The Value-Relevance of the differences between IAS and US GAAP

Introduction

This research was done by Veikko Fredrick Valli a student at the Swedish School of Economics and Business Administration. Department of Finance and Statistics. The research topic was "The value relevance of the differences between International Accounting Standards (IAS) and US GAAP(Generally Accepted Accounting Principles). The research objective was to examine if there are value-relevant differences between GAAP and IAS when applied to non US companies that provided 20-F filling to Securities and Exchange Commission (SEC) containing reconciliation for earnings and book values from IAS to US GAAP.

The association of these sets of earnings and book values according to IAS and US GAAP were examined in this study

The research objective

The objective of this study was to answer whether there are value-relevant differences between IAS and US GAAP after the IASC comparability and Improvements Project. This study does not answer questions relating to the information content of the 20-F fillings other than the reconciliations form IAS to US GAAP of earnings and book value.

The thesis structure

This thesis started with introduction giving a brief description on issues that led to the research question and later stated the research objective and the limitation of the research. Section 2, which went through the framework of financial statements.The accounting differences between IAS and US GAAP are presented in section 3, valuation of accounting differences are discussed in section 4, previous researches done are presented on section 5, a description of sample companies and data used in the study are provided in section 6. The methodology used to conduct the research is presented in section 7 and section 8 discusses the results. A summary concluding the findings of the research is in section 9

The Empirical Study

Methodology

The research hypotheses and regression analysis were used to study the value relevance of IAS and US-GAAP. The ordinary squares regression method was used in all of the regressions. Incremental and differential tests of value-relevance was also involved. Two main hypotheses were examined in this study

H0= The difference between IAS and US GAAP is not value relevant

H1= The difference between IAS and US GAAP is value relevant

The two regressions were used to test the value relevance of IAS and US GAAP measures. The first is un-deflated value regression and the second is a price per share regression where the variables are deflated by the number of shares

Data

22 sample companies were chosen and data obtained from them. Companies publishing 20-F forms were searched on NYSE and NASDAQ stock exchange lists of listed foreign companies. The search was limited to those countries that allow IAS statements according to IASC status report for different countries. The accounting standards used were identified by going through annual reports and the reconciliation data for the companies was found from their published 20-F documents or their annual reports. The sample contains companies whose fiscal year does not coincide with the calender year.

Bias in the sample

  • The sample does not include companies opted to change to another accounting standard other than IAS
  • The sample does not include companies using IAS that have gone bankrupt
  • 10 of the 22 companies in the sample are from China, differences in the companies operating environment may affect the accounting numbers.

    • Descriptive Statistics for Reconciliation

    The following descriptive statistics were used for reconciliation.
  • IAS earnings are on average slightly larger than US GAAP earnings
  • US GAAP book value is on average 3.94% lower than IAS

    • Results

    The results presented are from the regressions and there are discussions about autocorrelation, hetoroskedasticity and multicollinearity as possible violation and disturbances in the regressions. The US GAAP reconciliations did not seem to provide information beyond IAS amounts and were neither valued differently by investors. The differences between the accounting standards do not seem to be value-relevant.

    Contribution of the thesis

    This thesis provided a timely evidence on the value-relevance of the US GAAP reconciliations. The market value and prices per share regressions used in this study have been advocated as preferable in previous value-relevance and methodological researches.

    Comparison to previous studies

    This study is some what different from previous studies.Harris & Muller found value relevant differences between IAS and US GAAP during 1992 - 1996. From 1995 -1996 the results were similar to this research results

    Suggestions for further research

    Possible improvement would have been to add an event study based on the filling deadline of the 20-F documents or the annual report containing US GAAP reconciliations of companies using IAS