【英语财经】中国应从新兴市场阵营中“出列” Redefining EM: Country clusters offer new matrix

The question of what constitutes an emerging market has haunted me for decades. Since 1989, the World Bank has defined an emerging market as a country with a GDP per capita ceiling of $13,000 or less. But gross domestic product is simply a “wealth” statistic. A country with $12,999 in GDP per capita is very different from one with $1,500.

Thus a new, multi-dimensional method of classifying emerging markets is called for, one that takes into account a cluster of different indicators to produce a ranking of countries’ socio-economic maturation.

Having traded in and taught about these countries for more than 20 years, I have witnessed immense progress among many emerging markets, but often this progress has not been fully appreciated by either the providers of financial indices or investors.

I started asking myself, besides GDP, what are the other dimensions that define a country’s maturation? Are there more objective ways to analyse and cluster countries? Are not countries more nuanced than the blunt descriptions “advanced” and “emerging”? My gut feeling was yes, but I needed an objective, “big data” framework to answer these big questions.

Last year I probed these questions using Ward’s Method, a quantitative approach that clusters things based on the “minimum variance” of certain factors — that is, grouping things whose qualities were statistically most similar.

My idea was to batch 100 countries (both advanced and emerging) into 10 clusters. Group 10 would have the highest scores awarded for nine specific criteria, denoting more mature countries that may be more resilient to potential shifts and shocks. Group 1 would have the lowest scores, signifying those countries that are less mature and more vulnerable to an array of risks.

The members of each group were arrived at by assigning numerical values to each country reflecting the nine distinct categories. Thus a country with a high per capita GDP would be given a high score in the category concerned with this metric. The other eight categories include two for population size/wealth and competitiveness, three for credit ratings, stock market penetration and currency valuations, and three for heath, education and political climate.

I compared year-end 2003 and year-end 2013, offering a view on the changes that took place over an important decade of globalisation that embraced the five years leading up to the 2008 crisis, and the five years following it.

The results were fascinating.

First, the numerical scores from five economies did not cluster with any group because of outsized endowments in one or more areas. These countries included the US, which defied the cluster because of its large and wealthy population; China, which did not cluster because of its huge population; India, which did not fit because of its large population and many differences with China; Hong Kong, which was anomalous because of its combination of high wealth, strong financial development and small population; and Qatar, which is an outlier because of its wealthy and small population.

Second, most “emerging” countries rose up the ranking during the decade while some “advanced” economies slipped. The result was a bunching-up in the 5th-8th clusters from 27 countries in 2003 to 47 in 2013. Among the big advancers were Ghana, which climbed from Group 1 to Group 6, while big losers included Iceland, Ireland, Italy and Spain, which all slipped from Group 10 down to Group 8. Cyprus and Greece slid from Group 9 to Group 7. The two big surprises came from the Middle East, with Kuwait and the United Arab Emirates dropping from Group 8 to Group 4.

----------------------------

Clusters at year-end 2013 The world appears to have begun to converge; clusters 10-5 have increased from 46 countries in 2003 to 56 in 2013

1 Bangladesh, Nigeria, Pakistan

2 Honduras, Kenya, Rwanda, Uganda, Zambia

3 Egypt, Jordan, Lebanon, Morocco, Sri Lanka, Venezuela, Vietnam

4 Azerbaijan, Kazakhstan, Kuwait, Oman, United Arab Emirates

5 Argentina, Bolivia, Dominican Republic, Ecuador, El Salvador, Ghana, Guatemala, Jamaica, Namibia, Paraguay, Serbia, Ukraine

6 Brazil, Colombia, Indonesia, Malaysia, Mexico, Philippines, Russia, South Africa, Thailand, Turkey

7 Bulgaria, Costa Rica, Croatia, Cyprus, Greece, Hungary, Latvia, Mauritius, Panama, Peru, Romania, Trinidad and Tobago, Uruguay

8 Czech Republic, Estonia, Iceland, Ireland, Italy, Lithuania, Malta, Poland, Portugal, Slovak Republic, Slovenia, Spain

9 Chile, South Korea, Singapore, Taiwan

10 Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Israel, Japan, Luxembourg, Netherlands, New Zealand, Norway, Sweden, Switzerland, UK

Both the developed, western economies and ‘emerging markets’ have shifted towards more middle ground

As of 2013, countries such as Chile, South Korea, Singapore and Taiwan stood on the heels of the more developed western bloc

Traditionally safe countries such as Spain have shifted into clusters with more ‘risky’ emerging markets such as Lithuania

Ward’s minimum variance method, 1963. 10 = most prosperous countries as of 2013; 1 = least prosperous countries as of 2013

Source: Loomis Sayles

----------------------------

One insight the study throws up is that it may be instructive to view the development of countries as a continuum in which each stage of socio-economic progress may not be perceptibly different from each other, though the extremes are quite distinct.

