SEARCH

SEARCH BY CITATION

Keywords:

  • cross-listing;
  • operating performance;
  • Level II;
  • Level III

Abstract

  1. Top of page
  2. Abstract
  3. 1. INTRODUCTION
  4. 2. BACKGROUND, MOTIVATION, AND DEVELOPMENT OF THE HYPOTHESES
  5. 3. RESEARCH DESIGN
  6. 4. EMPIRICAL RESULTS
  7. 5. CONCLUSIONS
  8. REFERENCES

Abstract:  In this paper we examine the operating performance of non-US firms that enter major US stock exchanges using American Depositary Receipt (ADR) programs. Our dataset consists of 108 capital-raising and non-capital-raising firms from twenty four countries, cross-listed on major US stock exchanges during the period 1994–2004. We provide evidence that capital-raising cross-listed firms experience improvements in their operating performance after the listing, relative to a non-cross-listed matched sample of firms and relative to the pre-listing period, whereas non-capital-raising cross-listed firms out-perform a non-cross-listed matched sample of firms for both the pre-listing and the post-listing periods. These results suggest that the type of ADR program conveys information about changes in the post-listing operating performance. Moreover, both capital-raising and non-capital-raising cross-listed firms have positive abnormal returns due to the cross-listing and these abnormal returns are positively related with the post-listing abnormal changes in operating performance, suggesting that the market anticipates the post-listing abnormal changes in operating performance. Results are robust after adjusting for various firm and country risk characteristics.


1. INTRODUCTION

  1. Top of page
  2. Abstract
  3. 1. INTRODUCTION
  4. 2. BACKGROUND, MOTIVATION, AND DEVELOPMENT OF THE HYPOTHESES
  5. 3. RESEARCH DESIGN
  6. 4. EMPIRICAL RESULTS
  7. 5. CONCLUSIONS
  8. REFERENCES

The increasing interest on cross-listings around the world facilitates a long debate among academics and practitioners concerning the implications of the cross-listing for firms, shareholders and managers. Prior studies examine cross-listing implications with respect to firms' valuation, cost of capital, access to capital, information environment, firms' visibility and managerial incentives, among others.1 However, one important issue that has not yet been addressed in the cross-listing literature, is whether cross-listing affects firms' operating performance.

The purpose of this study is to examine the relation between cross-listing and operating performance. Doidge et al. (2004) suggest that cross-listed firms experience a revaluation of their growth opportunities that may be driven by an increase in the magnitude of firms' growth opportunities and a reduction in both firms' cost of capital and cash flows expropriated by managers. However, possible revaluation of firms' growth opportunities due to the cross-listing is expected to affect operating performance after the listing. This study links cross-listing with operating performance and in doing so examines the following research questions: (i) Does the post-listing operating performance of cross-listed firms increase relative to the performance of non-cross-listed matched sample of firms and relative to its pre-listing operating performance? (ii) Are the operating performance patterns of the capital-raising cross-listed firms different relative to the performance patterns of non-capital-raising cross-listed firms? (iii) Do changes in market expectations due to the cross-listing reflect post-listing changes in firms' operating performance?

To address the aforementioned research questions, we use a sample of 108 capital-raising and non-capital-raising firms from twenty four countries that were cross-listed on US stock exchanges during the period 1994–2004 and a country-industry-size matched sample of non-cross-listed firms. We provide evidence for economically and statistically significant operating performance improvements for capital-raising firms after the listing, relative to a non-cross-listed matched sample of firms and relative to the pre-listing period. Moreover, non-capital-raising cross-listed firms are found to out-perform a non-cross-listed matched sample for both the pre-listing and the post-listing periods. Further analysis that compares the abnormal operating performance around the cross-listing between capital-raising and non-capital-raising cross-listed firms indicates that after controlling for differences in size, growth, industry and country characteristics the matched-adjusted operating performance is higher for capital-raising firms relative to the non-capital-raising firms after the listing. Overall, these results suggest that the growth opportunities of the capital-raising cross-listed firms are enhanced more from a cross-listing in the US, relative to the growth opportunities of the non-capital-raising cross-listed firms.

We then examine whether matched-adjusted changes in operating performance after the listing relate with the twelve month pre-listing abnormal returns. A positive relation between these variables would suggest that the market recognises additional opportunities or re-values existing opportunities due to the cross-listing.2 Moreover, such a relation is expected to strengthen the belief that cross-listing effects leads the observed post-listing improvements in operating performance of the cross-listed firms. Results show that after controlling for firm and country risk characteristics, pre-listing abnormal returns are positively associated with post-listing matched-adjusted changes in operating performance for both capital-raising and non-capital-raising firms.

This paper makes three important contributions to the literature. First, we show significant differences in operating performance between capital-raising and non-capital-raising firms, suggesting that cross-listing effects may not affect in the same manner all exchange-listed firms. Second, we document that capital-raising through cross-listing conveys information about subsequent operating performance. Interestingly, prior studies provide evidence that other capital-raising events such as seasoned equity offerings and initial public offerings are followed by a significant decline in operating performance after the offering (Loughran and Ritter, 1997; and Jain and Kini, 1994). Consequently, capital-raising through cross-listing differs from both initial and secondary capital-offerings. Finally, we provide further evidence on where the revaluation effect of cross-listings may come from by considering operating performance.

The remainder of the paper is organised as follows: Section 2 reviews the related literature, motivates the paper and develops the research hypotheses. Section 3 provides details on the sample and data. Section 4 evaluates the empirical results. Concluding remarks are provided in Section 5.

2. BACKGROUND, MOTIVATION, AND DEVELOPMENT OF THE HYPOTHESES

  1. Top of page
  2. Abstract
  3. 1. INTRODUCTION
  4. 2. BACKGROUND, MOTIVATION, AND DEVELOPMENT OF THE HYPOTHESES
  5. 3. RESEARCH DESIGN
  6. 4. EMPIRICAL RESULTS
  7. 5. CONCLUSIONS
  8. REFERENCES

Doidge et al. (2004) provide evidence that the Tobin's q ratio of the cross-listed firms on major US stock exchanges exceeds by about 37% the q ratio of firms from the same country that do not list on US stock exchanges. Differences in valuation between cross-listed and non-cross-listed firms were mainly attributed to the fact that the growth opportunities of cross-listed firms are re-valued due to the cross-listing. There are three main explanations for the revaluation of growth opportunities due to the cross-listing: (i) the cross-listing helps to improve the magnitude of firms' growth opportunities, (ii) the cross-listing decreases firm's cost of capital and (iii) the cross-listing helps to reduce cash flow expropriation by managers.

As far as the relation between cross-listing and the magnitude of firms' growth opportunities, Pagano et al. (2002) suggest that firms cross-list to capitalise on product market reputation. For example, firms may find it easier to place their shares in foreign markets because local investors already trust them as consumers. Moreover, Pagano et al. (2002) suggest that cross-listings may serve as an advertisement for the firm's products and thereby increase sales by raising consumer demand and by improving relationships with suppliers and employees. Consistent with these arguments, Baker et al. (2002) provide evidence that a cross-listing in the US is associated with greater visibility since they found significant increases in the number of media ‘hits’. Moreover, Lee (2003) shows that US cross-listing announcements are associated with negative abnormal returns for the competitors of the announcing firm. The aforementioned empirical evidence is consistent with the notion that cross-listing improves both the magnitude of growth opportunities and the ability of the firm to exploit its growth opportunities.

As far as the relation between cross-listing and the cost of capital, prior studies provide evidence that cross-listing in the US leads to greater analysts' coverage, more accurate earnings predictions and improved corporate governance relative to firms that do not cross-list, suggesting that cross-listed firms experience an improvement in their information environment (Lang et al., 2003; Huijgen and Lubberink, 2005; and Charitou et al., 2007). Easley and O'Hara (2004) suggest that reduced information asymmetry between management and market participants, lowers the firm's cost of capital. Empirically, this view is supported by Hail and Leuz (2004) who provide evidence that cross-listing reduces cost of capital, even after controlling for traditional risk factors and country effects.3

Finally, as far as the relation between cross-listing and managerial expropriation, Doidge (2004) provides evidence that voting premium, which is a measure of private control benefits, is lower for cross-listed firms and that this difference is larger for firms from countries with poor protection of outside investors. This result may suggest that managers expropriate a lower fraction of cash flows after the listing.

Based on the aforementioned discussion, it seems possible that US cross-listings affect the firm's value of growth opportunities. These growth opportunities, however, are expected to affect positively the post-listing operating performance. These arguments lead us to the following hypothesis:

H1: Cross-listed firms experience improvements in their operating performance after the cross-listing.

Cross-listing per se is an investment opportunity for both capital-raising and non-capital-raising firms. However, to the extent that the need for external financing for some of the cross-listed firms relates with existing growth opportunities, it is likely that capital-raising firms will experience greater operating performance improvement compared to the non-capital-raising firms. Cross-listing on US stock exchanges signal a firm's commitment to protect minority investors, which allows firms to raise capital at a lower cost in their home markets to exploit existing growth opportunities (Coffee, 2002; La Porta et al., 2000 and 2002; and Lombardo and Pagano, 2000a and 2000b). Therefore, capital-raising through cross-listing may be seen as a signal for higher growth opportunities compared to the non-capital-raising firms' growth opportunities. The higher the growth opportunities prior to the listing are, the higher the expected operating performance improvements after the listing. These arguments lead us to the following hypothesis:

H2: Cross-listed firms' operating performance improvements after the listing are more pronounced for capital-raising firms relative to the non-capital-raising firms.

Prior studies provide evidence for positive abnormal returns around cross-listing on US stock exchanges. Researchers examined both long-run abnormal returns and the market reaction around the cross-listing announcement. Regarding long-run abnormal returns, Foerster and Karolyi (1999) found that cross-listed firms earn cumulative abnormal returns of 19% during the year before listing, and an additional 1.2% during the listing week, but incur a loss of 14% during the year following the listing. As far as the market reaction around the cross-listing announcement is concerned, Miller (1999) found positive abnormal returns. In both studies, the market reaction was shown to be more positive for capital-raising firms compared to the non-capital-raising cross-listed firms.4Coffee (2002) attributed partly the difference in abnormal return performance between capital-raising and non-capital-raising firms to the fact that capital-raising firms are voluntarily subjected to the strict liability provisions of Section 11 of the Securities Act of 1933. In principle, this liability provides added credibility to the information released by firms since firms face increased liability for any material misrepresentation or omission. Coffee (2002) suggests that this added credibility is responsible for the stronger positive market reaction for capital-raising firms.

