Methodology for Estimating Tax and EPS Impacts of Repatriation and Border Adjustments on S&P 500

Methodology Repatriation impact We estimated cumulative overseas profits for the S&P 500 excluding Financials and Real Estate via three sources: 1) Bloomberg data on cash held overseas (if disclosed/available}, 2) BofAML analyst estimates on cumulative overseas profits or cash, and 3) our own estimate if neither (1) or (2) are available, based on (Total cash x [% of foreign sales +10%J]) for companies with at least 5% foreign sales. We apply the effective tax rates proposed by Trump (10%) and the Blueprint (8.75%) to our estimated $1.2tn in overseas cash to determine taxes due for the S&P 500 ex. Financials & Real Estate. We estimate that 100% is brought back given that the tax is mandatory. To compute the one-time tax impact to GAAP EPS, we divide the cumulative tax impact by the S&P 500 divisor, after first excluding the impact from several large multinationals (e.g. AAPL) which already provision a portion of their overseas profits for US taxes and have effective US tax rates well above 35%. Note that for companies for which our analysts provided estimates, we asked them to provide cumulative overseas profits if possible, but in most cases this number reflects overseas cash. Thus, taxes paid could be slightly higher than we estimate given that both the Blueprint and Trump’s plans suggest a mandatory tax on all accumulated overseas profits, some of which may be permanently reinvested; here, the Blueprint suggests a lower 3.5% tax for retained earnings not held in cash/equivalents, suggesting that any additional taxes payable that we are not capturing are likely to be small. To calculate the % EPS impact from buybacks, we subtract the amount of taxes payable from total cash brought back for the S&P 500 ex. Financials & Real Estate (and for each sector) and divide this by the market cap for the S&P 500 (and for each sector.) We multiply this % impact for the overall index by our 2018E EPS of $137 to determine the potential EPS impact, applying various buyback scenarios (10-100%). We use 50% as a base case scenario, which is lower than the 80% brought back during the 2004 tax holiday, to be conservative. Border adjustments impact on EPS We estimate the costs of goods imported and exported for each company using the latest company filings, conversations with analysts, industry research, management commentary and the Input-Output accounts data published by the U.S. Department of Commerce. For more information about the Input-Output accounts data, please refer to the Bureau of Economic Analysis website at http://www.bea.gov/industry/io_annual.htm. When using the estimates based on Input-Output accounts, we adjust them to account for the varying foreign exposures of different S&P 500 industries. For companies that source products through importers (e.g. retailers), we only included the costs of goods estimated to be directly imported by each company, as taxes pertaining to those imported goods should be paid by the importers themselves. We estimate the earnings impact of border adjustments based on the additional taxes that companies would pay on the costs of imported goods sold in the US (non- deductible) and any reduction in taxes related to the production costs for exported goods (deductible). To calculate this, we multiple the net value of imported COGS by the assumed tax rate. Note that the net impact can be either positive or negative depending on whether the company is a net importer or exporter. We assume a 50% haircut to the impact to account for from alternate sourcing, currency rates and pricing power. Lower corporate tax rate impact on EPS We estimate the normalized effective domestic tax rate of a company based on conversations with the analysts or the median 5-year domestic income tax rate. If a company had a negative tax rate for a particular year, we exclude the tax rate of that BankofAmerica <2” 24 Equity Strategy Focus Point | 29 January 2017 Merrill Lynch HOUSE_OVERSIGHT_023092