卡在中等收入陷阱？不存在的！_风闻

《世界报业辛迪加》3月27日刊登耶鲁大学教授、摩根士丹利亚洲区前主席斯蒂芬·罗奇文章《中国没有落入中等收入陷阱》

文：Stephen Roach

译：马力

长期以来，人们一直对中国有一个很固定的印象，那就是中国是一个经济快速增长的国家，这种固有印象的产生当然是有原因的。对于一个像中国这样大的经济体来说，几十年如一日保持年经济增速在10%以上，这在全世界来说是个前所未有的现象，而中国却做到了：从1980到2011年，中国经济一直在高速增长。然而如今这个奇迹似乎就要无法延续了。2012年以后，中国的经济增速下滑到了7.2%，而最近李克强总理在其年度《政府工作报告》中把对2019年中国经济的增长目标调到了6%-6.5%。

对于那些看衰中国的人们来说，这的确是一个颇为得意的时刻。毕竟，中国总理的增长目标下限6%意味着2019年的中国经济增速与当年高速增长时代相比可能会下降40%之多。这使得那些关于“中国将落入中等收入陷阱”的预测听起来越发有道理了，这个“陷阱”意味着中国这样一个经济高速增长的发展中国家在快要接近高收入国家门槛时会突然落入低速增长的轨道。关于这一经济现象的初步研究显示：当一个国家的人均收入达到1.6万-1.7万美元（这是根据购买力平价标准换算后得出的数据）这个区间时，预期将出现2.5个百分点的经济增速下降。根据国际货币基金组织（IMF）的数据，中国的人均收入在2017年进入了这个区间，而中国在2011年之后已出现增速下降，这就更加令人悲观了。

在我攻读经济学的学生时代，我学到的第一个道理就是要谨慎对待数据分析的结果，其中可能隐藏着严重的错误。而“中等收入陷阱”就是一个处理大量数据后得出错误结论的典型例证。其实，只需一个数据库和一台性能强大的计算机，我就能“证明”任何一个以“分析性假说”的面目示人的经济理论。如今“中国已落入中等收入陷阱”的说法广为流传，而我在这里要提出五点来反驳这一说法。

首先，“中等收入陷阱”这个概念本身可能就是不成立的。“中等收入陷阱”这一概念是兰特·普里切特（Lant Pritchett）和劳伦斯·萨默斯（Lawrence Summers）对全球125个经济体在1950年至2010年这个时间段的经济表现进行实证研究后提出的。不过他们能得出的最可靠的结论应该仅限于“经济发展本身有增速的不连续性和向均值回归（译注：根据“均值回归理论”，经济增速、房产价格等社会现象以及气温、降水等自然现象，无论高于或低于均值，都会表现出以很高的概率向均值回归的倾向。根据该理论，上涨或者下跌的趋势不管其延续的时间有多长都不可能永远持续下去，均值回归的规律最终一定会应验）的强烈倾向（a strong tendency for growth discontinuities and mean reversion）”。最近在北京举行的中国发展高层论坛上，萨默斯在分析快速增长的发展中经济体可能面临哪些结果时进一步发展了自己的理论，他认为一些国家出现的均值回归放缓现象只不过体现了经济本身有一种缩小“后奇迹差距”的倾向（a tendency to close a post-miracle gap）。很显然，这种周期性增长不连续现象的统计规律（the statistical regularity of such periodic growth gaps）与增长陷阱的长期困境（the permanent quagmire of a growth trap）是截然不同的两个概念。

第二点，人均收入1.6万-1.7万美元的陷阱门槛也许是一个看起来很有道理的表述，然而在动态的全球经济中，这一表述并没有特别大的意义。自2012年有关中等收入陷阱的早期研究结果发表以来，全球经济的规模已经增长了约25%，中等收入陷阱的门槛很可能也会随之水涨船高。正是由于这个原因，最近的研究表明，即便存在中等收入陷阱，这个陷阱也不应用一个绝对的收入数字来衡量，而是应该用发展中经济体相对于高收入经济体的趋近程度来定义。从这个角度来分析，当发展中经济体的人均收入接近高收入经济体的20%-30%时，危险会逐渐逼近。考虑到2019年中国按购买力平价标准计算的人均GDP将达到美国的30%左右，现在是时候开始担心了！

