No alpha from repatriation and multiples actually compressed Our analysis of the top 15 repatriating companies (from Table 7) suggests that while these companies initially outperformed both the S&P 500 and equal-weighted S&P 500 from November 2004-April 2005 (by 6ppt and 5ppt, respectively), they subsequently underperformed during the remainder of 2005 and early 2006 (Chart 6). From the end of September 2004 through year-end 2006, these stocks were up 27% on average, in-line with the overall S&P 500, and below the equal-weighted benchmark’s 36% return. And multiples for these stocks compressed over the majority of this period, both on an absolute basis and relative to the benchmark (Chart 7). Chart 6: Cumulative relative performance (equal-weighted) vs. S&P 500 Chart 7: Fwd. P/E of Top 15 repatriating companies — absolute and and EW S&P 500 of Top 15 repatriating companies, 9/30/04-12/31/06 relative to the S&P 500 median fwd. P/E, 9/30/04-12/3 1/06 106.0 23.0 1.50 104.0 22.0 102.0 a5 unD 100.0 ‘ 1.30 20.0 28.0 1.20 96.0 19.0 94.0 18.0 , 1.10 92.0 : 170 1.00 90.0 16.0 0.90 eeseceeeeeee2e2eeese 86.0 o 6 &§ SS FS R6E ESTERS SSS SESESESSSSSESESESESSSESESES “we oS = “we oS = wi ie SBSSGSFSGRESSSOABSSSOSCSEE SSS OASES BOOZE SPSLES>FZZOZS SP SLES FP ZGOZS —— Top 15 Fwd P/E (LHS) Source: U.S. Senate Permanent Subcommittee on Investigations survey (for Top 15 repatriating Source: U.S. Senate Permanent Subcommittee on Investigations survey (for Top 15 repatriating stacks), FactSet, Bloomberg, BofA Merrill Lynch US Equity & US Quant Strategy stacks), FactSet, BofA Merrill Lynch US Equity & US Quant Strategy Post-repatriation cash use: will this time be different? Valuations, Investor preference, growth & leverage ratios suggest less buybacks While any potential restrictions on the use of repatriated earnings are still unknown, we suspect that a pick-up in buybacks is likely, but that a lower proportion will be used for buybacks today than during the last repatriation holiday.