In this study, we extend prior literature on the cross-listing return performance by examining the relation between cross-listing return performance and subsequent abnormal changes in operating performance. Specifically, we examine whether the market is aware of the signalling effect of the cross-listing. Given that in hypothesis 1 we expect cross-listed firms to experience improvements in their operating performance after the cross-listing, we suggest that market expectations due to the cross-listing incorporate this information. In that case, we expect a positive relation between the abnormal return performance due to cross-listing and the subsequent abnormal changes in operating performance. These arguments lead us to the following hypothesis:

H3: The higher the abnormal changes in operating performance after the listing are, the higher the abnormal returns due to the cross-listing.

3. RESEARCH DESIGN

  1. Top of page
  2. Abstract
  3. 1. INTRODUCTION
  4. 2. BACKGROUND, MOTIVATION, AND DEVELOPMENT OF THE HYPOTHESES
  5. 3. RESEARCH DESIGN
  6. 4. EMPIRICAL RESULTS
  7. 5. CONCLUSIONS
  8. REFERENCES

In this section we discuss (i) the depositary receipts market, (ii) the dataset and (iii) the control sample.

(i) The Depositary Receipts Market

Firms can list on US stock exchanges either by listing their shares through ADRs or by listing their shares directly on these exchanges. An ADR represents an ownership interest in a specified number or fraction of securities that have been deposited with a depositary. The deposited securities are typically equity securities of a foreign issuer, and the depositary is usually a US bank or trust company. ADRs were developed primarily to facilitate the transfer of ownership of foreign securities in the United States and the conversion of foreign currency dividends into US dollars, as an alternative to purchasing ordinary shares on foreign markets.

Currently, there are four types of ADR programs: Level I, Level II, Level III and the 144A Rule. The least costly way for a company to cross-list its shares, is to establish a Level I DR program. Level I DR programs trade in the US over the counter (OTC) market and firms do not have to comply with the US GAAP or full SEC disclosure. Level II DR programs are traded on the NASDAQ or NYSE, and are used by companies seeking greater liquidity and investor recognition. However, Level II DR programs entail greater indirect and direct costs since they must be partially reconciled to US GAAP and report quarterly, using Form 20-F. Level III DR programs trade on the NASDAQ or the NYSE and firms raise new equity capital. Level III DR programs are required to complete Form 20-F and fully reconcile to US GAAP. Finally, firms that adopt Rule 144A DRs raise equity capital via private placement without being obliged to meet the reporting and disclosure requirement of a US public offering.

(ii) Dataset

The dataset consists of non-US firms cross-listed on major US stock exchanges using Level II or Level III DR programs during the period 1994–2004. The cross-listed firms were obtained from the website of the Bank of New York and from the Lexis Nexis database. Financial data were retrieved from the Global Vantage database.5 Daily stock returns were obtained from the Datastream database. Listing and announcement dates were identified from the stock markets' webpages and the Lexis Nexis database.

The initial sample of Level II and Level III cross-listed firms, obtained from the Bank of New York for the period 1994–2004, consists of 366 firms. From this sample we excluded various firms for the following reasons: (a) twenty firms that cross-list in the US using preferred stocks or warrants, (b) forty five firms that belong in the financial sector since these firms operate in a regulated environment and their characteristics differ substantially from those of non-regulated firms, (c) nine firms with multiple cross-listings; in order to avoid using overlapping data and to reduce dependence for the statistical tests among firms we include only the earliest cross-listing, (d) sixty four firms without financial data available in the Global Vantage database, (e) sixty eight firms with missing operating performance measures at least one year prior to the listing; operating performance measures are needed to facilitate the comparison between pre- and post-listing periods, (f) thirty seven firms with missing returns during the cross-listing; returns are needed to examine whether cross-listing benefits are partly attributed to market expectations on future abnormal profitability,6 (g) eighteen firms that cannot be matched at least on the one-SIC industry code. These criteria/restrictions lead us to a sample of 105 firms. We then employed the Lexis Nexis database as a complementary source to examine the accuracy of the information gathered from the Bank of New York. In this procedure we identified three cross-listed firms that were not initially included in the aforementioned sample but they meet all the aforementioned criteria. We added these firms to our initial sample resulting in a final sample of 108 firms. From this sample 79 firms cross-list on US stock exchanges using Level II DR program and 29 firms cross-list on US stock exchanges using Level III DR program. For these firms, we define the fiscal year that cross-listing took place as the event year. We then collected data for the three-year period around the event year.

In Table 1 we present descriptive statistics for the final sample of cross-listed firms. Results are presented by country, year, and by industry. Panel A of Table 1 presents descriptive statistics by country. Our sample consists of cross-listed firms from twenty four countries. Moreover, we do not generally observe temporal concentration of the sample firms in any country, having the largest occurrence of cross-listings from the UK (21 firms or 19.44% of the sample), Japan (11 firms or 10.19% of the sample) and France (10 firms or 9.26% of the sample). The diversity of country of origin of cross-listed firms suggests that country is not a factor that might affect our results. Panel B of Table 1 shows a classification of the sample based on the year of listing. Most firms cross-list in the years 2000, 2001 and 2002 (17 firm or 15.74% of the sample, 21 firms or 19.44% of the sample and 20 firms or 18.52% of the sample, respectively). Finally, in Panel C of Table 1 we present descriptive statistics by industry, with the cross-listing activity taking place largely within the communication industry (18 firms or 16.67% of the sample) and the chemicals industry (17 firms or 15.74% of the sample).

Table 1.  Descriptive Statistics for Exchange-Listed ADRs Firms on US Stock Exchanges, 1994–2004
Panel A: Cross-Listed Firms by Country of Incorporation
CountryLevel II Frequency%Level III Frequency%Total Frequency %
ARGENTINA11.2700.001 0.93
AUSTRALIA11.2700.001 0.93
BELGIUM11.2700.001 0.93
BRASIL78.8600.007 6.48
CHILE33.8000.003 2.78
DENMARK11.2700.001 0.93
FINLAND22.5313.453 2.78
FRANCE911.3913.4510 9.26
GERMANY78.8613.458 7.41
GREECE11.2700.001 0.93
HONK KONG33.8000.003 2.78
INDIA22.53310.345 4.63
ITALY11.2713.452 1.85
JAPAN45.06724.141110.19
KOREA00.00310.343 2.78
MEXICO67.5913.457 6.48
NETHERLANDS33.8013.454 3.70
NEW ZELAND11.2700.001 0.93
NORWAY11.2700.001 0.93
SOUTH AFRICA33.8013.454 3.70
SPAIN11.2700.001 0.93
SWITZERLAND45.0613.455 4.63
TAIWAN00.00413.794 3.70
UK1721.52413.792119.44
Total7910029100108100
Panel B: Cross-Listed Firms by Year of Listing
YearLevel II Frequency%Level III Frequency%Total Frequency%
199422.5326.904 3.70
199511.27310.344 3.70
199611.2700.001 0.93
199767.5913.457 6.48
19981012.6600.0010 9.26
1999911.39413.791312.04
20001316.46413.791715.74
20011620.25517.242119.44
20021316.46724.142018.52
200356.3313.456 5.56
200433.8026.905 4.63
Total7910029100108100
Panel C: Cross-Listed Firms by Industry
Code/IndustryLevel II Frequency%Level III Frequency%Total Frequency%
  1. Notes: This table presents descriptive statistics for 108 exchange-listed firms in the US obtained from the Bank of New York and the Lexis Nexis database. Panel A presents information for the country of incorporation and the type of listing. Panel B presents information for the year of listing. Panel C presents information by industry (2-digit SIC codes) for the cross-listed firms.

10/Metal mining33.8026.9054.63
13/Oil and gas extraction22.5300.0021.85
20/Food products78.8600.0076.48
21/Tobacco manufacture11.2700.0010.93
23/Apparel and other finished products11.2700.0010.93
26/Paper products11.2700.0010.93
27/Printing and publishing22.5300.0021.85
28/Chemicals1316.46413.791715.74
29/Petroleum refining22.5300.0021.85
32/Store, clay, glass, concrete33.8000.0032.78
33/Primary metal11.2713.4521.85
35/Computer equipment22.5326.9043.70
36/Electrical equipment22.53724.1498.33
37/Automotives33.8013.4543.70
38/Meas. Inst. Photo11.2713.4521.85
44/Water transportation11.2713.4521.85
45/Scheduled air trans22.5300.0021.85
48/Communication1215.19620.691816.67
49/Electric and gas56.3300.0054.63
50/Durable goods11.2700.0010.93
51/Nondurable goods22.5300.0021.85
53/General merchandise stores11.2700.0010.93
54/Food stores22.5300.0021.85
58/Eating and drinking places11.2700.0010.93
59/Retail stores11.2700.0010.93
73/Business services56.33413.7998.33
99/Other22.5300.0021.85
Total7910029100108100

(iii) Control Sample

Consistent with prior literature, as a benchmark against which we compare changes in operating performance around the cross-listing we construct a control sample of firms as follows: for each firm cross-listed in a particular year (the year prior to the listing), we select all the non-cross-listed firms for the corresponding year that belong in the same home country, and have the same SIC industry code.7 Among those firms, we selected one matched firm that has the closest total assets with the cross-listed firms' total assets during the corresponding year. If we do not find firms in the same SIC industry code, we search for non-cross-listed firms that belong in the same three-SIC industry code, or alternatively, belong in the same two-SIC or one-SIC industry code. This procedure yields a total of 108 control firms. For this sample we collected data for the three-year period around the event year.8

4. EMPIRICAL RESULTS

  1. Top of page
  2. Abstract
  3. 1. INTRODUCTION
  4. 2. BACKGROUND, MOTIVATION, AND DEVELOPMENT OF THE HYPOTHESES
  5. 3. RESEARCH DESIGN
  6. 4. EMPIRICAL RESULTS
  7. 5. CONCLUSIONS
  8. REFERENCES

In this section, we examine the operating performance, levels and changes, of Level II and Level III firms around the cross-listing. Moreover, we investigate the return performance of the cross-listed firms around the listing and finally, we examine whether return performance relates with post listing abnormal operating performance.

