Sep 17, 2008
SSE.LT Research: Corporate Governance and Information Disclosure in Baltics
Does corporate governance matter? Here, in Baltics? Particularly, do listed companies of Estonia, Latvia and Lithuania can increase their share prices through better corporate information disclosure practices? These questions were raised last spring in Stockholm School of Economics in Riga with help of Riga Stock Exchange by my partner Inga Zarecka, currently a student of Heriot-Watt University of Edinburgh, and me in Bachelor thesis writing process. Our intention was to evaluate numerically the effect of corporate governance on stock returns of listed companies in Baltic States.
Apart from identification of general links between information disclosure and corporations valuation, one can find stated questions especially interesting under current poor economic environment of the region. Good knowledge about companies is the first step to attract foreign capital to the countries, which is so required today for Baltic economies. But results we have found during the study process are not very promising. As could be expected more transparent companies might gain benefit which is greater than costs of preparing report which is expressed in excess stock returns. However, the opposite effect was observed in empirical analysis of the paper. The group of companies with poor reporting standards significantly outperform the companies with high quality corporate information disclosure.
Research Question
We were interested whether the companies of Baltic stock exchanges gain an extra advantage in their share price values through “best practice” reporting. In order to answer this question the authors investigated whether investors earn positive returns on a long “high information disclosure level” portfolio and a short “low information disclosure” portfolio.
Theoretical Background
There exist several authors who researched the relationship between corporate governance and market returns in different countries. You can skip the section if you are not interested in information disclosure emperical analysis development.
Drobetz et al. (”Corporate Governance and Expected Stock Returns: Evidence from Germany”, February 2003) find empirical evidence of their hypotheses that better governed corporations are associated with higher returns in Germany. Longing good governance firms and shorting bad governance firms would result a 12% annual return to its investors. From the asset pricing perspective corporate governance aspects have significant role, since it implies additional risks of failure and investors require additional returns for that, according to Drobetz et al. conclusions.
Russia may be an extreme case for measuring the corporate governance effect because of generally low quality of governing standards and practice. It leaves a great room for variations in attracting financial resources in the stock market – either by employing good corporate governance practice, or simply looting the firm. Black (”Does corporate governance matter? A crude test using Russian Data”, June 2001) in his study of Russian market conducts a crude analysis of 16 largest listed companies using the basic data on its capitalization levels and corporate governance ranking. The correlation measured by Black is substantially high – 90% with t-statistic 7,63.
Bauer et al (”Empirical evidence on corporate governance in Europe: The effect on stock returns, firm value and performance”, September 2003) has studied the effect of corporate governance on stock returns, firm value and performance in Europe. They all note the similar conclusion: better corporate governance practice has lesser impact on stock returns in developed economies and is more significant in emerging markets. In other words – “the lower governance standards, the stronger is the relationship between governance and firm value”.
Gompers et al. (”Corporate governance and equity prices”, 2003) created an index of corporate governance according to companies’ shareholders and management power. Then, after sorting each company with respect to the index, the author creates several portfolios: democracy and dictatorship as a base correspondingly with good and bad corporate governance practices. The author compares two portfolios and makes a conclusion that the democracy portfolio outperformed the dictatorship one by higher returns of 0,7% per month what is 8,5% per year.
Anete Pajuste’s study “Corporate Governance and Stock Market Performance in Central and Eastern Europe” (2002) makes a valuable contribution to the research topic. She has done a solid analysis of the stock market developments in CEE countries over the period of 1994 – 2001. One important conclusion of Pajuste’s research in that law enforcement quality plays a greater role in market evolution than the quality of law itself. Moreover, she conducts risk-return analysis and discovers that due to high ownership concentration and low transparency, Baltic stock markets showed a negative dependency between risk and return during the studied period: with higher risks investors were obtaining lower returns.
Anete Pajuste also completed another study in cooperation with Erik Berglof (2005) “What do Firms Disclose and Why? Enforcing Corporate Strategy in Central and Eastern Europe”. In this paper the authors evaluate “to which extent rules and regulations relating to corporate governance disclosure are being implemented and enforced in individual corporations in CEE”. Pajuste and Berglof found evidence that higher voluntary disclosure is associated with a high level of financial resources. They claim that the direct costs of information disclosure outweigh the benefits of attracting minority shareholders.
