For a long time we have been long-term copper bulls. In theory one might imagine thatdon't believe that a 50% tariff on copper is a logical policy, and we don't think it will beimplemented. Our logic follows.Trump has declared massive tariffs before and then not implemented themThe most recent announcement was not an official document or even an executiveorder but rather a reply to a question (US Copper Prices Surge to Record as Trump Callsfor 50% Tariff:"I believe the tariff on copper, we're going to make it 50%," Trump said inresponse to a reporter's question during a Cabinet meeting.... Commerce Secretary HowardLutnick said the copper tariffs will be put in place later in July, or in August.).The US relies on imports for half its copper consumption (Exhibit 1).90+% of imported copper is from Chile, Canada and Peru (Exhibit 2).A 50% tariff on 900 kt of imported copper (nearly a million tons) at a global price of$10,000/t is $4.5 billion added to the purchase price.TheUSispme to only two primary smelters (Exhibit 3)Thgcost of building a smelter is perhaps $6 bln per million tons capacity (using FCXeGresik as an example). The timeline of building a smelter from scratch is perhaps 5 years.Globally, smelters are oversupplied, and smelter economics are terrible (negativetreatment charges/refining charges) [Exhibits 4 & 5].operational during a Trump presidency with poor marginsTherefore, the tariff incents no proper economic action but rather simply adds costto US manufacturers.Therefore, we think logic ultimately prevails, and the policy is radically transformedCanada and Peru from the tariffs solves the problem)In the meantime, we also highlight that ever decreasing volumes from the LME CopperexhangeintotheUSriskspoorpricediscoveryandapocopper prices. [Exhibits 6-8].If we are correct, once the market understands that no significant copper tariffs will beimplemented, we should see softening of copper price (from high levels on the LME andrecords levels on COMEX). In such a scenario, copper pure plays like ANTO and FCXwould soften (and FcX more given it benefits from copper tariffs in theory given its USsupply of SX-EW copper and smelted copper). eonLMEwww.bernsteinresearch.com The US imports half its copper.EXHIBIT 1: The US relies on imports for half its copperPrimaryefinedPeriodcopperproductionDecemberJanuary-DecemberJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberJanuary-Decemberas % of 2024Source: USGS; Bloomberg; Bernstein analysis...froma handful of countriesEXHIBIT 2: 90+% of imported copper comes from Chile, Canada and PeruRefined copper imports by countryCountry20232024Chile531,000650,000Canada128,000139,000Peru79,50062,100Congo (DRC)11,80031,600Mexico14,00017,300Zambia2,0402,760Other262,330Japan1,8801,520Germany2,240716Belgium0363China46A91Finland4议口27South Korea57United Kingdom012France56Italy0Total771,000908,000Source: USGS; ICSG; Bloomberg; Bernstein analysis...with only two domestic smelters...GLOBAL METALS & MINING EXHIBIT 3: The US is home to only two primary smeltersCOPPER(Data in thousand metric tons, copper content, unless otherwise specified)Domestic Production and Use: In 2024, the recoverable copper content of U.S. mine production was an estimated1.1 million tons, a decrease of 3% from that in 2023, and was valued at an estimated s10 billion, slightly greater thandomestic output; copper was also mined in Michigan, Missouri, Montana, Nevada, New Mexico, and Utah. Copper.was recovered or processed at 25 mines (17 of which accounted for more than 99% of mine production),refineries. A new secondary smelter and secondary refinery were expected to start up by yearend. Refined copperand scrap were consumed at about 30 brass mills, 14 rod mills, and several hundred foundries and miscellaneousbuilding construction, 42%; electrical and electronic products, 23%; transportation equipment, 18%; consumer andgeneral products, 10%; and industrial machinery and equipment, 7%.Salient Statistics-United States:Production:Mine, recoverableRefinery:Primary (from ore)Secondary (from scrap)Copper recovered from old (post-consumer) scraptImports for consumption:Ore and concentratesRefinedExports:Ore and concentratesRefinedConsumption:Reported, refined copperApparent, primary refined copper and copper from old scrap3Price, annual average, cents per pound:U.S. producer, cathode (COMEX + premium)COMEX, high-grade, first positionLondon Metal Exchange, grade A, cashStocks, refined, held by U.S. producers, consumers, and metalexchanges, yearendEmployment, mine and plant, numberNet import reliance4 as a percentage of apparent consumptionSource: USGS; Bernstein analysisSmelting is a bad business. Who would build one in the 'fog of war'?GLOBAL METALS & MINING 2020202120221,2001,2301,230872922930434940161169152211126769197323833443514148271,6801,7501,7201,6601,9601,820286.7432.3410.8279.9424.3400.7279.8422.5399.81181178411,00011,40012,000384441BERNSTEIN SOCIETE CENERALE CROUP 2 primary2023202