(i) The Operating Performance of the Exchange-Listed Firms

(a) Levels of Operating Performance Measures

Similar to Loughran and Ritter (1997) and Papaioannou et al. (2003) we measure operating performance by focusing on operating return plus depreciation on assets/sales and on operating return on assets/sales. In addition we examine performance with respect to capital expenditures to assets. Furthermore, cross-listed firms' operating performance may be affected from changes in firms' efficiency or from changes in firms' sales. To shed more light on whether cross listing affects firms' efficiency or facilitates changes in sales we also report total sales and total operating cash flows.

In Table 2 we report yearly median and mean operating performance measures for the cross-listed firms, the non-cross-listed matched firms (control firms), as well as matched-adjusted operating performance measures defined as the difference between cross-listed and control firms' measures.9 Panel A of Table 2 reports results for Level II firms. Hypothesis 1 suggests that a positive trend in operating performance measures should be observed from the pre-listing period to the post-listing period. In contrast to hypothesis 1, no significant positive trend in operating performance is observed for Level II firms. Specifically, both cross-listed and control firms are profitable for the period under examination (year −3 to year +3). Significant differences between the cross-listed and control firm's operating performance measures emerge from year −1 to year +2. During that period the median matched-adjusted operating return plus depreciation on assets (on sales) ranges from 0.017 to 0.087 (0.014 to 0.117). Similarly, the median matched-adjusted operating return on assets ranges from 0.010 to 0.023. Most of these figures are significant at conventional levels suggesting that Level II firms are more profitable relative to the control sample during each of the years −1 to +2. In contrast, the median matched-adjusted operating returns on sales are not significant at conventional levels in each of the years 0, +1 and +2. Furthermore, both cross-listed and control firms have positive capital expenditures to assets during the period under examination. The median matched-adjusted capital expenditures on assets are positive and significant only for years −2, −1 and year +1. Finally, cross-listed firms have higher sales and operating cash flows in each of the years under examination. Untabulated results show no significant differences in asset turnover ratio, suggesting that cross-listed firms do not use their assets more efficiently compared to the control firms. Overall, the results suggest that Level II firms become more profitable relative to the control sample a year prior to the listing. These results leave open the question on whether cross-listing enhances Level II firms' operating performance or not.

Table 2.  Operating Performance Measures and Firm Characteristics per Year Around the Listing, 1994–2004
Panel A: Operating Performance and Firm Characteristics Around the Listing for Level II Firms
  Operating Return plus Depreciation on AssetsOperating Return plus Depreciation on SalesOperating Return on AssetsOperating Return on SalesCapital Expenditures on AssetsSalesOperating Cash Flows
Year −3Cross-Listed0.082***0.094***0.036***0.039***0.054***1980.004***269.497***
 0.085***0.0660.043***−0.0080.063***7250.849***926.897***
Control0.074***0.077***0.033***0.039***0.037***2373.477***181.130***
 0.056***0.1220.0130.0500.057***4959.372***570.712***
Difference0.009−0.0030.0000.0050.000184.691*84.852***
 0.029−0.0560.030−0.0580.0062291.477**356.185***
No of Firms   65  65   65  65   656565
Year −2Cross-Listed0.095***0.118***0.041***0.061***0.051***1937.985***273.868***
 0.080***0.0220.032**−0.0750.058***8288.550***1263.713***
Control0.077***0.091***0.035***0.047***0.039***2125.768***203.971***
 0.062***0.112*0.023*0.0480.048***5586.684***608.528***
Difference0.0140.0100.0060.0100.008*262.3285**105.513***
 0.018−0.0900.009−0.1230.0102701.866**655.185***
No of Firms   74  74   74  74   747474
Year −1Cross-Listed0.078***0.105***0.040***0.057***0.053***1960.668***239.973***
 0.074***0.130***0.032**0.0490.065***8186.588***1070.607***
Control0.076***0.086***0.029***0.036***0.042***2344.392***271.564***
 0.049***−0.0420.006−0.1190.053***5451.629***620.111***
Difference0.014**0.022**0.009**0.011**0.004**229.609**69.262***
 0.0250.172*0.0260.1680.0122734.959**450.496***
No of Firms   79  79   79  79   797979
Year 0Cross-Listed0.079***0.101***0.038***0.049***0.052***2239.666***320.559***
 0.070***0.0530.024**−0.0290.060***9696.390***1350.666***
Control0.067***0.096***0.025***0.029***0.042***2240.062***214.873***
 0.012−0.132−0.032−0.2110.059***3856.943***615.479***
Difference0.018*0.0100.0100.0090.005427.825***120.648***
 0.0580.1850.0560.1820.0015839.447**735.187***
No of Firms   79  79   79  79   797979
Year +1Cross-Listed0.094***0.117***0.045***0.063***0.049***2590.617***409.168***
 0.0480.045−0.007−0.0430.057***10485.590***1405.843***
Control0.080***0.095***0.035***0.041***0.041***2323.88***242.312***
 0.039−0.024−0.014−0.1050.050***6120.982***689.273***
Difference0.017**0.0140.018**0.0110.006*567.549***133.819***
 0.0090.0690.0070.0620.0074364.608**716.570***
No of Firms  79  79  79  79  797979
Year +2Cross-Listed0.087***0.111***0.047***0.052***0.040***2766.231***394.613***
 0.090***0.0780.040***−0.0090.052***11080.140***1420.771***
Control0.077***0.092***0.031***0.037***0.038***2517.007***252.815***
 0.066***0.0280.017−0.0470.050***6343.857***707.972***
Difference0.022***0.0190.023**0.0140.000595.259***183.249***
 0.0240.0500.0230.0380.0024736.283**712.799***
No of Firms  77  77  77  77  777777
Year +3Cross-Listed0.085***0.114***0.039***0.054***0.037***3520.530***488.324***
 0.077***0.097**0.0270.0080.057***14089.640***1683.659***
Control0.075***0.098***0.033***0.044***0.028***2905.928***236.137***
 0.052***0.1180.0020.0390.040***6746.221***869.652***
Difference0.0020.0210.0100.0100.000730.537***173.611***
 0.025−0.0210.025−0.0310.0177343.419***814.007**
No of Firms  72  72  72  72  727272
Panel B: Operating Performance and Firm Characteristics Around the Listing for Level III Firms
  Operating Return plus Depreciation on AssetsOperating Return plus Depreciation on SalesOperating Return on AssetsOperating Return on SalesCapital Expenditures on AssetsSalesOperating Cash Flows
  1. Notes: This table shows medians (means) in the top row (bottom row in italics) of various operating performance measures and firm characteristics for cross-listed firms, for non-cross-listed matched firms (control firms) and for the differences between cross-listed and control firms' measures. Difference measures are defined to be cross-listed firm operating performance measure minus control firm operating performance measure. Control firms were matched by country, year, industry and the closest total assets. Panel A (Panel B) reports the results for Level II (Level III) firms and the corresponding control firms. The variables examined are the following: operating income plus depreciation to assets, operating return plus depreciation on sales, operating return on assets, operating return on sales, capital expenditures on assets. We also present the following firm characteristics: sales and operating cash flows (in millions of $US). No of Firms indicates the number of firms available in a particular year. The Wilcoxon matched-pair signed-ranks test and the paired t-test were used to test for the significance of the results. Significance (two-tailed) is denoted by ***, ** and *, for p < 0.01, p < 0.05, and p < 0.10 respectively.

Year −3Cross-Listed0.075**0.137**0.046**0.071**0.037***698.769***49.657***
 0.053*0.0560.004−0.0550.076***9248.923*1432.997
Control0.087***0.129***0.036***0.058***0.033***569.157***79.224***
 0.091***0.177***0.052***0.086***0.065***4802.944*1054.947
Difference−0.018−0.058−0.025−0.0160.00087.060*−1.092
 −0.038−0.121−0.048−0.1410.0114445.979378.050
No of Firms 21 21 21 21 21 21 21
Year −2Cross-Listed0.086***0.157***0.049***0.078**0.065***560.915***71.141***
 0.096***0.0850.046*−0.0050.108***7564.065*936.549**
Control0.083***0.114***0.039***0.052***0.036***514.323***63.305***
 0.100***0.154***0.058***0.082***0.065***4455.055**550.556**
Difference−0.012−0.050−0.029−0.0390.012*45.053*6.712
 −0.004−0.069−0.012−0.0870.0433109.010385.993
No of Firms 26 26 26 26 26 26 26
Year −1Cross-Listed0.073***0.134***0.037***0.068**0.042***1054.144***58.401***
 0.091***0.118**0.040*0.0250.068***7561.396*1038.762*
Control0.107***0.139***0.056***0.070***0.032***525.727***61.716***
 0.0590.119**0.0160.0500.053***4640.695**610.863*
Difference−0.009−0.0140.001−0.0060.00088.750**12.682
 0.032−0.0010.024−0.0250.0152920.701427.899
No of Firms 29 29 29 29 29 29 29
Year 0Cross-Listed0.116***0.179***0.041***0.052***0.064***1146.998***195.573***
 0.124***0.0870.063**−0.0170.084***8495.467*1720.577**
Control0.082***0.133***0.032***0.052***0.047***721.196***96.154***
 0.085***0.151***0.044***0.0800.074***5040.088**689.184**
Difference0.0190.0400.0150.0020.000297.690***53.549**
 0.039−0.0640.019−0.0970.0103455.3791031.393*
No of Firms 29 29 29 29 29 29 29
Year +1Cross-Listed0.112***0.224***0.060***0.067**0.064***1249.405***118.553***
 0.117***0.186***0.0390.0470.095***9225.929**2002.110**
Control0.078***0.084***0.0410.0390.039***675.084***52.160***
 0.0360.086**−0.0080.0120.053***5940.281**827.894**
Difference0.033**0.052*0.022*0.0130.012**326.939***75.300***
 0.081*0.100*0.0470.0350.042*3285.6481174.216**
No of Firms 29 29 29 29 29 29 29
Year +2Cross-Listed0.107***0.153***0.048***0.077***0.045***1546.529***276.996***
 0.115***0.193**0.048**0.0360.074***9616.310**1630.268**
Control0.089***0.130***0.051***0.067***0.032***762.099***73.621***
 0.085***0.133***0.041***0.0490.054***6725.778**823.494**
Difference0.02**0.036*0.005−0.0080.008389.605**77.602***
 0.030*0.060*0.007−0.0130.0202890.532806.774**
No of Firms 29 29 29 29 29 29 29
Year +3Cross-Listed0.114***0.194***0.069**0.104**0.037***799.003***180.067***
 0.112***0.120*0.0330.0270.047***8943.489963.002
Control0.074***0.111***0.023**0.044**0.027***461.547***37.141***
 0.067***0.109***0.032*0.051**0.039***5495.384360.860
Difference0.063*0.0500.0370.0490.000199.015*161.810***
 0.045*0.0110.001−0.0240.0083448.105602.142*
No of Firms 21 21 21 21 21 21 21