Indeed, because many “emerging” countries have closed many socio-economic gaps and converged with “advanced” ones, it is now tough to say where “emerging” ends and “advanced” begins.

Another insight is that so-called emerging markets are now generally more mature than they were a decade ago and too big for investors to ignore. They contribute more than 50 per cent of global output on a purchasing power parity basis, most have robust “investment grade” credit ratings and have fast-growing financial markets.

China’s dramatic rise is also noteworthy. It has a credit rating similar to or better than most advanced economies (AA minus), while also having the world’s second-largest stock market and an economy that ranks second in size to the US in nominal terms.

These dimensions — combined with the fact that its economy is almost as large as all other emerging markets combined — argue for removing China from its current classification as an emerging market and creating a separate category for it alone.

The only country that is remotely comparable to China is the US.

Peter Marber is head of emerging market investments for Loomis, Sayles & Company. He also teaches at Harvard University and his latest book is ‘Brave New Math: Information, Globalization, and New Economic Thinking in the 21st Century’

几十年来一直困扰我的一个问题是：要具备哪些要素才能算作新兴市场。自1989年以来，世界银行(WB)对新兴市场的定义是：人均国内生产总值(GDP)在1.3万美元及1.3万美元以下的国家。但是，GDP只统计“财富”。人均GDP为12，999美元的国家和人均GDP为1500美元的国家之间差异巨大。

因此，人们呼吁采取一种新的、多个维度的划分新兴市场的方法。这种方法要将一组不同的指标考虑在内，以便得到相关国家社会经济成熟程度的排名。

在逾20年的时间里，我曾在这类国家里开展过交易，还曾从事过有关这些国家的教学。我曾见证过许多新兴市场的巨大进步，然而这些进步往往得不到金融指数提供商和投资者的充分认识。

我开始问我自己，除了GDP以外，定义一国社会经济成熟程度的其他维度是什么？有没有更客观的办法对各国进行分析和聚类？比起简单化的“发达国家”和“新兴国家”的描述，国与国的差别难道不应该更微妙一些么？我的直觉是，这些问题的答案是肯定的。不过，我需要一个客观的“大数据”框架来回答这些大问题。

去年，我用沃德法(Ward's Method)对这些问题开展了研究。沃德法是一种根据特定因子的“最小方差”对事物进行聚类的定量分析方法。也就是说，这种方法会把统计上最相似的事物分到同一组。

我的想法是把包括发达国家和新兴国家在内的100个国家分成10个聚类。10号组在9项具体标准上的分数最高，表示该组成员是更成熟的国家，对潜在的转变和冲击抵抗力较强。1号组则分数最低，表示这些国家不够成熟，更易受到一系列风险的冲击。

根据这9项不同标准类别为每个国家赋予分值，由此将这些国家分成10组。这样，人均GDP较高的国家，会在与该指标有关的标准类别中得到高分值。在其他8个类别中，有两个分别是人口的规模，财富及竞争力，还有三个是信用评级、企业上市率和货币估值，另外三个则是卫生保健、教育和政治氛围。

我比较了2003年底和2013年底的结果，这为观察这段全球化的重要十年里发生的变化提供了一个视角。这十年包括了2008年金融危机发生前的那5年，以及危机发生后的那5年。

由此得到的结果是令人着迷的。

首先，由于在一个或多个领域的能力太过突出，有五个经济体的各项分值无法与任何小组形成聚类。美国是这五个国家之一，它巨大而又富有的人口令它无法与别国聚成一类。中国也无法形成聚类，原因是其巨大的人口。印度同样无法聚类，因为它在拥有巨大人口的同时，还与中国存在许多不同之处。香港也无法聚类，它的异常是由于它集中了巨大的财富、高度发达的金融以及极少的人口这三大特征。还有一个无法聚类的国家是卡塔尔，它的独特性则是由于它的财富及极少的人口。