OMX Baltic Viewpoint
We conducted an interview with an OMX Riga stock exchange representative which was held in order to find out what actions stock exchanges take in order to motivate companies to disclose more information. OMX notes the demand side problem in which minority investors are not actively participating in governing the company. Firstly, they might be unaware of the right to communicate with a CEO directly and discuss issues. Secondly, they might be confused and lack self-confidence in doing this.
Companies also have difficulty being open to their investors. The problem represents the supply side of the equity market. The majority of companies listed on the Baltic Stock Exchange (especially in Latvia) are there not of their own will, but are forced to be there due to a mandatory listing process during privatization. Costs of full transparency for an organization outweigh the benefits it might gain.
Intermediaries also play a role in improving information disclosure and capital market development in general. Since companies like banks or brokers directly communicate with investors and are familiar to their needs and preferences. However, as explained by OMX, they do not actively promote the possibility of raising finance at stock exchanges, rather offer bank loans.
Research Methodology
The general idea of the research is to construct three portfolios of different information disclosure index level, calculate the average returns of each weighted on the capitalization level in the group and conduct a regression with portfolio returns as a dependant variable and the Fama and French three factor model components as explanatory variables.
RP – RF= t+1 * RMRFt + 2 * SMBt + 3 * HMLt + t
Where RP – RF is a portfolio return minus risk-free return of the market, RMRF – market premium rate, calculated as weighted average of all stocks present on the stock exchanges at the time less one-month risk free rate; SMB – return on portfolio of companies with small capitalization minus the return on portfolio of companies with big capitalization; HML – return on portfolio of companies with high book-to-market ratio minus the corresponding with low book-to-market ratio and t, is a measure of randomness. t is a measure of return of an investor into the specific portfolio.
Data Sample
Our sample included 100 companies which currently are or were listed on Baltic stock exchanged during the period of June 2000 and June 2007. The list of the companies include:
| Tallinn Stock Exchange | Riga Stock Exchange | Vilnius Stock Exchange |
| Total of 18 companies | Total of 39 companies | Total of 43 companies |
| 1.Baltika (BLT1T) | 1.Latvijas Balzams (BAL1R) | 1.Alita (ALT1L) |
| 2.Eesti Ehitus (EEH1T) | 2.Balozi (BLZ1R) | 2.Anyksciu vynas (ANK1L) |
| 3.Estiko (ESOET) | 3.Brivais Vilnis (BRV1R) | 3.Apranga (APG1L) |
| 4.Eesti Telekom (ETLAT) | 4.Ditton Pievadkezu Rupnica (DPK1R) | 4.Alytaus Tekstile (ATK1L) |
| 5.Harju Elekter (HAU1T) | 5.Rigas Farmaceitiska Fabrika (FRM1R) | 5.Dvarcioniu Keramika (DKR1L) |
| 6.Hansapank (HPA1T) | 6.Grindeks (GRD1R) | 6.Grigiskes (GRG1L) |
| 7.Kalev (KLV1T) | 7.Grobina (GRZ1R) | 7.Gubernija (GUB1L) |
| 8.Merko Ehitus (MKO1T) | 8.Latvijas Gaze (GZE1R) | 8.Invalda (IVL1L) |
