(Vo.12,No,23) ANALYSIS Executive Summary a Analysis compares pan-European results to country-specific results. This paper presents original Deloitte Touche re- search analyzing the issue whether pan-European comparables sets provide statistically different arms-length ranges of results compared to an arms-length range of results formed by European country-specific comparable companies a widely accepted statistical methods used. The approach is scientific in that the paper in our analysis. Arm'slength ranges of comparable companies are generated for al teet epted statistical Economic Cooperation and Development transfer pricing principles, conservative statistIc to test the reasonableness of these assumptions. The approach assumptions and statistical a Economlc evidence indicates that European markets are integrating or have integrated. As a prelude to the statistical analysis, the authors briefly review whether the critical legal factors that allow undistorted economic activities between he European states'markets are already in place, so that it is reasonable to test whether Europe is an integrated market for transfer pricing purposes. Further, economic studies that test the results of these European Union initiatives are re- viewed as to whether EU product and financial markets have converged in prices, intra-EU trade, profits, etc. The evidence demonstrates that the most significant market impediments among EU member states have been abolished, and that the foundation for a highly integrated European market is already in place the economic evidence indicates that Eu- ropean markets are integrating (or have integrated)and that in many ways the levels of price or profit differences among European countries are not meaningfully different from the differences found in integrated markets, such as japan or the United States B Economic evidence indicates that European markets are integrating or have integrated As a prelude to the statistical analysis, the authors briefly review whether the critical legal factors that allow undistorted economic activities betw for transfer pricing purposes. Further, economic studies that test the results of these European Union initiatives are re- viewed as to whether EU product and financial markets have converged in prices, intra-EU trade, profits, etc. The legal gnificant market impediments among EU member states have been abolished, and ed European market is already in place the economic evidence indicates that Eu- opean markets are integrating nd that in many ways the levels of price or profit differences amon European countries are not fferences found in integrated markets, such as Japan or the United states Two separate sets of tests. The paper's statistical analysis comprises two separate series of tests covering nine indus- is per formed to develop pan-European and country-specific arms-length ranges i tests, a detailed comparability analysis comparability analysis practice. The second series of tests takes a broader view of comparability, in that the arms-length ges are calculated from companies identified as comparable based principally on industry classification codes n Results clearly show that European arms-length ranges do not statistically differ from country-specif ranges. The results of the statistical analysis under both series of tests clearly show that European arm's-I do not statistically differ from country-specific arm'slength ranges in almost all cases. Specifically, out of ed testing the statistical equality of upper and lower quartiles of arms-length ranges using 95 per m s, 219 tests( approximately 94 percent of the tests) generate resuits supporting the equality of interquartile In other words, it is highly likely that a country-specific comparability analysis and a pan-European comparability would result in interquartile arm's-length ranges of results that were not statistically different at a 95 percent a When differences occur no pattern is evident. Further, when the country-specific arms-length range was statistically different from the pan-European arms-length range, there was no obvious bias or pattern of profit levels indicating that a particular European country's arm' s-length range of results is always statistic ally different from the rest of Europe. m Results clearly indicate that pan-European comparability searches produce reliable arm' s-length ranges. Pan-European comparability analysis may not be appropriate in all situations: the specific facts and circumstances in each case should be assessed in making this determination. However, the papers results clearly indicate that pan-European comparability searches do produce reliable arm'slength ranges of resuits relative to country-specific arm's-leng ges. Thus, the use of pan- European studies should generally be promoted, because they generate reliable resuits, as indicated by this study 41404 Copyright 2004 TAX MANAGEMENT NC,. a subsidiary of The Bureau of Nats cnai Affairs, inc. TMTR ISSN 1063-2069