This paper studies the quality of mandatory financial forecasts which announced by Taiwan lists and OTC companies ¡]hereinafter called ¡§the companies¡¨¡^. Does outside financial analyst can offer the public users more useful financial forecast than the companies¡H What¡¦s the difference on precision between financial forecasts issued from the companies and outside analysts¡HBesides, this paper also tests if the companies would perform earnings management for meeting the public expectation. The empirical results indicate that for public investors, the mandatory financial forecasts issued by the companies didn¡¦t show better accrual precision than the financial forecast issued by outside analysts. There wasn¡¦t any significant difference on degree of differential between financial forecast prepared by outside analysts and the companies. The risk for using both of the financial forecasts are the same. That means the investors can rely on none of the companies and outside analysts¡¦ forecasts to do investment decision. And the companies performed earning management to meet the consistent expectation of earning arose from market which can avoid the bad result when the actual financial report differ from forecast. In summary, the mandatory financial forecasts in Taiwan can not offer timely useful information to investors. Most of them only increase the preparation cost and control cost. Some of the companies even use them as the tool to collect money from public for personal welfare. We recommend the government authorities to consider the benefit of voluntary financial forecast system in UAS when they are reviewing the mandatory financial forecast system. The voluntary financial forecast system may solve the problems of incomplete predictability of financial forecast and asymmetry of information, finally increase the usefulness of financial forecasts.

Submitted by Cristiane Shirayama (cristiane.shirayama@fgv.br) on 2011-06-03T16:25:31Z No. of bitstreams: 1 66080100248.pdf: 372702 bytes, checksum: 48ec88dac699d9c817a805f2729c2be1 (MD5) / Approved for entry into archive by Vera Lúcia Mourão(vera.mourao@fgv.br) on 2011-06-03T16:54:03Z (GMT) No. of bitstreams: 1 66080100248.pdf: 372702 bytes, checksum: 48ec88dac699d9c817a805f2729c2be1 (MD5) / Approved for entry into archive by Vera Lúcia Mourão(vera.mourao@fgv.br) on 2011-06-03T17:04:01Z (GMT) No. of bitstreams: 1 66080100248.pdf: 372702 bytes, checksum: 48ec88dac699d9c817a805f2729c2be1 (MD5) / Made available in DSpace on 2011-06-03T19:03:56Z (GMT). No. of bitstreams: 1 66080100248.pdf: 372702 bytes, checksum: 48ec88dac699d9c817a805f2729c2be1 (MD5) Previous issue date: 2011-02-07 / O objetivo desse trabalho é a comparação das previsões obtidas através da base de dados da pesquisa Focus do Banco Central do Brasil, para as Projeções de Mercado de câmbio, juros e inflação. O trabalho não visa avaliar ou julgar modelos, mas sim utilizar os mesmos como um suporte na avaliação da comparação preditiva. Desse modo procuramos avaliar e/ou comparar se os desvios entre as projeções de mercado e os indicadores efetivamente realizados são melhores do que os desvios entre um modelo econométrico e os mesmos indicadores efetivamente realizados. De um modo geral veremos que a pesquisa Focus do Banco Central do Brasil tem um poder preditivo satisfatória para todos os índices, se comparado com as previsões obtidas através dos modelos econométricos em estudo. / This work aim to compare the predictions obtained through the database of Research Focus from Central Bank of Brazil for the expectations of market exchange rates, interest rates and inflation. The paper does not intend to assess or judge models but use them as a support in the evaluation of the predictive comparison. Thus we evaluate and / or compare whether the deviations between market expectations and the indicators actually realized are better than the deviations between an econometric model and the same indicator actually realized. Generally we will see that the database of Research Focus from Central Bank of Brazil had a satisfactory predictive power for all indexes, compared with forecasts obtained from the econometric models in this study.