings, whereas for the other types of cross-listings the valuation effects are primarily, if not solely, attributable to an expansion in firms’ growth opportunities. This study makes several contributions. First, we use a novel approach to provide evidence that cross-listings on U.S. exchanges are associated with a statistically significant, yet economically plausible reduction in the cost of capital. Related studies produce estimates based on realized stock returns (or dividend yields) that are too large to be attributable solely to a reduction in the cost of capital. For instance, Errunza and Miller (2000) report an average effect of over 1,000 basis points, and Sarkissian and Schill (2009) estimate an effect of 800 basis points. Both of these findings are difficult to reconcile with the valuation effects around cross-listings and probably reflect the difficulty of estimating cost of capital effects around cross-listings from realized stock returns or dividend yields. To compute expected returns from realized returns, one needs fairly long time-series, yet such data are not available for many cross-listed stocks. Moreover, as cross-listings are major corporate events, it is difficult to obtain equilibrium estimates for expected returns, especially considering that cross-listings change firms’ exposures to the global market portfolio (Foerster and Karolyi, 1999). Our approach shows a way to overcome these difficulties. Second, we provide evidence that cross-listings are associated with significant valuation effects stemming from revisions in cash flow (or growth) expectations. A concern is that analyses using market values, Tobin's q, or stock returns pick up these effects, even when they are not related to cross-listing per se, thereby potentially overstating the benefits of U.S. cross-listings. However, shocks to firms’ growth opportunities that are concurrent but not incidental to the cross-listing are unlikely to be persistent, whereas changes in growth or the cost of capital that are related to the cross-listing should be sustained. Thus, by documenting a sustained reduction in the cost of capital, we corroborate earlier findings based on firms’ valuations. Moreover, our evidence on large and sustained benefits from U.S. exchange listings contributes to the recent debate about post-SOX regulatory costs and its consequences for the competitiveness of U.S. capital markets ( Zingales, 2007; Piotroski and Srinivasan, 2008; Doidge, Karolyi, and Stulz, 2009). Third, our analysis is based on one of the largest panels of U.S. cross-listings. Prior studies are generally based on smaller samples that are constructed as of one point in time.2 The panel approach not only mitigates survivorship bias, but also facilitates fixed-effects and difference-in-differences estimation, thereby mitigating concerns about unobserved heterogeneity and (time-invariant) selection bias. Finally, we provide evidence that the cost of capital effects differ systematically across regulatory consequences of different types of American Depositary Receipts (ADRs) and across firms from home countries with different institutional frameworks. This evidence corroborates earlier evidence on the bonding hypothesis (e.g., Reese and Weisbach, 2002; Doidge, Karolyi, and Stulz, 2004), which several recent studies have questioned (e.g., Siegel, 2005; Gozzi, Levine, and Schmukler, 2008). The remainder of the paper is organized as follows. Section 2 reviews the literature and develops our hypotheses. In Section 3, we describe the sample and the construction of the dependent and independent variables. Section 4 presents the main results including robustness checks. In Section 5, we show that the cost of capital effects differ depending on firms’ home-country legal institutions and analyze the relative magnitude of cash flow and cost of capital effects. Section 6 concludes. In Appendix A, we delineate the estimation procedure of the implied cost of capital measures.
نتیجه گیری انگلیسی
In this paper, we examine the cost of capital effects of U.S. cross-listings for a large panel of ADR firms from 45 countries. Prior research shows significant valuation effects of cross-listings in the U.S. However, it provides little evidence on the mechanism(s) by which cross-listing affects firm value. One important question is whether and to what extent these benefits stem from a reduction in firms’ costs of capital, as the bonding hypothesis or disclosure theory would suggest. Alternatively, it is possible that the valuation effects merely reflect that firms cross-list when they experience an expansion in their growth opportunities, even though the latter is unrelated to the cross-listing per se. Thus, understanding the sources of the valuation benefits is important. This issue is, for instance, at the heart of the debate about the Sarbanes-Oxley Act and its potential consequences on foreign firms’ cross-listing or delisting decisions. We use cost of capital estimates implied by current market prices and analyst forecasts, rather than estimates based on realized returns or dividend yields, as this approach makes an explicit attempt to account forrevisions in growth expectations around cross-listings. We find strong evidence that cross-listings on U.S. exchanges are associated with a significant decrease in firms’ costs of equity capital, after controlling for traditional proxies for risk, analyst forecast bias, and firm-fixed effects. The magnitude and the ranking of the effects seem plausible with exchange listings experiencing a reduction in cost of capital between 70 and 120 basis points, followed by OTClistings with about 30 to 70 basis points. Firms that access U.S. markets via private placements exhibit no change or a small increase in cost of capital. These cost of capital effects are sustained over time, and we find no evidence that Sarbanes-Oxley has diminished the benefits of U.S. cross-listings. We do not find similar cost of capital effects for listings on the London Stock Exchange. When we investigate cross-sectional differences in the cost of capital effects, we find that firms from countries with weak disclosure regulation and weak protection of outside investors against self-dealing by corporate insiders benefit the most from cross-listing on U.S. exchanges. Overall, the ranking of the cost of capital effects across ADR types (from Level III to Level I) and the cross-sectional results lend support to the bonding hypothesis. Our evidence of sustained cost of capital effects also mitigates concerns that valuation effects documented by prior studies stem merely from growth shocks that are concurrent but unrelated to cross-listing per se. That said, there are significant valuation effects stem- ming from changes in (analysts’) growth expectations for all three cross-listing types, indicating that the change in firm value around U.S. cross-listings is a combination of both growth and cost of capital effects. For cross-listings on U.S. exchanges, the reduction in cost of capital accounts for more than half of the increase in value around cross- listings, whereas for the other ADR types the valuation effects around cross-listings are primarily attributable to an expansion in firms’ growth opportunities. Finally, several caveats are in order. First, cross-listings representmajorcorporateeventsmakingitdifficultto compute equilibrium cost of capital estimates. While our methodology should be well equipped to estimate the long- run consequences of cross-listings for firms’ costs of equity capital, it is possible that the events leading up to the cross- listing affect the calculation of our proxies. Second, while we interpret our evidence as indicating that cross-listings reduce the cost of capital, we cannot preclude that causality runs the other way (i.e., firms choose to cross-list after they experience a reduction in their cost of capital). In this case, however, we would not expect the cost of capital effects to differ across ADR types and countries in the documented way. There also remain concerns about self-selection, despite our efforts to control for the fact that cross-listed firms differ from non- cross-listed firms in observable and unobservable ways. Lastly, cost of capital and growth effects are likely to be intertwined in a non-trivial way. Thus, the results from our decomposition of the two effects should be interpreted cautiously.