评论
打印
President Trump’s announcement on Thursday about trade and investment interventions against China were a little less hawkish than expected, and the Chinese response was at the dovish end of the spectrum of possible retaliatory actions. If we have seen the first shots in a global trade war, then there are many rounds of escalation still to come, probably taking years to play out.
美国总统特朗普上周四宣布针对中国采取的贸易和投资干预措施的强硬程度略低于预期，而中国的反应在可能的报复行动范围内属于温和的。如果说我们见证了全球贸易战的头几枪，那么还有多轮升级尚未到来，可能需要数年时间才能结束。
Nevertheless, markets are — justifiably — taking this long-term risk far more seriously than other geopolitical problems in recent years. Investors need to judge the probability of a full-scale global trade war and the economic damage from such an outcome. Asset prices will be volatile as assessments of these two variables change.
然而，市场对待这一长期风险的重视程度远远超过对待近年来其他地缘政治问题，这是合理的。投资者需要判断全球贸易战全面打响的可能性以及可能由此带来的经济损失。随着市场对这两个变量的评估变化，资产价格将出现波动。
This column will discuss the second of these variables — the scale of the economic damage from a large increase in tariffs in the global economy.
本专栏将讨论两个变量中的第二个——全球经济在关税大幅上调情况下面临的经济损失规模。
Last week, I discussed the impact of generalised tariffs on US imports, assuming no retaliation from other trading nations. The conclusion was that a 20 per cent US tariff, imposed on imported goods from east Asia, would reduce US real gross domestic product by about 0.6 per cent.
上周，我讨论了美国对进口商品广泛征收关税的影响，前提是假设其他贸易国不采取报复措施。结论是，美国对从东亚进口的商品征收20%的关税，将使美国实际国内生产总值(GDP)减少约0.6%。
That might seem manageable, but what would happen after retaliation against US exports? And is such retaliation a rational response to President Trump’s action?
这似乎在可控范围内，但是贸易国对美国出口商品采取报复措施后会发生什么呢？此类报复措施是对特朗普举动的理性反应吗？
The first question is how the rest of the world will be affected by the initial impact of US tariffs. The IMF has published simulations on its global economic model, showing the impact of a 10 per cent extra US tariff on all imports of goods from the rest of the world. This cuts US GDP by about 1 percentage point in the long term, and it also reduces GDP in the rest of the world by 0.3 per cent. There is no transfer of output from one region to another: everyone loses.
第一个问题是，世界其他地区将如何受到美国关税最初效果的影响。国际货币基金组织(IMF)公布的全球经济模型模拟，显示了美国对从世界其他地区进口商品额外征收10%关税的影响。从长远来看，这将使美国GDP缩减约1%，并使世界其他地区GDP缩减0.3%。经济产出并不会从一个地区转移至另一个地区：每个人都面临损失。
Inevitably, the economic damage in China, the EU and other countries would result in political pressures to retaliate by imposing tariffs on US exports. These tariffs will hit some US sectors, like agriculture, very hard, inviting successive rounds of retaliation.
不可避免的是，中国、欧盟和其他国家的经济损失将引发政治压力，迫使其采取报复措施，对美国出口商品征收关税。这些关税将严重打击美国农业等行业，引起连番报复。
This raises an interesting issue. Assume that the economic impact of tariffs by other countries is basically symmetrical to the original US tariffs. Under these circumstances, the US will clearly suffer output losses from retaliation by other countries, but so too will those imposing the tariffs. Why would countries make a policy choice from which they themselves suffer economic losses?
这引出了一个有趣的问题。假设其他国家的关税措施带来的经济影响基本上与最初美国关税的影响对称。在这种情况下，美国显然会因其他国家的报复措施而遭受GDP损失，但那些征收关税的国家也会遭受损失。各国为什么要做出让自己蒙受经济损失的政策决定？
Perhaps those affected by the US tariffs, especially China, will not make a self-damaging choice, and will instead simply make the best of the new situation.
或许那些受到美国关税影响的国家——尤其是中国——不会做出自毁决定，反而只会充分利用这种新形势。
Unfortunately, it already seems that China’s leaders will decide that their overall welfare, taking account of economic costs and political benefits, is to impose retaliatory measures. It is well known that, in this game [1], akin to a classic “Prisoners’ Dilemma”, the players are likely to gravitate towards a bad “Nash” equilibrium, even when a good equilibrium is available through co-operative behaviour. Both adopt protection.