Panel B of Table 2 reports the results for Level III firms. Consistent with hypothesis 1, the operating performance of Level III firms is not significantly different relative to the control sample prior to the listing. Significant differences in operating performance between cross-listed and control firms emerge after the listing. Specifically, both Level III and control firms are profitable firms across the period under examination. Furthermore, all the median matched-adjusted operating performance measures (except capital expenditures to assets and total sales) are not significant in each of the years −3, −2 and −1. After the listing, the median matched adjusted operating return on assets is significant in each of the years +1, +2 and +3 and it ranges from 0.021 to 0.063. A similar pattern is observed for the other operating performance measures, although on some occasions the median matched-adjusted measures are not significant at conventional levels. Overall, the results suggest that cross-listing enhances the ability of Level III firms to increase their operating performance. To examine whether differences in operating performance measures between cross-listed and control firms after the listing are due to the ability of the cross-listed firms to use their assets more efficiently relative to the control sample, we examine for possible differences in their asset turnover ratio. Untabulated results show no significant differences in asset turnover ratio between cross-listed and control firms, suggesting that cross-listing does not enhance the firms' ability to use their assets more efficiently compared to the control firms.

We then concentrate on the long-run operating performance of the cross-listed firms, rather than on yearly measures, by splitting our sample into (a) the pre-listing period that consists of years −3, −2 and −1, (b) the listing period that consists of year 0 and (c) the post-listing period that consists of years +1, +2 and +3. To prevent possible bias in the comparison of the pre-listing, the listing and the post-listing periods we use firms with full data for the period −3 to +3.10 This restriction reduces the sample size for both Level II and Level III firms to 51 and 12 firms, respectively. Results are presented in Table 3. Panels A and B report results for the Level II and Level III firms, respectively. Overall, these results are stronger than those reported in Table 2. During the pre-listing period, Level II firms seem to be more profitable relative to the control sample, have higher capital expenditures, higher sales and higher operating cash flows. During the listing and post-listing periods, Level II firms continue to be more profitable relative to the control sample, although differences between the two samples become weaker. In contrast to Level II firms, Level III firms are not more profitable relative to the control sample prior to the listing. During the listing and the post-listing periods, Level III firms become more profitable relative to the control sample, have higher capital expenditures, higher sales and higher operating cash flows.

Table 3.  Operating Performance Measures and Firm Characteristics for the Pre-Listing, Listing and the Post-Listing Periods, 1994–2004
Panel A: Operating Performance and Firm Characteristics for Level II Firms During the Pre-Listing, the Listing and the Post-Listing Periods
  Operating Return plus Depreciation on AssetsOperating Return plus Depreciation on SalesOperating Return on AssetsOperating Return on SalesCapital Expenditures on AssetsSalesOperating Cash Flows
Pre-Listing PeriodCross-Listed0.093***0.119***0.039***0.061***0.054***2427.280***379.740***
 0.089***0.104*0.042***0.0120.068***8475.675***1210.542
Control0.077***0.091***0.031***0.045***0.042***2989.520***336.280***
 0.062***0.139***0.017*0.0650.060***5922.618***684.453
Difference0.013**0.017***0.007**0.012***0.006**113.287**104.075***
 0.027**−0.0350.025**−0.0530.0082553.057***526.089***
No of Obs. 153 153 153 153 153 153 153
No of Firms 51 51 51 51 51 51 51
Listing PeriodCross-Listed0.075***0.110***0.036***0.049***0.049***2647.030***419.900***
 0.075***0.096**0.029*0.0080.058***10443.700***1519.719***
Control0.075***0.097***0.026***0.029***0.042***2982.880***300.130***
 0.059*0.104*0.0070.0190.066***6622.824***730.751***
Difference0.0120.0150.0100.0090.000348.367*128.623***
 0.016−0.0080.022−0.011−0.0083820.876788.968**
No of Obs. 51 51 51 51 51 51 51
No of Firms 51 51 51 51 51 51 51
Post-Listing PeriodCross-Listed0.083***0.112***0.032***0.051***0.041***4368.050***599.850***
 0.070***0.071*0.019−0.0210.052***11864.890***1484.496***
Control0.079***0.098***0.032***0.041***0.040***3028.180***311.570***
 0.059***0.142***0.0030.0580.052***6864.930***845.990***
Difference0.006*0.0150.001*0.0080.001562.551***184.982***
 0.011−0.0710.016−0.0790.0004999.960***638.506***
No of Obs. 153 153 153 153 153 153 153
No of Firms 51 51 51 51 51 51 51
Panel B: Operating Performance and Firm Characteristics for Level III Firms During the Pre-Listing, the Listing and the Post-Listing Periods
  Operating Return plus Depreciation on AssetsOperating Return plus Depreciation on SalesOperating Return on AssetsOperating Return on SalesCapital Expenditures on AssetsSalesOperating Cash Flows
  1. Notes: This table shows medians (means) in the top row (bottom row in italics) of various operating performance measures and firm characteristics for cross-listed firms, for non-cross-listed matched firms (control firms) and for the differences between cross-listed and control firms' measures. Difference measures are defined to be cross-listed firm operating performance measure minus control firm operating performance measure. Control firms were matched by country, year, industry and the closest total assets. Panel A (Panel B) reports the results for Level II (Level III) firms and the corresponding control firms. We use firms with full data for the period −3 to +3, which reduce the initial sample to 51 Level II firms and 12 Level III firms for each year. We define pre-listing period to refer to years −3, −2 and −1 (153 and 36 firm-year observations for Level II and Level III firms, respectively), listing period refers to year 0 (51 and 12 firm-year observations for Level II and Level III firms, respectively) whereas post-listing period refers to years +1, +2 and +3 (153 and 36 firm-year observations for Level II and Level III firms, respectively). Number of obs. indicates the number of firm-year observations available in a particular period. No of Firms indicates the number of firms available in a particular period. The variables examined are the following: operating return plus depreciation to assets, operating return plus depreciation on sales, operating return on assets, operating return on sales, capital expenditures on assets. We also present the following firm characteristics: sales and operating cash flows (in millions of $US). The Wilcoxon matched-pairs signed-ranks test and the paired t-test were used to test for the significance of the results. Significance (two-tailed) is denoted by ***, ** and *, for p < 0.01, p < 0.05, and p < 0.10 respectively.

Pre-Listing PeriodCross-Listed0.086***0.161***0.062***0.122***0.058***262.710***37.340***
 0.089***0.172**0.041*0.0050.095***8961.385*97.957***
Control0.096***0.148***0.057***0.070***0.041***232.260***21.950***
 0.071*0.104**0.0370.088**0.053***4803.498**544.798**
Difference−0.0060.000−0.023−0.0200.007**−11.6102.055
 0.0180.0680.004−0.0830.042**4157.887−446.841
No of Obs. 36 36 36 36 36 36 36
No of Firms 12 12 12 12 12 12 12
Listing PeriodCross-Listed0.106***0.195**0.071*0.0780.038***508.200***108.150***
 0.137**0.185***0.0690.0720.076**10456.550683.455
Control0.053*0.075**0.0170.0320.047***399.030***46.090**
 0.054**0.092***0.0200.0510.095**5412.107427.245
Difference0.053**0.1050.0330.087−0.006103.53784.592*
 0.083*0.093***0.0490.021−0.0195044.443256.210
No of Obs. 12 12 12 12 12 12 12
No of Firms 12 12 12 12 12 12 12
Post-Listing PeriodCross-Listed0.117***0.204***0.060***0.089***0.043***601.380***128.500***
 0.123***0.184***0.0380.074**0.076***11426.610*975.700**
Control0.079***0.123***0.047*0.065**0.028***405.110***41.400***
 0.044*0.092***0.0080.0270.040***5948.664**516.945**
Difference0.049***0.063***0.022*0.035*0.010***120.285***76.285***
 0.079**0.092***0.0300.0470.036**5477.946*458.755**
No of Obs. 36 36 36 36 36 36 36
No of Firms 12 12 12 12 12 12 12
(b) Changes in Operating Performance

In the previous subsection we presented results using the levels of operating performance measures for both cross-listed and non-cross-listed matched firms (control firms). In this subsection, to emphasise the cross-listing effect on operating performance measures we present post-listing changes in operating performance measures for both cross-listed firms and control firms relative to year −1, as well as matched-adjusted changes in operating performance measures defined as the difference between cross-listed and control firms' measures. Results are presented in Table 4. Panel A reports the results for the Level II firms. Consistent with the aforementioned results of Tables 2 and 3, Level II firms do not experience any positive changes in their operating performance measures. Specifically, neither the median matched-adjusted changes in operating return plus depreciation on assets/sales, nor the median matched-adjusted changes in operating return on assets/sales are statistically significant at conventional levels. Similarly, the median matched-adjusted changes in capital expenditures on assets are insignificant. Median changes in total sales for Level II firms range from 0.065 to 0.444, and are significant in each of the years −1 to 0, −1 to +1, −1 to +2 and −1 to +3. Median changes in total sales for the control firm's range from 0.019 to 0.264, and they are significant in each one of the year's −1 to +1, −1 to +2 and −1 to +3. Median matched-adjusted changes in total sales are positive and significant in each of the years −1 to +1, −1 to +2 and −1 to +3; however, their effect on operating cash flows is ambiguous in the sense that while for years −1 to 0 and −1 to +1 the median matched-adjusted changes in operating cash flows are positive, for years −1 to +2 and −1 to +3 the median matched-adjusted changes in operating cash flows are negative. Overall, the results suggest that Level II firms do not experience significant changes in their operating performance relative to year −1.

