中文摘要：

矿业是经济社会发展的基础产业。在当前中国经济进入新常态和矿业长期低迷的形势下,要合理推进矿业供给侧改革,必须理清经济增长、城镇化与矿产资源消费之间的长期关系,需要依次回答：是否存在长期关系,这些关系有什么作用特征,这些作用的变化趋势如何,不同矿产资源是否存在显著的区别,针对这些区别如何提出应对策略。基于上述问题的判断,本文选取9种矿产资源,利用多种方法进行了组合分析。应用脱钩指数刻画了1978—2015年中国经济增长与矿产资源消费的脱钩关系,发现呈现一种周期性的脱钩和复钩过程;选取经济增长、城镇化和任意一种矿产资源消费为变量,建立了VAR模型,揭示各变量间的因果关系,将显著的因果关系分为三种类型,包括循环因果链（三种变量之间存在显著的接替性循环变化的关系,包括煤炭、锌、原铝和钾盐等）、两两互为因果（某两种变量之间存在一种或两种显著的双向交替变化关系,包括原铝、锌、钾盐、石油、精炼铅、精炼镍等）和单向因果（某两种变量之间仅存在一种显著的先后变化关系,包括铁矿石、精炼铜、精炼镍）等;运用脉冲响应函数刻画了上述关系的长期变化趋势,发现矿产资源消费的长期效应的大小及其波动频率有显著差异,大多数矿产可以持续效应10年以上（如煤炭、石油、锌、原铝等消费变动对GDP变动的影响）,少数响应仅在5—6年内显著（如铁矿石、钾盐、精炼铅等消费变动对GDP变动的影响）,之后效应会逐渐消失。基于此,结合当前实际情况,提出了中国矿业供给侧改革发展战略选择,包括以去产能为抓手、提高资源效率、坚持底线思维、建立政府适当干预下的资源价格形成机制、促进矿业经济创新等政策建议。

英文摘要：

Mineral industry is a fundamental industry in social-economic development. With the concern on the situation that China ’ s economy enters into New Normal and mineral industry suffers a prolonged slump, the supply-side structural reform in mining industry is promoted through clarifying the long-term relationship between economic growth, urbanization and the consumption of mineralresources. Some questions need to be answered sequentially does the long-term relationship exist？ Does the relationship has any interacting characteristics？ What is the changing trend that these interactions have？ Are there any significant difference between different mineral resources？ And howto propose adaptable strategies according to these differences？ Relying oquestions, this study selects nine mineral resources and does an integrated analysis by combining is used to describe the decoupling relationship between economic growth and the consumption of mineral resources during the period from 1978 to 2015 , and a periodic decoupling-recoupling process is found. Economic growth, urbanization and the consumption of one of nine mineral resources are selected as variables to reveal tlieir causal relationship through Vector Athe causal relationship can be divided into three groups： circular causality chain, where a recurrent changing relationship can be significantly^ found from three variables, including coal, zinc, aluminum and sylvite; reciprocal causation, where one or two alternate changing relationship can be significantly found from two of three variables, including zinc, aplumbum and refined nickel; one-way causation, where only one successive changing relationship can be significantly found from two ofthree variables, including iron ore, refined copper and refined nickel. The long-term trend of tliese relationshipResponse Function （IRF）, where significant differences in sizeand frequency