遗憾的是，事实似乎是，考虑到经济损失和政治利益，中国领导人将认定采取报复性措施才符合他们的整体利益。众所周知，在这场类似于经典“囚徒困境”的博弈中，玩家可能会倾向于达成不利的“纳什均衡”，即使玩家可以通过合作行为实现有利的均衡。双方都采取保护措施。
Let us now assume that the worst happens, and all countries impose a retaliatory 10 per cent extra tariff on all imported goods from the US. All global trade in goods would then be subject to a rise of 10 percentage points in applied tariffs. This is a fairly severe definition of an all-out “trade war”; at present, the average level of applied tariffs is only around 2 per cent in the advanced economies and 9 per cent in the emerging economies.
现在让我们设想一下最糟糕的情况出现，所有国家都对从美国进口的所有商品额外征收10%的报复性关税。全球所有商品贸易适用的关税都将提高10个百分点。这是对全面“贸易战”的相当严格的定义；目前，发达经济体适用关税的平均水平仅为2%左右，新兴经济体为9%。
Protectionism has lately fallen so far from the radar of macroeconomists that there are few published simulations of the impact of a trade war with widespread retaliation. However, the OECD published a brief account of the impact of generalised tariffs of 10 percentage points in 2016. The summary of their results is as follows:
保护主义近来已经远远超出了宏观经济学家的研究范围，目前几乎没有对涉及广泛报复措施的贸易战影响的模拟。然而，2016年经合组织(OECD)发表了对全面征收10%进口关税的影响的简述。其结论摘要如下：
The tariff increases are estimated to reduce global trade volumes by around 6 per cent, and real GDP is reduced by 1.4 per cent, with the US, China and Europe each suffering a drop in output of 1.7-2.2 per cent. Since these trade and output losses are likely to be spread over several years, the size of this shock would seem manageable, and maybe a bit less than investors currently fear.
关税上调预计将使全球贸易量缩减6%左右，实际GDP下降1.4%，其中美国、中国和欧洲均承受1.7%-2.2%的产出缩减。由于这些贸易和产出损失可能分摊到数年，这种冲击的规模似乎是可控的，可能略低于投资者目前担心的程度。
Other simulations are broadly similar. Goldman Sachs economists have published results that imply output losses of around 0.9 per cent over two years for the US and Europe in the case of a 10 percentage point increase in global tariffs, along with a 20 per cent drop in equity prices. China’s output loss is only 0.5 per cent. Inflation rises by a few tenths of a per cent, but then drops back to target as monetary policy tightens and unemployment rises. (See also this World Bank report.)
其他模拟大致相同。高盛(Goldman Sachs)经济学家公布的结果表明，在全球关税提高10个百分点的情况下，美国和欧洲在两年内的产出损失约为0.9%，同时股票价格下跌20%。中国的产出损失仅为0.5%。通胀率会 上升千分之几，但之后随着货币政策收紧以及失业率上升，通胀率将回落至目标范围。（另见世界银行(World Bank)报告。）
There are two obvious ways in which the economic impact could be much worse:
还有两种明显因素可能让经济影响严重得多：
The huge growth in global value chains early in this century (ie trade in parts and components rather than finished products) could leave the trading system vulnerable to much greater temporary disruption if there are bankruptcies and dislocations in companies within the GVCs. Tariffs could be imposed every time components criss-cross a frontier, greatly increasing the effective costs of production in GVCs.
本世纪初全球价值链（即零部件贸易而不是成品贸易）的巨幅增长可能使得贸易体系容易受到更大的暂时性混乱的冲击，比如全球价值链中的企业出现破产或紊乱。每一次当零部件跨过边境时，都可能被征收关税，大大增加全球价值链的实际生产成本。
Uncertainty about future tariffs and their effects could lead to postponement of capital investment decisions, resulting in a much larger immediate hit to GDP than shown in the above simulations.
关于未来关税及其影响的不确定性可能导致投资者推迟资本投资决策，从而造成对GDP比上述模拟所呈现的大得多的直接影响。
In summary, the available (though not very convincing) evidence suggests that the global output losses from a trade war might reach 1-3 percentage points over several years, with a left tail that could be much worse. But there are huge uncertainties, both economic and political. If this strategic game gets stuck in a bad equilibrium, then markets might become disposed to assume the worst.
总之，现有（虽然不是很有说服力）证据表明，贸易战带来的全球产出损失可能在几年内达到1%-3%，而采取左尾检定估测的情况可能会更糟糕。经济和政治上都存在着巨大的不确定性。如果这场战略博弈陷入不利的均衡状态，那么市场可能倾向于做最坏的假设。