Table 4.  Changes in Operating Performance Measures and Firm Characteristics from Year −1 to Years 0, +1, +2 and +3, 1994–2004
Panel A: Changes in Post-Listing Operating Performance and Firm Characteristics for Level II Firms Relative to Year −1
  Operating Return plus Depreciation on AssetsOperating Return plus Depreciation on SalesOperating Return on AssetsOperating Return on SalesCapital Expenditures on AssetsSalesOperating Cash Flows
Year −1 to 0Cross-Listed0.007−0.0120.0340.008−0.0020.065***−0.331***
 0.0570.0260.0120.0520.324**0.195***0.733***
Control0.0070.029−0.0130.055−0.0010.019−0.206
 0.1920.171−0.104−0.0360.584**0.047−0.087
Difference−0.048−0.081−0.018−0.080−0.0610.045*0.069***
 −0.135−0.1450.1160.088−0.2600.148**0.820***
No of Firms797979797979  79
Year −1 to +1Cross-Listed0.1140.0770.0760.063−0.0540.182***−0.147***
 0.1640.0950.0750.0410.371**0.344***0.932***
Control−0.0180.0620.0420.114−0.0870.056**−0.210
 −0.6510.238**−0.176−0.0170.808**0.137***0.156
Difference0.085−0.0550.1530.013−0.0560.062**0.007***
 0.815−0.1430.2510.058−0.4370.207***0.776***
No of Firms 79 79 79 79 79 79  79
Year −1 to +2Cross-Listed0.0580.061−0.0300.001−0.1290.311***0.134***
 0.316**0.231**0.2750.2460.284*0.490***1.340***
Control0.0450.0820.0070.062−0.1480.172***0.301**
 0.4130.195*−0.0520.0160.4260.291***0.489***
Difference0.1330.0850.1800.053−0.1090.094*−0.016*
 −0.0970.0360.3270.230−0.1420.199**0.851***
No of Firms 77 77 77 77 77 77  77
Year −1 to +3Cross-Listed0.0180.007−0.072−0.029−0.279***0.444***−0.371***
 0.216*0.120−0.024−0.0780.0460.676***1.175***
Control0.0200.1490.0620.130−0.182***0.264***0.522
 0.1390.248**−0.0830.0950.1140.297***0.383**
Difference0.043−0.0460.071−0.0080.0000.116**−0.221**
 0.077−0.1280.059−0.173−0.0680.379***0.792**
No of Firms 72 72 72 72 72 72  72
Panel B: Changes in Post-Listing Operating Performance and Firm Characteristics for Level III Firms Relative to Year −1
  Operating Return plus Depreciation on AssetsOperating Return plus Depreciation on SalesOperating Return on AssetsOperating Return on SalesCapital Expenditures on AssetsSalesOperating Cash Flows
  1. Notes: This table shows medians (means) in the top row (bottom row in italics) of changes in operating performance measures and firm characteristics from year −1 to years 0, +1, +2 and +3 for both cross-listed and non-cross-listed matched firms (control firms) and differences between cross-listed and control firms' measures. Difference measures are defined to be cross-listed firm's changes in operating performance measure minus control firm's changes in operating performance measure. Control firms were matched by country, year, industry and the closest total assets. Panel A (Panel B) reports the results for Level II (Level III) firms and the corresponding control sample of firms. The variables examined are the following: operating return plus depreciation to assets, operating return plus depreciation on sales, operating return on assets, operating return on sales, capital expenditures on assets. We also present the following firm characteristics: sales and operating cash flows. No of Firms indicates the number of firms available in a particular period. The Wilcoxon matched-pairs signed-ranks test and the paired t-test were used to test for the significance of the results. Significance (two-tailed) is denoted by ***, ** and *, for p < 0.01, p < 0.05, and p < 0.10 respectively.

Year −1 to 0Cross-Listed0.253**0.194*0.201*0.3400.0170.176***1.388*
 0.503**0.2650.651*0.5150.1780.517***1.097**
Control−0.098−0.021−0.1020.0130.208*0.0930.987
 0.4780.0880.2780.3152.355**0.159*0.635*
Difference0.289**0.2420.377*0.285−0.423*0.123**0.021
 0.0250.1770.373*0.200−2.177*0.358**0.462
No of Firms 29 29 29 29 29 29  29
Year −1 to +1Cross-Listed0.179*0.1830.286*0.3310.0220.300***1.318***
 0.921**0.363*0.4980.3890.538*0.641***1.599***
Control−0.092−0.100−0.261−0.249−0.2730.100**0.829**
 0.726−0.054−0.137−0.3062.099*0.188**0.758**
Difference0.2770.3190.547*0.565*0.0140.186***0.000
 0.1950.417**0.635*0.695**−1.5610.453**0.841
No of Firms 29 29 29 29 29 29  29
Year −1 to +2Cross-Listed0.1280.182*−0.2340.190−0.1060.391***−0.235**
 0.487*0.431**0.4930.5400.1230.790***1.023**
Control−0.185−0.073−0.139−0.061−0.2210.188**0.040**
 0.4820.0170.025−0.0151.878*0.226**1.198**
Difference0.284**0.291***0.3650.313*−0.2100.380***0.004
 0.0050.414**0.4680.555*−1.755*0.564***−0.175
No of Firms 29 29 29 29 29 29  29
Year −1 to +3Cross-Listed0.1960.079−0.0350.108−0.610**1.003***0.809**
 0.8900.210*−0.225−0.2280.5551.227***1.801**
Control−0.413−0.275−0.414−0.546−0.616**0.217**0.106
 0.592−0.191−0.606−0.575*0.2090.305**1.232**
Difference0.562**0.454**0.5540.6540.0000.735***0.788
 0.2980.401*0.3810.3470.3460.922***0.569
No of Firms 21 21 21 21 21 21  21

Panel B of Table 4 reports the results for Level III firms. Overall, Level III firms experience improvements in their post-listing changes in operating return measures relative to year −1. Specifically, the median changes in operating return plus depreciation on assets, relative to year −1, range from 0.128 to 0.253. Similarly, the median changes in the other operating performance measures relative to year −1, are positive. Note also that some of these operating performance changes are significant at conventional levels. At the same time, the median changes in operating performance measures for the control sample are in general negative but insignificant. As a result, the median matched-adjusted changes in operating performance, relative to year −1, are positive and in some occasions significant. More specifically, the median matched-adjusted changes in operating return plus depreciation on assets (on sales) range from 0.179 to 0.562 (0.183 to 0.454) and they are significant in each one of the periods −1 to 0, −1 to +2 and −1 to +3 (−1 to +2 and −1 to +3). In this line, the median matched-adjusted changes in operating return on assets (on sales) range from 0.365 to 0.554 (0.285 to 0.654) and they are significant in each one of the periods −1 to +1, −1 to +2 (−1 to +2 and −1 to +3). Finally, as expected, operating performance improvements for Level III firms are supported by higher sales after the listing in each one of the years −1 to 0, −1 to +1, −1 to +2 and −1 to +3.

(c) Multivariate Analysis

In the previous subsection we provided evidence that Level III firms experienced improvements in operating performance whereas no such improvements were observed for Level II firms. To test hypothesis 2, that the operating performance improvements for Level III firms are more pronounced relative to operating performance improvements for Level II firms, we employ a multivariate regression model. In this model we control for differences between Level II and Level III firms due to the matched character of the sample i.e. firm characteristics and for differences with respect to firms' country characteristics and the industry. Specifically, we run various specifications of the following model:

  • image

where ‘MOPM’ are various matched-adjusted operating performance measures defined as the difference in operating performance measure between cross-listed and non-cross-listed matched firms (control firms). These measures are (i) the matched-adjusted operating return plus depreciation on assets, (ii) the matched-adjusted operating return plus depreciation on sales, (iii) the matched-adjusted operating return on assets and (iv) the matched-adjusted operating return on sales, respectively. The explanatory variables are as follow: Listing is a dummy variable that equals one if the observation is for the years after the listing and zero otherwise. To the extent that cross-listing enhance firms' operating performance the coefficient of the variable Listing is expected to be positive. Moreover, to provide insights on possible differences in operating performance between Level II and Level III firms; we interact the variable Listing with the variable Type which is a dummy variable that equals one for Level III firms and zero for Level II firms. A positive interaction term will provide support to hypothesis 2 that the operating performance improvements for Level III firms are more pronounced relative to operating performance improvements for Level II firms.

We also control for imperfect matching with respect to the growth opportunities and the size of the firms using the variables M_MB defined as the matched-adjusted market to book ratio (i.e. the difference in the market to book ratios between cross-listed and control firms) and M_Size defined as the matched-adjusted size (i.e. the difference in logarithmic total assets between cross-listed and control firms). Moreover, we use country-level variables taken from La Porta et al. (1998), to control for possible variation between Level II and Level III firms with respect to the degree of anti-director rights AR, the efficiency of the judicial system EJ and the quality of the accounting standards AS. Finally, we control for industry effect using dummy variables for certain industries i.e. communications and chemicals.11

Table 5 presents the results of the aforementioned regression models using robust standard errors.12Listing is insignificant across all models suggesting that after controlling for imperfect matching, country characteristics and industry effect, there is no relation between cross-listing and operating performance. Consistent with hypothesis 2, Listing*Type is positive and significant when the dependent variables are the matched-adjusted operating return plus depreciation on assets/sales and the matched-adjusted operating return on sales (ρ= 0.01, ρ= 0.05 and ρ= 0.10, respectively). There is also a positive relation between the variable Listing*Type and the matched-adjusted operating return on assets, however, the relation is not statistically significant. As far as the control variables are concerned, M_MB and EJ are insignificant in all four-regression models. M_Size is positively related with matched-adjusted operating return plus depreciation on assets/sales and with matched-adjusted operating return on assets/sales variables (ρ= 0.01, ρ= 0.01, ρ= 0.01 and ρ= 0.01, respectively). Furthermore, AS is positively related with matched-adjusted operating return plus depreciation on assets/sales and with matched-adjusted operating return on assets (ρ= 0.05, ρ= 0.05 and ρ= 0.05, respectively), whereas AR is negatively related with matched-adjusted operating return on sales (ρ= 0.05). These results suggest that country characteristics affect significantly firms' operating performance. More specifically, matched-adjusted operating performance is higher, the better the accounting standards and the worse the anti-directors rights.

