Submitted by Maykon Nascimento (maykon.albani@hotmail.com) on 2015-12-14T18:37:06Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Essays on life cycle voluntary disclosure and the cost of capital of brazilian companies.pdf: 1743467 bytes, checksum: 506c8fe8bffb13d82c87125c7dd355fb (MD5) / Approved for entry into archive by Morgana Andrade (morgana.andrade@ufes.br) on 2016-01-07T11:31:23Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Essays on life cycle voluntary disclosure and the cost of capital of brazilian companies.pdf: 1743467 bytes, checksum: 506c8fe8bffb13d82c87125c7dd355fb (MD5) / Made available in DSpace on 2016-01-07T11:31:23Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Essays on life cycle voluntary disclosure and the cost of capital of brazilian companies.pdf: 1743467 bytes, checksum: 506c8fe8bffb13d82c87125c7dd355fb (MD5) Previous issue date: 2015 / CAPES / Esta dissertação investiga como os estágios do ciclo de vida impactam no disclosure voluntário, no custo de capital próprio e também na relação entre eles. Utilizando uma amostra de empresas não-financeiras entre 2008 e 2014, e seguindo o modelo de Gebhardt et al. (2001) para desenvolver o Custo de Capital Implícito (ICC), usando a base de dados I/B/E/S, da Thomson Reuters®, encontra-se que, em média, as empresas são beneficiadas com a redução do custo de capital por meio da melhora no nível de evidenciação voluntária. Usando a métrica de ciclo de vida de Dickinson (2001), encontra-se que o nível de disclosure voluntário cresce até atingir o estágio de maturidade e então esse nível decresce. Ainda, o ICC se difere significativamente entre os estágios de ciclo de vida. Contudo, não são encontradas diferenças significativas entre os estágios de crescimento, maturidade e shake-out, embora o nível de ICC diminua depois de sair do estágio inicial e ele cresça novamente no estágio de declínio. Finalmente, encontra-se que a melhora no nível de disclosure voluntário nos estágios de crescimento e maturidade é mais bem compensada com uma redução maior no nível de custo de capital. Os resultados para o disclosure voluntário são robustos usando uma série de testes de sensibilidade, muito embora os resultados não tenham se mantido para o custo de capital usando proxies diferentes. / This thesis investigates how life cycle stages impact on voluntary disclosure, cost of equity capital and also on the relationship between them. Using a sample of non-financial Brazilian listed firms covered by analysts between 2008 and 2014, and following the model by Gebhardt et al. (2001) to develop the Implied Cost of Capital (ICC) using I/B/E/S database, from Thomson Reuters®, I find that companies on average are benefitted by the reduction of the cost of capital via improvement in voluntary information level. Using Dickinson’s (2001) life cycle measure, I find that voluntary disclosure level grows until reaching the maturity phase and then it declines. Moreover, the ICC significantly differs across life cycle stages. However, I do not find differences among growth, maturity and shake-out, although the level of the ICC fades out after leaving the initial stage and it increases again in decline stage. Finally, I find that the improvement of voluntary disclosure level in growth and maturity stages is better compensated with more reduction in cost of capital. The results for voluntary disclosure are shown to be robust using a series of sensitivity tests, however the expectation about cost of capital did not hold using different proxies.

Ce travail propose une analyse de l’évaluation des capitaux propres de l’entreprise obtenue des modèles d’évaluation de la rente économique. La qualité des estimations dépend de la pertinence des hypothèses et du calibrage empirique de ces modèles. Dans un premier temps, les fondements théoriques des différentes hypothèses sur la croissance et/ou la persistance de la rente économique et de son influence sur l’évaluation des fonds propres sont analysés. S’appuyant sur le modèle des accroissements anormaux du résultat, nous montrons que la rente économique du nouvel investissement ou de l’accroissement anormal du résultat devrait converger vers zéro du fait des forces concurrentielles. Cette proposition, différente de ce qui est couramment admis suite au travail d’Ohlson et Juettner-Nauroth, apparaît influencer sensiblement l’estimation des fonds propres de l’entreprise. Une analyse asymptotique des multiples présentés dans le deuxième chapitre de cette thèse souligne ces aspects en dehors des biais et des difficultés que pourraient amener les mesures empiriques du phénomène. Le travail propose notamment une explication de la surévaluation systématique des fonds propres obtenue par le modèle des accroissements anormaux du résultat. Par la suite, une méthode d’estimation de la persistance de l’accroissement anormal du résultat par firme est proposé afin d’améliorer le calibrage de ce modèle. Dans le dernier chapitre, un modèle analytique synthétisant les deux grands types d’approche concernant la modélisation de la rente, le modèle des résultats résiduels et le modèle des accroissements anormaux du résultat, est proposé. Une équation de valorisation linéaire en est dérivée et est mobilisée afin de caractériser le contenu informationnel des prévisions de la variation du bénéfice à court terme avancées par les analystes financiers. / This research proposes an analysis of equity valuation of firm, obtained from the models valuing economic rents. The quality of the estimations depends on the pertinence of the hypothesis and the empirical calibration of these models. In the first place, the theoretical foundations of different hypotheses on the growth rate and/or the persistence level of economic rent and the influence of these hypotheses on equity valuation are analyzed. Through the abnormal earnings growth model, we show that the economic rent of the new investment or the abnormal earnings growth should converge towards zero due to the market competition. This proposition, different from Ohlson and Juettner-Nauroth’s assumption often adopted in prior studies, is found having sensitive influence on equity valuation. The asymptotic analysis of valuation ratios in the second chapter of this thesis underlines these aspects while being immune from the bias and the difficulties that the empirical studies on this issue could induce. This research notably proposes an explanation for the problem of systematic overvaluation of equity by the abnormal earnings growth model. In the second place, to improve the calibration of the abnormal earnings growth model, a method is proposed to estimate the firm-specific persistence level of the abnormal earnings growth. In the last chapter, an analytical model is proposed to synthesize the two main approaches concerning rent modeling: the residual income valuation model and the abnormal earnings growth model. A linear valuation equation is derived from the analytical model and mobilized to characterize the value relevance of the short-term earnings variation forecasted by financial analysts.