| 9.Norma (NRM1T) | 9.Kurzemes Atslega (KA11R) | 9.Klaipedos Baldai (KBL1L) |
| 10.Rakvere Lihakombinaat (RLK1) | 10.Kurzemes CMAS (KCM1R) | 10.Klaipedos Juru Kroviniu Kompanija (KJK1L) |
| 11.Saku Olletehas (SKU1T) | 11.Kvadra Pak (KVD1R) | 11.Klaipedos Nafta (KNF1L) |
| 12.Starman (SMN1T) | 12.Liepajas Autobusu Parks (LAP1R) | 12.Kauno Energija (KNR1L) |
| 13.Sampo Pank (SPO1T) | 13.Latvijas Juras Medicinas Centrs (LJM1R) | 13.Kauno Tiekimas (KTK1L) |
| 14.Tallink Grupp (TAL1T) | 14.Latvijas Krajbanka (LKB1R) | 14.DFDS Lisco (LBS1L) |
| 15.Tallinna Farmaatsiatehase (TFA1T) | 15.Liepajas Metalurgs (LME1R) | 15.Lietuvos Dujos (LDJ1L) |
| 16.Tallinna kaubamaja (TKM1T) | 16.Lode (LOD1R) | 16.Lietuvos Elektrine (LEL1L) |
| 17.Tallinna Vesi (TVEAT) | 17.Daugavpils Lokomotivu Remonta Rupnica (LOK1R) | 17.Lietuvos Energija (LEN1L) |
| 18.Viisnurk (VNU1T) | 18.Latvijas Kugnieciba (LSC1R) | 18.Lifosa (LFO1L) |
| 19.Latvijas Tilti (LTT1R) | 19.Lietuvos Juru Laivininkyste (LJL1L) | |
| 20.Nordeka (NKA1R) | 20.Limarko Laivininkystes kompanija (LLK1L) | |
| 21.DnB Nord Banka (NLB1R) | 21.Linas (LNS1L) | |
| 22.Olainfarm (OLF1R) | 22.Mazeikiu Nafta (MNF1L) | |
| 23.Rigas Autoelektroaparatu Rupnica (RAR1R) | 23.Mazeikiu Elektrine (MZE1L) | |
| 24.Rigas Elektromasinbuves Rupnica (RER1R) | 24.DnB Nord Bankas (NDL1L) | |
| 25.Rigas Juvelierizsradajamu Rupnica (RJR1R) | 25.Pramprojektas (PRM1L) | |
| 26.Rigas Kugu Buvetava (RKB1R) | 26.Panevezio Statybos Trestas (PTR1L) | |
| 27.Rigas Raugs (RRA1R) | 27.Pieno Zvaizgdes (PZV1L) | |
| 28.VEF Radiotehnika RRR (RRR1R) | 28.Rytu Skirstomieji Tinklai (RST1L) | |
| 29.Rigas Starptaustiska Autoosta (RSA1R) | 29.Rokiskio Suris (RSU1L) | |
| 30.Rigas Transporta Flote (RTFL) | 30.Sialiu Bankas (SAB1L) | |
| 31.SAF Tehnika (SAF1R) | 31.Sanitas (SAN1L) | |
| 32.Siguldas CMAS (SCM1R) | 32.Snaige (SNG1L) | |
| 33.Saldus Mezrupnieciba (SMA1R) | 33.Snoras (SRS1L) | |
| 34.Strencu MRS (SMR1R) | 34.Stumbras (STU1L) | |
| 35.Tosmares Kugubuvetava (TKB1R) | 35.TEO it AB (TEO1L) | |
| 36.Talsu Mezrupnieciba (TMA1R) | 36.Ukio Bankas (UKB1L) | |
| 37.VEF (VEF1R) | 37.Utenos Trikotazas (UTR1L) | |
| 38.Ventspils Nafta (VNF1R) | 38.Vilniaus Baldai (VBL1L) | |
| 39.Valmieras Stikla Skiedra (VSS1R) | 39.Vilniaus Degtine (VDG1L) | |
| 40.Vilkyskiu Pienine (VLP1L) | ||
| 41.Vilniaus Vingis (VNG1L) | ||
| 42.VST (VST1L) | ||
| 43.Zemaitijos Pienas (ZMP1L) |
Information Disclosure Index
We evaluated the quality of information disseminated by the listed firms in their quarterly, semi-annual and annual reports as well as news posted on their websites and OMX announcement pages. The information disclosure index depicts value of disclosed corporate information. The full questionnaire developed by OMX was used for this purpose. The maximum of 55 points could be collected by a company with 25 maximum possible assigned to annual reports, 20 assigned to periodic reports and only 10 assigned to corporate announcements available in press.
After estimating information disclosure index for every company, the weighted average returns were calculated per each of three portfolios: with high information disclosure level, with intermediate information disclosure level and with low information disclosure index. Historical distribution is the following.