Table 5.  Results of Regression of Matched-Adjusted Operating Performance Measures Relative to the Listing, the Type of ADR and Control Variables, 1994–2004
VariablesMatched-Adjusted Operating Returns plus Depreciation on AssetsMatched-Adjusted Operating Returns plus Depreciation on SalesMatched-Adjusted Operating Returns on AssetsMatched-Adjusted Operating Returns on Sales
  1. Notes: This table presents results of pooled regression estimates with robust standard errors adjusted for heteroskedasticity bias of the following model:

    • image

    MOPM are matched-adjusted operating performance measures defined as the difference in operating performance measure between cross-listed and non-cross-listed matched firms (control firms). These measures are (i) the matched-adjusted operating return plus depreciation on assets, (ii) the matched-adjusted operating return plus depreciation on sales, (iii) the matched-adjusted operating return on assets, (iv) the matched-adjusted operating return on sales. The explanatory variables are as follows: Listing is a dummy variable that equals one if the observation is for the years after the listing and zero otherwise. Type is a dummy variable that equals one for Level III firms and zero for Level II firms. M_MB is the matched-adjusted market to book ratio defined as the difference in the market to book ratios between cross-listed and control firms. M_Size is the matched-adjusted size defined as the difference in logarithmic total assets between cross-listed and control firms. Anti-director rights (AR) is an index that aggregates six different shareholder rights. Efficiency of the judicial system (EJ) is an assessment of the efficiency and integrity of the legal environment as it affects business. The accounting standards (AS) rating is an index created by examining and rating companies' annual reports for their inclusion or exclusion of 90 items. AR, AS and EJ were taken from LaPorta et al. (1998). Industry Dummies are dummy variables that equal 1 for certain industries (i.e. communication and chemicals) and 0 otherwise. In this regression we use firms with full data during the period −3 to +3. This requirement reduces the sample to 51 Level II firms and 12 Level III firms for each year or 357 firm-year observations for Level II firms and 84 firm-year observations for Level III firms. Number of observations indicates the number of firm-year observations available. F-Statistic assesses the joint significance of the explanatory variables. Adjusted R2 is an indicator of the goodness of fit of the model in the population. White statistic (x2) test for heteroskedastic disturbances. The t-statistic for the significance of the coefficients is in italics below the coefficient estimates. Significance is designated by *** at 1%, ** at 5% and * at 10%.

Intercept−0.068*−0.261***−0.092**−0.201
−1.82−3.00−2.28−1.64
Listing−0.101−0.028−0.008−0.03
−1.240.153−0.88−1.19
Listing*Type0.041***0.057**0.0180.063*
2.682.011.061.74
M_MB0.0020.0030.0010.002
1.391.561.030.87
M_Size0.021***0.055***0.022***0.047***
3.234.113.272.65
AR0.000−0.009−0.003−0.026**
0.02−1.26−0.83−2.52
AS0.001**0.003**0.002**0.003
2.012.271.981.21
EJ−0.0020.005  00.012
−0.610.770.161.3
Industry Dummies Yes  Yes  Yes  Yes
No of Observations441441441441
F-Statistic3.30***5.25***2.76***3.74***
Adjusted R20.0750.1190.0440.059
White statistic (x2)138.852***176.574***168.561***183.299***

Overall, these results are consistent with hypothesis 2 where we hypothesised that the improvements in operating performance measures should be more pronounced for capital-raising firms. These results may suggest that cross-listing effects may not affect in the same manner all exchange-listed firms. Moreover the observed improvements in operating performance for capital-raising firms contradict prior results on other capital-raising events. In particular, for both initial public offerings and seasoned equity offerings a negative operating performance is observed subsequent to the event (Jain and Kini, 1994; and Loughran and Ritter, 1997). Consequently, our results suggest that capital-raising through cross-listing is different from both initial and secondary capital-offerings.

(ii) Return Performance of the Cross-Listed Firms

In the previous section we provided evidence of the operating performance of the exchange-listed firms around the cross-listing. In this section we provide evidence of the return performance of exchange-listed firms around the cross-listing. In particular we provide evidence for the long-run return performance of the cross-listed firms by the type of listing. We calculate the matched-adjusted returns (AR) and the cumulative matched-adjusted returns (CAR). Similar to Ritter (1991), Loughran and Ritter (1995) and Foerster and Karolyi (2000), we estimate the monthly raw returns as the geometric capital gain return beginning either from the listing date or four months after the previous fiscal year, whichever comes later in fixed length intervals, regardless of calendar month-end dates. Specifically, the monthly return is defined as follows:

  • image

where Ri,m is the monthly return over the fixed 21-day interval for stock i, Ri,d is the return for stock i on day d, measured in US dollars. The monthly returns are calculated for the twelve-month period prior to the event and the twelve-month period after the event. Next, we use the monthly stock returns to calculate the matched-adjusted returns in the following way:

  • image

where Rim is the geometric capital gain return of the matched firm during the corresponding month m. We then compute for each month an average matched-adjusted return (AR), across firms as:

  • image

Finally, we compute the cumulative matched-adjusted returns (CAR) as follows:

  • image

where t1 and t2 are the beginning and ending periods. To facilitate the cross-sectional analysis of the long-run performance of the cross-listed firms, we calculate the holding period matched-adjusted returns. For each stock, we calculate its geometric return over the entire period of interest (twelve month period prior to the event) as:

  • image

where t1 and t2 are the beginning and ending periods.

Results in Table 6 show the raw returns, cumulative returns, abnormal returns and cumulative abnormal returns for both cross-listed and non-cross-listed matched firms. For the twelve-month period prior to the listing cumulative returns for Level II firms are 10.2% whereas for the control firms are −15.3%, resulting in 34.3% cumulative abnormal returns. For the twelve-month period after the listing both Level II and control firms experience negative cumulative returns (−17.1% and −14.7%, respectively), resulting in −3.6% cumulative abnormal returns. As far as Level III firms is concerned, for the twelve-month period prior to the listing, cumulative returns are 40.8% whereas for the control firms cumulative returns are only 3.5%. As a result, cumulative abnormal returns for the twelve-month period prior to the listing are 21.8%. However, in contrast to Level II firms, Level III firms do not underperform control firms after the listing. Specifically, the cumulative abnormal return for Level III firms after the listing is 5.2%.

Table 6.  Return Performance Around the Cross-Listing
Panel A: Return Performance Around the Listing for Level II Firms
MonthCross-Listed FirmsControl FirmsDifference
Raw ReturnsCum ReturnsRaw ReturnsCum ReturnsARCAR
  1. Notes: Raw returns (cum returns) for each type of exchange-listed ADRs are calculated as geometric capital gains returns over 21 fixed-day interval (months) around the listing date or four months after the previous fiscal year, whichever comes later regardless of calendar month-end dates. Abnormal returns (AR) and cumulative abnormal returns (CARs) are calculated as the difference between the cross-listed firm's stock return and the corresponding non-cross-listed firms' (control firms') return for the corresponding period. Missing data during some of the months are dealt with by assuming that their return is equal to the control firms' return, by assigning these firms AR of zero during those months. Panel A presents results for seventy-nine Level II firms whereas Panel B presents results for twenty-nine Level III firms. Data are taken from Datastream database on a daily basis. T-statistic for the raw returns and the abnormal returns are computed as Rt*√nt/sdt, where the Rt is the average return for month t, nt is the number of observations for month t and sdt is the cross-sectional standard deviation of returns for month t. t-statistics for the cumulative returns and the cumulative abnormal return in month t, CARt, are computed as CARt*√nt/csdt where csdt is computed as csdt=[t*var+2*(t−1)*cov]1/2 where var is the average (over either the pre-issuance or post-issuance period) cross-sectional variance and cov is the first-order autocovariance of the Rt series. Significance is designated by *** at 1%, ** at 5% and * at 10%.

−120.0070.007−0.011−0.0110.0100.010
−110.0050.013−0.017−0.0280.034*0.044
−10−0.0120.0010.002−0.0260.0140.059
 −90.0420.0430.018−0.0080.0410.102
 −80.0190.0630.002−0.0060.0180.123
 −70.0390.1050.003−0.0030.0430.171
 −6−0.0190.084***−0.024−0.0270.0070.179***
 −5−0.0080.075−0.025*−0.0520.0110.191**
 −4−0.0140.059−0.027**−0.078**0.0140.208***
 −3−0.0110.048−0.010−0.087**−0.0030.203***
 −20.0310.081−0.053***−0.136**0.088**0.310***
 −10.0190.102***−0.020−0.153**0.0250.343***
  1−0.017−0.017−0.041**−0.0410.0210.021
  2−0.001−0.018−0.003−0.045−0.0070.013
  3−0.011−0.029−0.017−0.061*0.0060.019
  4−0.031−0.0590.012−0.050−0.045−0.027
  50.004−0.055−0.004−0.0540.018−0.010
  6−0.009−0.0630.002−0.051−0.008−0.018
  7−0.029*−0.091−0.011−0.062*−0.018−0.036
  8−0.021−0.1100.006−0.057−0.001−0.037
  9−0.021*−0.129***−0.002−0.059−0.015−0.051
 10−0.025−0.151**−0.050**−0.1060.000−0.051
 11−0.004−0.154***−0.025−0.128**0.030−0.022
 12−0.020−0.171**−0.021−0.147**−0.014−0.036
 
Panel B: Return Performance Around the Listing for Level III Firms
MonthCross-Listed FirmsControl FirmsDifference
Raw ReturnsCum ReturnsRaw ReturnsCum ReturnsARCAR
 
−120.0240.0240.0210.021−0.012−0.012
−110.057*0.082−0.0080.0130.0380.025
−100.0070.0890.0290.042−0.0120.013
 −90.074*0.170**0.0350.0790.0340.048
 −80.0120.183**−0.0070.072**0.0070.055
 −70.063**0.258***0.0560.132−0.0210.033
 −60.086**0.367***0.073*0.2150.0330.068
 −50.064**0.455***−0.0070.207**0.077***0.150*
 −4−0.0010.453***−0.043**0.154**−0.0040.145**
 −3−0.0080.442***−0.0630.0810.0480.201**
 −2−0.043*0.380***−0.053*0.0240.0050.207***
 −10.0200.408***0.0110.0350.0100.218***
  1−0.026−0.026−0.039−0.0390.0250.025
  2−0.007−0.033−0.066**−0.102***0.0420.068*
  30.015−0.019−0.011−0.112***0.0200.089
  4−0.017−0.035−0.051−0.158**0.0140.104
  5−0.016−0.051−0.046−0.196**0.0100.115**
  60.010−0.0420.015−0.184**−0.0080.106
  70.026−0.0170.042−0.150*−0.0020.103
  8−0.009−0.0250.041−0.115−0.0410.058
  9−0.052−0.0760.004−0.112−0.0340.022
 100.015−0.0620.031−0.0840.0190.042
 11−0.026−0.0860.001−0.0840.0110.053
 12−0.003−0.0890.003−0.081−0.0010.052

(iii) The Relation Between Return Performance and Future Operating Performance

In the previous section we provided evidence that both Level II and Level III firms have positive abnormal returns for the twelve-month period prior to the cross-listing. In this section, we use the twelve-month pre-listing abnormal returns as proxy for cross-listing benefits because in many cases, during the announcement, investors may not be certain on whether a cross-listing will finally take place or not. For example, some companies may initially announce that they will examine the possibility for a cross-listing in the near future. Some other companies may announce the progress of the cross-listing procedures several times, something that gradually increases the certainty with respect to the final outcome. Finally, there is always the possibility to withdraw a prior taken cross-listing decision. Such kind of uncertainty may introduce noise to the cross-listing announcement returns which is an alternative measure of cross-listing benefits. The twelve-month pre-listing abnormal return, given its broader nature, may eliminate the aforementioned noise. The disadvantage of the twelve-month pre-listing abnormal return, however, is that it exacerbates the possibility that a firm characteristic or event other than the cross-listing is related to stock returns. To address this concern, we make sure that our results are qualitatively similar using both the twelve-month pre-listing abnormal returns and announcement returns as proxy for cross-listing benefits.