第三点，不应该用同样的标准来看待所有的增长放缓现象。一个国家的GDP总额是多个部门、多个产业、多种产品的制造活动的总结果。中国各个行业发生的结构性转变可能会造成增长不连续的表象，而这一表象也许只不过是经过深思熟虑的再平衡战略的结果。考虑到中国正从高速增长的制造业等第二产业转向增长相对较慢的服务业等第三产业，如今中国经济的表现正符合上面的表述。从某种程度上来讲，这种转变是中国再平衡战略的预期结果，增长有所放缓不应引起过于严重的忧虑。

第四点，中国经济增速放缓是意味着中国已落入“中等收入陷阱”还只是一种单纯的增长不连续现象？我想在当前中国经济发展遭遇的严峻挑战面前，这些都并不重要。在追上占据技术高地的发达经济体之后，中国会变成一个怎样的经济体？这就是为何中国会决意从引进技术转向走自主创新之路的原因。对于一个寻求进军科技前沿的发展中经济体来说，中等收入与高收入之间的差别不过是相对的（middle-income versus high-income status is a relative comparison）。尽管周期性的外部干扰（如去杠杆、全球经济放缓甚至贸易战等）会产生暂时影响，然而若能追赶占据技术高地的发达经济体并加入其中来共同推动科技进步，中国终将获得回报。中国领导人已经在“使中国在2050年之前晋升为一个高收入经济体”的目标阐述中涵盖了上述内容。

最后一点，当你着眼于一个国家的长期发展前景时，生产率增长要远比GDP增长更加重要。因此，我更担心中国会陷入生产率增长陷阱，而不是GDP增长陷阱。不过，中国的一个研究团队最近对中国全要素生产率进行的一项研究可以缓解我们的焦虑。与普里切特和萨默斯的研究结果一样，这项对中国全要素生产率的最新研究也揭示了中国在过去40年里的一些增长不连续现象。不过在过去5年里，中国全要素生产率的发展趋势是令人满意的：年均增长率约为3%，其中第三产业的全要素生产率增长尤为强劲。因此，尽管最近总体GDP增长在放缓，但由服务业引领的中国经济再平衡却正在对整体经济的生产率增长产生意义重大的积极影响。

现在的问题是中国是否能够维持其全要素生产率继续增长（考虑到中国日益强大的自主创新能力以及由不断壮大的受过良好教育的劳动力所支撑的服务业的持续增长，中国全要素生产率继续增长的可能性是非常大的）并享受资本存量持续增加带来的好处。中国这项最新的研究显示，如果能够做到这两点，那么中国在未来5年内的潜在经济增长率就会保持在6%左右。这样一种结果将非常符合中国的长期抱负。

我们可以说：是的，中国经济增长率10%的时代已经结束了，这也是不可避免的。然而我们有充分的理由相信，中国经济增长正在从量变转向质变，这才是故事真正的精彩之处。这意味着，面对“落入中等收入陷阱”的普遍担忧，中国大可以无视其存在。

No Middle-Income Trap for China

There has always been a fixation on Chinese economic growth. And with good reason. For a large economy, sustaining annual growth rates of 10% over several decades is unprecedented. And yet that’s exactly what China did from 1980 to 2011. But now the miracle is over. Since 2012, annual growth has slowed to 7.2%, and Premier Li Keqiang’s recent annual “work report” set a growth target of just 6-6.5% for 2019.

For the vast legions of China doubters, this is a “gotcha” moment. After all, the lower bound of the premier’s target implies a 40% deceleration from the “miracle” trend. This seems to vindicate warnings of the dreaded “middle-income trap” – the tendency of fast-growing developing economies to revert to a much weaker growth trajectory just when they get their first whiff of prosperity. The early work on this phenomenon was precise in terms of what to expect: as per capita income moved into the $16,000-$17,000 range (in 2005 dollars at purchasing power parity), a sustained growth deceleration of around 2.5 percentage points can be expected. With China having hit that income threshold in 2017, according to International Monetary Fund estimates, its post-2011 slowdown looks all the more ominous.