其次，在这十年里，多数“新兴”国家排名上升了，而部分“发达”经济体的排名则下滑了。其结果是，第5到第8聚类中的国家数量从2003年的27个上升到2013年的47个。有一个排名上升很快的国家是加纳，它从1号组上升至6号组。而排名下滑严重的国家则包括冰岛、爱尔兰、意大利和西班牙，它们都从10号组滑落至8号组。塞浦路斯和希腊则从9号组滑落至7号组。两个意外则来自中东——科威特和阿联酋都从8号组滑落至4号组。

以下为2013年底的聚类情况，世界似乎已开始趋同；10至5号组已从2003年的46个国家增长到2013年的56个国家。

1 孟加拉国、尼日利亚、巴基斯坦

2 洪都拉斯、肯尼亚、卢旺达、乌干达、赞比亚

3 埃及、约旦、黎巴嫩、摩洛哥、斯里兰卡、委内瑞拉、越南

4 阿塞拜疆、哈萨克斯坦、科威特、阿曼、阿拉伯联合酋长国

5 阿根廷、玻利维亚、多明尼加共和国、厄瓜多尔、萨尔瓦多、加纳、危地马拉、牙买加、纳米比亚、巴拉圭、塞尔维亚、乌克兰

6 巴西、哥伦比亚、印度尼西亚、马来西亚、墨西哥、菲律宾、俄罗斯、南非、泰国、土耳其

7 保加利亚、哥斯达黎加、克罗地亚、塞浦路斯、希腊、匈牙利、拉脱维亚、毛里求斯、巴拿马、秘鲁、罗马尼亚、特立尼达和多巴哥、乌拉圭

8 捷克共和国、爱沙尼亚、冰岛、爱尔兰、意大利、立陶宛、马耳他、波兰、葡萄牙、斯洛伐克共和国、斯洛文尼亚、西班牙

9 智利、韩国、新加坡、台湾

10 澳大利亚、奥地利、比利时、加拿大、丹麦、芬兰、法国、德国、以色列、日本、卢森堡、荷兰、新西兰、挪威、瑞典、瑞士、英国

发达西方经济体与“新兴市场”都向中间靠拢了。

到2013年，智利、韩国、新加坡和台湾等国已紧随更发达的西方国家之后。

传统上稳居发达国家之列的西班牙等国，已经与立陶宛等地位更“不稳”的新兴市场归于同一个聚类。

沃德最小方差法，1963。10=2013年最繁荣的国家；1=2013年最不繁荣的国家

来源：Loomis Sayles

---

这项研究得出的一条见解是，这样看待国家的发展可能富有启迪性，即将国家的发展视为一个连续的过程，每一个社会经济进步阶段与上一阶段可能都没有明显差别，但第一个和最后一个阶段可能差异非常显著。

实际上，由于许多“新兴”国家缩小了与“发达”国家之间的社会经济差距，与后者趋同，现在很难为“新兴”画一条终止线，为“发达”画一条起始线。

另一条见解是所谓的新兴市场大体上比10年前更为成熟，已经大到投资者不能忽视。以购买力平价(PPP)来衡量，它们贡献了全球产出的逾50%，许多新兴市场拥有健康的“投资级别”信用评级，还拥有快速增长中的金融市场。

中国的急速崛起也值得注意。中国拥有与大多数发达经济体差不多乃至更胜一筹的信用评级(AA-)，同时还拥有世界第二大股市，其经济体量以名义值计算居世界第二、仅次于美国。

这些指标，再加上中国经济规模几乎相当于其他所有新兴市场的总和，说明应该将中国从其现在所属的新兴市场分类中移出，为其单辟一个类别。

唯一可以与中国相比较的国家是相距遥远的美国。

本文作者是Loomis, Sayles & Company新兴市场投资主管。他还在哈佛大学(Harvard University)执教。他的最新著作是《美丽新数学：21世纪的信息、全球化和新经济思考》(Brave New Math: Information, Globalization, and New Economic Thinking in the 21st Century)

译者/何黎