Using the aforementioned proxies for cross-listing benefits we examine hypothesis 3 (results using announcement returns are not tabulated). In doing so, we examine the determinants of the abnormal return performance of the cross-listed firms using various multivariate regression analysis models. With this type of analysis we are able to control for country specific factors and for firm risk characteristics and filter out spurious correlation between various explanatory variables. Specifically, we estimate the following model:

  • image

where BHR[−1,−12] is the matched-adjusted buy-hold return for the twelve-month period prior to the listing. The explanatory variables are as follows: M_ΔOperating Return Plus Depreciation on Assetst=−1,T is the matched-adjusted change in the operating income plus depreciation on assets from year −1 to year T (where T= 0, +1, +2, +3) defined as the difference between cross-listed and non-cross-listed matched firms (control firms).13 A positive coefficient will provide support to hypothesis 3 that the higher the abnormal returns due to the cross-listing the higher the abnormal changes in operating performance after the listing. We then control for the effect of imperfect matching on matched-adjusted buy-hold returns with respect to the growth opportunities and the size of the firms using the variables M_Tobin's qt=−1,T defined as the matched-adjusted Tobin's q ratio (i.e. the difference between the market value of equity less the book value of equity plus the book value of assets between cross-listed and control firms, measured in year −1) and the M_Sizet=−1 defined as the matched-adjusted size (i.e. the difference in logarithmic total assets between cross-listed and control firms, measured in year −1). Finally, La Porta et al. (1998, 2000 and 2002) suggest that country law and regulations, through their relation with the level of investor protection, may affect company valuations. Therefore, we control for the degree of anti-director rights AR the efficiency of the judicial system EJ and the quality of the accounting standards AS using country-level variables taken from La Porta et al. (1998).

Results are presented in Table 7. Note that each regression model is estimated using robust standard errors. M_ΔOperating Return Plus Depreciation on Assetst=−1,T relates positively with the twelve month pre-listing buy-hold abnormal returns in each of the years −1 to 0, −1 to +1 and −1 to +3 (ρ= 0.01, ρ= 0.10 and ρ= 0.01, respectively). Consistent with hypothesis 3, these results indicate that the market anticipates significant operating performance improvements for both Level II and III firms after the listing. As far as firm characteristics are concerned, M_Tobin's qt=−1,T, is not statistically significant in any specification, suggesting that differences in the tobon's q ratio between cross-listed and control firms do not relate with the matched-adjusted buy hold return for the twelve-month period prior to the listing whereas, M_Sizet=−1,T is negatively related with the twelve-month pre-listing buy-hold abnormal returns in all specifications. Finally, country factors, AR, AS and EJ, do not relate with the matched-adjusted buy-hold return for the twelve-month period prior to the listing.

Table 7.  Results of Regressions of Buy-and-Hold Returns Prior to the Listing on Matched-Adjusted Operating Performance Measures and Control Variables, 1994–2004
VariablesMatched-Adjusted Buy-Hold Returns for the Twelve Month Period Prior to the Listing
  1. Notes: This table presents results of regression with robust standard errors adjusted for heteroskedasticity bias of the following model:

    • image

    where BHR[−1,−12] is the matched-adjusted buy-and-hold return for the twelve month period prior to the listing. M_ΔOperating Return Plus Depreciation on Assetst=[−1, T] is the matched-adjusted change in the operating income plus depreciation on assets from year −1 to year T (where T= 0, +1, +2, +3) defined as the difference between cross-listed and non-cross-listed matched firm (control firm). Control firms were matched by country, year, industry and the closest total assets. M_Tobin's q is the matched-adjusted Tobin's q ratio (market value of equity less the book value of equity plus the book value of assets) defined as the difference in Tobin's q ratio between cross-listed and control firms measured at year −1. M_Size is the matched-adjusted size defined as the difference in logarithmic total assets between cross-listed and control firms measured at year −1. Anti-director rights (AR) is an index that aggregates six different shareholder rights. Efficiency of the judicial system (EJ) is an assessment of the efficiency and integrity of the legal environment as it affects business. The accounting standards (AS) rating is an index created by examining and rating companies' annual reports for their inclusion or exclusion of 90 items. AR, AS and EJ were taken from LaPorta et al. (1998). No of firms indicates the number of firms in the regression. F-Statistic assesses the joint significance of the explanatory variables. Adjusted R2 is an indicator of the goodness of fit of the model in the population. White statistic test for heteroskedastic disturbances. The t-statistic is in italics below the coefficient estimates. Significance is designated by *** at 1%, ** at 5% and * at 10%.

Intercept−0.577−0.635−0.612−0.432
−0.48−0.52−0.51−0.30
M_ΔOperating Return Plus Depreciation on Assetst=[−1, 0] 0.022***   
 8.19   
M_ΔOperating Return Plus Depreciation on Assetst=[−1, +1]  0.022*  
 1.67  
M_ΔOperating Return Plus Depreciation on Assetst=[−1, +2]  −0.053 
  −0.98 
M_ΔOperating Return Plus Depreciation on Assetst=[−1, +3]    0.028***
   3.06
M_Tobin's qt=−1−0.014−0.012−0.020−0.013
−0.35−0.29−0.49−0.31
M_Sizet=−1−0.511**−0.431*−0.435*−0.585**
−2.14−1.76−1.86−2.06
AR 0.019 0.036 0.081 0.066
0.260.641.070.67
AS 0.009 0.008 0.003 0.006
0.470.410.160.26
EJ 0.035 0.044 0.060 0.037
0.420.520.780.33
No of Firms  108  108  106   93
F-Statistic 14.16*** 1.42 0.92 5.03***
Adjusted R2 0.054 0.010 0.031 0.035
White statistic (x2) 50.330*** 42.45** 50.724*** 33.354

Overall, the results in this subsection support hypothesis 3. That is, the market expectations due to the cross-listing reflects post-listing changes in firms' operating performance. These results add insight into where the revaluation effect of cross-listings may come from, by considering operating performance. Moreover, they strengthen the belief that cross-listing affects operating performance and that our results are not due to self-selection bias introduced by the matched character of our sample choice.

5. CONCLUSIONS

  1. Top of page
  2. Abstract
  3. 1. INTRODUCTION
  4. 2. BACKGROUND, MOTIVATION, AND DEVELOPMENT OF THE HYPOTHESES
  5. 3. RESEARCH DESIGN
  6. 4. EMPIRICAL RESULTS
  7. 5. CONCLUSIONS
  8. REFERENCES

This study addresses three research questions: (i) Does the post-listing operating performance of cross-listed firms increase relative to non-cross-listed matched firms, and relative to its pre-listing operating performance? (ii) Are the operating performance patterns for capital-raising cross-listed firms different from those of non-capital-raising cross-listed firms? (iii) Do changes in market expectations due to the cross-listing reflect post-listing changes in firms' operating performance? To address these research questions, we used a sample of capital-raising and non-capital-raising firms that were cross-listed on US stock exchanges during the period 1994–2004 and a country-industry-size matched sample of non-cross-listed firms.

Consistent with our expectations and hypotheses, we provide evidence for economically and statistically significant operating performance improvements only for capital-raising firms after the listing, relative to non-cross-listed matched sample of firms, and relative to the pre-listing period. Moreover, the operating performance of the capital-raising firms after the listing is significantly different from the operating performance of non-capital-raising firms, even after controlling for firm characteristics, country and industry effects. These results suggest that the type of ADR conveys information about post-listing operating performance.

Finally, we provide evidence for positive and statistically significant abnormal returns for the twelve-month period around the cross-listing, for both capital-raising and non-capital-raising firms. After controlling for firm and country risk characteristics, abnormal returns for the twelve-month period prior to the listing are positively associated with post-listing matched-adjusted changes in operating performance suggesting that the market anticipates these operating performance improvements after the listing.

Overall, the results of this study are potentially useful to managers and shareholders, especially of high growth firms with limited opportunities to raise capital. For such firms, the results suggest that a cross-listing on US stock exchanges potentially helps them to raise the capital needed to support their growth opportunities, resulting in operating performance improvements after the listing. Moreover, the results are also useful to academic researchers for at least two reasons: First, the observed improvements in operating performance for capital-raising firms contradict prior results on other capital-raising events such as initial public offerings and seasoned equity offerings something that warrants a closer examination by future research. In equilibrium, cross-sectional variation in the benefits and the costs associated with both the different requirements to raise capital and the different firm characteristics may explain the observed variation in operating performance. Second, while most previous empirical studies focused on the relation between the revaluation effect of cross-listings and cost of capital, information environment, firms' visibility and managerial incentives, this paper expands our understanding on where the revaluation effect of cross-listings may come from by considering operating performance.

Footnotes
  • 1

    Karolyi (2006) provides a review of the cross-listing literature.