But one of the first things taught to economics graduate students, even back in my day, is to be wary of the perils of data mining. And the middle-income trap is a classic example of the pitfalls of endless number crunching. Give me a database and a powerful computer, and I can “validate” almost any economic relationship masquerading as an analytical conjecture. There are five key reasons to dismiss the now-widespread diagnosis that China is ensnared in the middle-income trap.

First, a middle-income trap may not even exist. That is the conclusion of Lant Pritchett and Lawrence Summers’s rigorous empirical study covering a broad cross section of 125 economies from 1950 to 2010. The best they could come up with is a strong tendency for growth discontinuities and mean reversion. At the recent China Development Forum in Beijing, Summers went further in assessing likely outcomes in rapidly growing developing economies, dubbing any mean-reverting slowdown as merely a tendency to close a “post-miracle gap.” Needless to say, the statistical regularity of such periodic growth gaps is very different from the permanent quagmire of a growth trap.

Second, a fixed trap threshold of $16,000-$17,000 may be a great literary device, but it makes little sense in a dynamic global economy. Since early research on the middle-income trap was published in 2012, the world economy has grown by about 25% – presumably boosting the moving target of a middle-income threshold by a comparable magnitude over that period. Largely for that reason, recent research has couched the trap not in terms of an absolute threshold, but as relative convergence to high-income countries. From this perspective, danger looms when developing economies’ per capita income approaches 20-30% of the level in high-income economies. Given that China will hit about 30% of America’s per capita GDP (in PPP terms) in 2019, it must be time to worry!

​

Third, not all growth slowdowns are alike. A country’s GDP is a broad aggregation of a multiplicity of activities across sectors, businesses, and products. Structural shifts from one sector to another can give the appearance of a growth discontinuity that may be nothing more than the outcome of a deliberate rebalancing strategy. This is very much the case with China today, given its shift from higher-growth manufacturing and other “secondary” industries to slower-growing services, or “tertiary” industries. To the extent that this shift is the intended result of China’s strategic rebalancing, a growth slowdown is far less alarming.

Fourth, the daunting challenges that China faces at this point in its economic development are far more important than whether its slowdown is a gap or a trap. What comes after the catch-up to advanced economies operating on the technological frontier? This is where China’s stated goal of shifting from imported to indigenous innovation comes in. Middle- versus high-income status is a relative comparison for developing economies seeking to operate on that frontier. Notwithstanding the temporary effects of periodic exogenous disturbances – such as deleveraging, global slowdowns, or even trade wars – catching up to the frontier and joining others pushing to move beyond it is the ultimate reward of economic development. That goal is enshrined in President Xi Jinping’s aspiration for China to achieve high-income status by 2050.

Lastly, productivity growth is far more important than GDP growth in determining a country’s development prospects. As such, I would be far more worried about China falling into a productivity trap than a GDP growth trap. A new study on total factor productivity by a team of Chinese researchers offers some comfort here. Like the work of Pritchett and Summers, this latest assessment of Chinese TFP growth reveals several discontinuities over the past 40 years. But the underlying trend for the past five years is encouraging: annual TFP growth of around 3%, with especially strong growth in the tertiary sector. So, notwithstanding the recent slowdown in aggregate GDP growth, services-led Chinese rebalancing is imparting meaningful productivity leverage to the economy as a whole.

The question now is whether China can sustain its recent TFP trajectory – a distinct possibility in light of an increasingly powerful shift to indigenous innovation and the sustained services-led productivity of a growing cohort of well-educated knowledge workers – and also reap the benefits of continued upgrading of its capital stock. If it can, the new Chinese study concludes that China’s potential GDP growth rate could hold at nearly 6% over the next five years. Such an outcome would conform quite closely with China’s longer-term ambitions.

So, yes, the days of 10% Chinese growth are over. That was inevitable. But there is good reason to believe that the real story is the shift in Chinese output from quantity to quality. That suggests that China will defy yet again widespread fears of a looming middle-income trap.

(End)