This research investigates the information content of the translation information resulting from exchange rate fluctuations. Two hypotheses are examined. The dollar movement hypotheses investigate whether there is a positive relationship between security valuation and the translation information and whether the market assigns different weights to translation gains and losses in both the depreciating and appreciating exchange rate environments. The geographic concentration hypothesis tests whether the market's response to the translation information is geographically sensitive. Prior research on SFAS No. 8 and SFAS No. 52 has concentrated on the price and trading volume responses to the deliberations and issuance of these two accounting statements. Soo and Soo (1994) examine the long-term effect of the disclosure requirement under SFAS No. 52 on MNEs' security prices from 1981 to 1987. However, they fail to address two important issues pertinent to the MNE research--the effects of exchange rate changes and the geographic concentration. The dollar movement hypotheses provide strong evidence that under both the appreciating and depreciating exchange rate environments, a positive relationship exists between security returns and the translation information when MNEs disclose translation losses in stockholders' equity. The findings also provide evidence for a positive or at least non-negative relationship between security returns and the translation information when MNEs disclose translation gains. The findings provide evidence that the positive relationship is greater in appreciating than in depreciating exchange rate environment for losses, but no evidence of such a difference exists for gains. The evidence also indicates that the market reacts more to the translation information when translation losses are reported than when translation gains are reported in both exchange rate environments. The examination of the impact of the geographic concentration of MNEs' foreign operations provides limited evidence to support the geographic concentration hypothesis. One possible explanation for the weak findings is that the larger degree of the aggregation of some of the geographic disclosures prevents the market from impounding the geographic information.

Faculty of Commerce School of Accountancy 9108665/p swartzg@soa.wits.ac.za / The debate on the determinants of firm value remains unresolved in finance research. This research report contributes to the debate by examining the validity of the Ohlson (1995) valuation model in South Africa. Using Johannesburg Securities Exchange data, this research report aims to identify whether the book value of assets, accounting (accrual) earnings, and abnormal cash dividends explain the behavior of South African share prices. The Ohlson (1995) model has been successfully tested in a number of recent studies (Collins, Pincus and Xie, 1999; Garrod and Rees, 1998; Collins, Maydew and Weiss, 1997; and Kothari and Zimmerman, 1995). This study attempts to extend this body of work in an emerging market context (South Africa), to determine whether the results obtained in developed markets also hold in an emerging market setting, where required rates of return are higher, liquidity is low, and capital is scarce. The research uses both cross sectional and panel data for 129 Johannesburg Securities Exchange listed companies over the 1992-2003 period to investigate the value relevance of the annual financial statements using the Ohlson (1995) model. Using cross-sectional data, the study indicates that the Ohlson (1995) model cannot be used for value prediction purposes, but does indicate that accrual accounting data is value relevant. However, using a panel data approach resulted in a statistically significant, robust, positive relationship for accounting earnings, book value of assets, and abnormal dividends in predicting firm value.