  • 2

    Our results are robust when we use the announcement returns as proxy for cross-listing benefits. We present the results using the twelve month pre-listing abnormal returns as proxy for cross-listing benefits because in many cases, during the announcement, investors may not be certain on whether a cross-listing will finally take place or not. For example, some companies may initially announce that they will examine the possibility for a cross-listing in the near future. Some other companies may announce the progress of the cross-listing procedures several times, something that increase gradually the certainty with respect to the final outcome of the decision. Such kind of uncertainty in conjunction with the possibility to withdraw a prior taken cross-listing decision, may introduce noise to the cross-listing announcement returns. The twelve month pre-listing return, given its broader nature, may eliminate the aforementioned noise. The disadvantage of the twelve month pre-listing abnormal return, however, is that it may exacerbate the possibility that a firm characteristic or event other than the cross-listing is related to stock returns.

  • 3

    Alternatively, cross-listing might reduce firm's cost of capital due to improvements in the integration among the markets. Specifically, theory suggests that stock prices for firms that cross-list from segmented markets are expected to rise and their subsequent expected returns should fall as an additional built-in risk premium compensating for these barriers that dissipates. Empirical evidence consistent with the market segmentation theory is provided by Foerster and Karolyi (1999).

  • 4

    Consistent with the aforementioned evidence, theoretical models and empirical evidence by Stapleton and Subrahmanuam (1977), Errunza and Losq (1985) and Alexander et al. (1987) demonstrate how companies from segmented markets that issue equity overseas can lower their cost of capital and increase the market value of their shares.

  • 5

    Similar to Lang et. al. (2003) we adjust foreign financial data to US dollars to enhance comparability across countries because cross-listed firms originate from various countries.

  • 6

    Note that results presented in Tables 2, 3, 4 and 5 are qualitatively similar with the inclusion of these firms.

  • 7

    We use these criteria to eliminate possible differences in accounting and litigation environment between the cross-listed firm and the non-cross-listed matched firm. For example, by using matched firms from the local market, we neutralise the possible confounding effects caused when the ADR program initiation coincides to internationalise the local capital market (Stulz, 1999). In addition, we account for industry effects since prior evidence by Ritter (1991) showed that some industries experienced significant decline in operating performance during the 1980s.

  • 8

    Note that the control sample for Level III firms is not taking into consideration the capital-raising activity. The rationale lies on prior literature on capital-raising events that suggest that capital-raising does not convey any information about subsequent firm operating performance (Healy and Palepu, 1990), something that reduces the need to control for capital-raising activity. Furthermore, for firms that proceed with capital-raising activities, either Seasoned Equity Offerings or Initial Public Offerings, a significant decline in operating performance is observed after the offering relative to non-capital-raising firms (Loughran and Ritter, 1997; and Jain and Kini, 1994). As a result, it will be more likely to capture improvements in operating performance of Level III firms if we match them with capital-raising non-cross-listed firms as opposed to a matching based on non-capital-raising non-cross-listed firms (as we did). In this sense, our matching procedure eliminates the aforementioned bias and may be viewed as acting against finding any improvements in operating performance. Thus, our matching procedure is conservative with respect to differences in the capital-raising activity.

  • 9

    Due to the skewness of accounting ratios, it is typical to discuss median values in studies examining operating performance. Jain and Kini (1994), Loughran and Ritter (1997) and Papaioannou et al. (2003), among others, discuss median values.

  • 10

    Note that this restriction may introduce survivorship bias in our results. Thus, we make sure that the results presented in this subsection are robust when we relax this restriction.

  • 11

    We have 35 firms in these industries accounting for 32.41% of the sample.

  • 12

    In these regressions we include only firms with full data during the years −3 to +3. On the one hand by including only firms with full data we control for possible bias in the results due to the inclusion of different firm-year observations in the pre-listing and post-listing periods. On the other hand, with this restriction our results may suffer from survivorship bias. Consequently, we make sure that when we relax the restriction the results are qualitatively similar.

  • 13

    Similarly, matched-adjusted operating return plus depreciation on sales and operating return on sales may proxy for market expectations on cross-listed firms post-listing operating performance. We examine the robustness of our results using the rest of the operating performance measures in the regression model and we make sure that the results are qualitatively similar.

REFERENCES

  1. Top of page
  2. Abstract
  3. 1. INTRODUCTION
  4. 2. BACKGROUND, MOTIVATION, AND DEVELOPMENT OF THE HYPOTHESES
  5. 3. RESEARCH DESIGN
  6. 4. EMPIRICAL RESULTS
  7. 5. CONCLUSIONS
  8. REFERENCES
  • Alexander, G., G. Eun and S. Janakiramanan (1987), ‘Asset Pricing and Dual Listing on Foreign Capital Markets: A Note’, Journal of Finance, Vol. 42, pp. 15158.
  • Baker, K., J. Nofsinger and D. Weaver (2002), ‘International Cross-Listing and Visibility’, Journal of Financial and Quantitative Analysis, Vol. 37, pp. 495521.
  • Charitou, A., C. Louca and S. Panayides (2007), ‘Cross-Listing, Bonding Hypothesis and Corporate Governance’, Journal of Business Finance & Accounting, Vol. 34, pp. 1281306.
  • Coffee, J. (2002), ‘Racing Towards the Top? The Impact of Cross-Listing and Stock Market Competition on International Corporate Governance’, Columbia Law Review, Vol. 102, pp. 1757831.
  • Doidge, C. (2004), ‘U.S. Cross-Listing and the Private Benefits of Control: Evidence from Dual Class Shares’, Journal of Financial Economics, Vol. 72, pp. 51953.
  • Doidge, C., A. Karolyi and R. Stulz (2004), ‘Why are Foreign Firms Listed in the U.S. Worth More?’, Journal of Financial Economics, Vol. 71, pp. 20538.
  • Easley, D. and M. O'Harra (2004), ‘Information and the Cost of Capital’, Journal of Finance, Vol. 59, pp. 155383.
  • Errunza, V. and E. Losq (1985), ‘International Asset Pricing Under Mild Segmentation: Theory and Test’, Journal of Finance, Vol. 40, pp. 10524.
  • Foerster, S. and G. Karolyi (1999), ‘The Effects of Market Segmentation and Investor Recognition on Asset Prices: Evidence from Foreign Stocks Listing in the United States’, The Journal of Finance, Vol. 54, pp. 9811013.
  • Hail, L. and C. Leuz (2004), ‘Cost of Capital and Cash Flow Effects of U.S. Cross-Listings’, Working Paper (ECGI).
  • Healy, P. and K. Palepu (1990), ‘Earnings and Risk Changes Surrounding Primary Stock Offers’, Journal of Accounting Research, Vol. 28, pp. 2548.
  • Huijgen, C. and M. Lubberink (2005), ‘Earnings Conservatism, Litigation and Contracting: The Case of Cross-Listed Firms’, Journal of Business Finance & Accounting, Vol. 32, pp. 1275309.
  • Jain, A. and O. Kini (1994), ‘The Post-Issue Operating Performance of IPO Firms’, Journal of Finance, Vol. 49, pp. 1699726.
  • Johnson, S., R. La-Porta, F. Lopez-de-Silanes and A. Shleifer (2000), ‘Tunneling’, American Economic Review, Vol. 90, pp. 2227.
  • Karolyi, A. (2006), ‘The World of Cross-Listings and Cross-Listings of the World: Challenging Conventional Wisdom’, Review of Finance, Vol. 10, pp. 6998.
  • La Porta, R., F. Lopez-De-Silanes, A. Shleifer and R. Vishny (1998), ‘Law and Finance’, Journal of Political Economy, Vol. 106, pp. 111355.
  • La Porta, R., F. Lopez-De-Silanes, A. Shleifer and R. Vishny (2000), ‘Investor Protection and Corporate Governance’, Journal of Financial Economics, Vol. 58, pp. 327.
  • La Porta, R., F. Lopez-De-Silanes, A. Shleifer and R. Vishny (2002), ‘Investor Protection and Corporate Valuation’, Journal Finance, Vol. 57, pp. 114770.
  • Lang, M., K. Lins and D. Miller (2003), ‘ADRs, Analysts and Accuracy: Does Cross-Listing in the U.S. Improve a Firm's Information Environment and Increase Market Value?’, Journal of Accounting Research, Vol. 41, pp. 31745.
  • Lang, M., J. Raedy and M. Yetman (2003), ‘How Representative are Cross-listed Firms? An Analysis of Firm Performance and Accounting Quality’, Journal of Accounting Research, Vol. 41, pp. 36386.
  • Lee, D. (2003), ‘Why does Shareholder Wealth Increase when Foreign Firms Announce their Listing in the U.S.?’, Working Paper ( University of Kentucky ).
  • Lombardo, D. and M. Pagano (2000a), ‘Legal Determinants of the Return on Equity’, Working Paper ( Stanford Law and Economics Olin and University of Salerno ).
  • Lombardo, D. and M. Pagano (2000b), ‘Law and Equity Markets: A Simple Model’, Working Paper ( Stanford Law and Economics Olin and University of Salerno ).
  • Loughran, T. and J. Ritter (1997), ‘The Operating Performance of Firms Conducting Equity Offerings’, Journal of Finance, Vol. 52, pp. 182350.
  • Miller, D. (1999), ‘The Market Reaction to International Cross-Listings: Evidence from Depositary Receipts’, Journal of Financial Economics, Vol. 51, pp. 10323.
  • Pagano, M., A. Roell and J. Zehner (2002), ‘The Geography of Equity Listings: Why do Companies List Abroad?’, Journal of Finance, Vol. 57, pp. 265194.
  • Papaioannou, G., N. Travlos and K. Viswanathan (2003), ‘The Operating Performance of Firms that Switch their Stock Listings’, Journal of Financial Research, Vol. 26, pp. 46986.
  • Reese, W. and M. Weisbach (2004), ‘Protection of Minority Shareholder Interests, Cross-Listings in the United States, and Subsequent Equity Offerings’, Journal of Financial Economics, Vol. 66, pp. 64104.
  • Ritter, J. (1991), ‘The Long-Run Performance of Initial Public Offerings’, Journal of Finance, Vol. 46, pp. 327.
  • Stapleton, R. and M. Subrahmanyam (1977), ‘Market Imperfections, Capital Market Equilibrium and Corporate Finance’, Journal of Finance, Vol. 32, pp. 30719.
  • Stulz, R. (1999), ‘Globalization of Equity Markets and the Cost of Capital’, Journal of Applied Corporate Finance, Vol. 12, pp. 825.