兩岸簽署貨幣清算協議之影響：以貿易人民幣結算為例

Abstract

隨著中國經濟持續高速成長, 國際貿易及投資快速擴展, 中國更積極推動貿\n易及投資以人民幣結算, 人民幣在兩岸、 區域甚至全球市場流通可望大幅\n增加。\n 1990年代以來, 兩岸經貿快速發展, 相形下, 雙方貨幣層面合作卻明顯\n幾經波折、 踟躕不前, 目前兩岸雖已開啟人民幣現鈔清算服務但尚未建立貨\n幣清算機制 , 限制台灣成為人民幣清算平台及後續金融影響力。\n 本文以 Devereux and Shi(2005) 為主架構, 討論兩岸簽署貨幣清算機\n制建立對貿易廠商影響, 主要發現如下:\n(1) 於國際貨幣交易需透過工具貨幣 (Vehicle Currency,VC) 現實下, 兩岸人民幣貿易結算無法完全規避匯率風險, 此與台灣中央銀行意見一致, 但當兩岸進出口貿易全數以人民幣結算時, 若台灣使用人民幣銷貨收入購買中\n國商品, 台灣對中國消費將完全不受匯率變化影響, 並非所有匯率風險皆由\n台灣承擔; 此外, 當中國完全使用人民幣支付自台灣進口商品時, 中國最適\n決策代數式和 D-S(2005) 雙邊貨幣直接兌換均衡 (BD) 部分相同, 自2012\n年6 月1 日起, 人民幣與日圓直接兌換亦正式於東京和上海外匯市場展開,\n結論因而別具政策意涵。\n(2) 數值分析發現, 當兩岸進出口貿易全數以人民幣結算時, 台灣及中國總\n消費對外國商品消費偏好非常不穩健, 對應所有θ, 中國總消費皆較台灣高,\n且隨θ上升快速增加, 由於兩岸皆較雙邊匯兌需透過美元及雙邊貨幣可直接兌換時消費更多商品, 故可知兩岸皆得利於以人民幣進行結算。 對於上述結\n論, 需要特別注意的是, 本文以包括台灣、 美國及中國, 且美國為工具貨幣\n(VC) 發行國之三國模型, 是在非常特例的情況下進行福利分析, 此時雙邊\n貨幣兌換均衡將與對稱交易均衡相同, 影響所及, 結論受校準時使用基準參\n數值影響相當大;\n(3) 本文依循 D-S(2005), 限制工具貨幣發行國美國以外各國僅能於跨期\n間持有美元, 故無法分析中國得以人民幣購買台灣商品下, 台灣累積一定金\n額人民幣存量之影響, 然為簡化分析, 排除任何調整過程, 亦為模型不合理\n處; 此外, 為簡化模型, 設定中國自台灣進口商品貨款中使用人民幣支付比\n率α為外生, 然而該比率實際上應受人民幣在岸、 離岸市場利差及匯差、 全\n球景氣變化增加持有美元意願等因素影響, 由於本文僅考慮貿易需求換匯\n未考慮套利、 套匯行為, 將留待未來進一步分析。

For the past three decades, a skyrocketing Chinese economy has supported its growing influences on international trade and investment,with Beijing’s active promotion for Renminbi investment and trade settlement, its greater circulation across the strait, in the Asia-Pacific\nregion or even globally is only to be expected.\n Compared with the significantly intensified and institutionalized cross-strait trade and economic exchanges since early 1990s, monetary cooperation across the strait has obviously stalled and progressed slowly. Currently with only a cross-strait cash settlement agreement,but not an establishment of a cross-strait currency settlement mech-\nanism clearly limits Taiwan’s prospect to become the next offshore Renminbi center and its future financial influence.\n This thesis is based on Devereux and Shi(2005) to query into the effect of a cross-strait currency settlement mechanism for firms engaged in international trade. We find that:\n\n First, under the reality that international currency trade still uti-\nlizes U.S. dollar as a vehicle, cross-strait trade settled in Renminbi\ncould not completely avoid exchange rate risks, which is in accordance\nwith the opinion from central bank, however, when cross-strait trade\nare all settled in Renminbi, Taiwan would not bears all the exchange\nrate risks if Taiwan’s imports from China were paid in its Renminbi\nsales revenue; Furthermore, when all of China’s imports from Taiwan\nare settled in Renminbi, some of the algebraic expressions representing\nChinese consumption and currency exchange decisions would coincide\nwith the bilateral deviation(BD) equilibrium in D-S(2005) entailing\nsome intriguing policy implications since starting from June 1,2012,\nyuan and yen can trade directly in Tokyo and Shanghai .\n\n Second,numerical analysis found that when cross-strait trade are\nall settled in Renminbi, though both sides’ total consumption are not\nrobust to changes in preference toward foreign goods ,θ, for all θ in the\nrelevant range , China’s total consumption is greater than Taiwan and\nthe difference is increase in θ, in addition, they both consume more\nthan when cross-strait remittance needs to be done via the U.S. dollar\nand when New Taiwanese dollar and Chinese yuan trade directly, it\nthus can be inferred that both Taiwan and China gain from Renminbi\ntrade settlement.Though cautions should be taken that a welfare anal-\nysis utilizing a three-country model consisting of Taiwan, China and the United States which further acts as the issuing country of vehicle\ncurrency is an extreme case, in that, bilateral deviation and symmet-\nrical trading equilibrium (STE) would be identical and the outcome\nhighly subjects to benchmark values used in calibration.\n\n Third,The framework is borrowed from Devereux and Shi(2005)\nwhich prohibits non-vehicle countries from holding currencies other\nthan U.S. dollar intertemporally, hence it could not provide any in-\nsights to the effects of the accumulation of a Renminbi pool in Taiwan\nunder the Renminbi trade settlement scheme, which together with the\nfact that to simplify analysis, no adjusting process is included ren-\nder the model unreasonable and unrealistic; In addition, to reduce\nmodel construction, α, the percentage of China’s imports from Tai-\nwan to be paid in Renminbi is exogenous while in reality it should\nbe endogenous and collectively determined by factors not exclusive\nto the spreads between CNY and CNH, differences in depositing and\nlending rates between the onshore and offshore market and investors’\ninclination which is affected by the outlook of the global economy to\nembrace the safe haven of U.S. dollar. As arbitrage is ruled out and\nonly currency exchanges for the purposes of trade is considered, all these drawbacks are to be improved upon in further studies.