| June 2001 | June 2002 | June 2003 | June 2004 | June 2005 | June 2006 | June 2007 | |
| Low ID | 5 (18%) | 15 (38%) | 28 (47%) | 31 (42%) | 25 (29%) | 13 (14%) | 18 (20%) |
| Medium ID | 18 (64%) | 16 (41%) | 19 (32%) | 33 (45%) | 46 (54%) | 61 (66%) | 50 (56%) |
| High ID | 5 (18%) | 8 (21%) | 13 (21%) | 10 (13%) | 14 (17%) | 19 (20%) | 21 (24%) |
| Total | 28 | 39 | 60 | 74 | 85 | 93 | 89 |
Findings
Daily changes of each portfolio accumulated into seven year investment strategy were modified in order to see the exact historical returns of three portfolios. Three blue lines depict return on portfolios of companies with High, Medium and Low information disclosure quality measured as indices. Grey area demonstrates the return of the whole market of three weighted Baltic stock exchanges. As seen from the graph, if one invested €100 in June 2000 into market portfolio distributing share purchases proportionally to market capitalization of each stock in this portfolio, he would withdraw €161 in June 2007. If invested the same amount in portfolio of companies with High information disclose, the outcome would be €333; for Medium disclosure – €375; Low disclosure – €654, resulting in the highest return.
Collected indices revealed that historically Estonian companies outperformed companies form Latvia and Lithuania in terms of information disclosure quality
Another interesting thing to discuss is composition specifics of three portfolios. In particular, this is a relationship between average historical book-to-market ratios of a company with its average historical information disclosure indices and average capitalization of each portfolio. Firstly, we see on the graph that firms with relatively higher information disclosure indices, or those which published more comprehensive reports comparing to the rest, usually have book-to-market ratios less than one. It means that market capitalization of such companies exceeds accounting values of their equity, thus stock prices are already high and do not attract many speculators. On the contrary, there is a larger number of firms with relatively lower information disclosure indices, and book-to-market ratios of such firms far exceeding one. It is logical, if a book value of one share is 2 or more times larger than market value of the same share, it might be underpriced in most cases. Consequently, rational investors will buy more and more of such stock unless it reaches its optimal book-to-market ratio. As a result, share prices rise.
As can be seen in the table the majority of statistically significant coefficients were obtained in the High information disclosure portfolio. An excess monthly return of 1,34% can be earned when investing in this portfolio. While the same amount of investment into Low information index portfolio would bring a 3.28% monthly return (at 5% significance level). We also see that Medium information disclosure excess returns is located between neighbour portfolios with intercept equal to 1.54%. The risk of each portfolio measured by market beta has also been distributed according to the values of information disclosure indices: High ID being the least risky, Low ID – the most and Medium ID located between the two. Both these indicators imply that we should expect the highest returns in terms of market related return (measured by market beta) and market excess return (measured by an intercept) on Low ID portfolio and the lowest on High ID portfolio. The difference between the two returns is statistically significant at 88% confidence interval.
| Alphat | RMRFt | SMBt | HMLt | R-squared | |
| -0.0194 | -0.2024 | 0.7941 | -0.2522 | ||
| High ID – Low ID | (-1.58) | (-0.87) | (1.40) | (-1.11) | 13.3% |
| 0.0134* | 0.3579** | -0.3693** | -0.1794 | ||
| High ID excess returns | (2.39) | (3.60) | (-3.93) | (-1.86) | 22.01% |
| 0.0154* | 0.3989** | -0.4793** | -0.0521 | ||
| Medium ID excess returns | (2.35) | (3.29) | (-3.75) | (-0.42) | 22.79% |
| 0.0328* | 0.5602* | -1.1634 | 0.0728 | ||
| Low ID excess returns | (2.53) | (2.19) | (-1.95) | (0.28) | 22.22% |
Conclusion
The obtained results were the opposite of the expectations prior to conducting a research. Despite all portfolios outperformed the weighted market index, the one consisting of companies with low-quality disclosure proved to be the most profitable. So, if one invested in Low information disclosure portfolio in June 2000 and closed the position in June 2007, he would obtain a return of 554%, implying 30.8% of annual growth. “Amazing, but true”, one would say.
All in all, at the first glance the answer to the question raised in the name of this thesis “To Read or not to Read an Annual Report?” would seem rather obvious and is “No, do not read”. However, as was noted in previous studies done on this topic and mentioned in the interview with A. Pajuste and L. Dubava, there is a lack of active non-institutional investors who are familiar with norms and regulations and who do not hesitate to push on companies with respect to their obligations. In other words, the result that we obtained is a direct outcome of developing, immature financial market. And as more investors are interested in corporate disclosure and companies are willing to disclosure, more mature and significant markets will appear. So it would be more appropriate to answer “Yes, you should read an annual report!” This would take effect in the long run and the situation with information disclosure would converge with more developed countries.
Rate opinion
