凯恩斯的宏观经济学：学术评价（3）
蒋闻铭
对凯恩斯理论的学术评价
凯恩斯的理论，本质上是一种肤浅的表象理论（nominal theory）。说肤浅，是因为他的分析没有触及到当时困扰整个自由资本主义体系的根本问题。
现代生产体系的基本特征在于对消费品的大规模生产。自由资本主义体系将注意力主要集中在生产问题上，但在一定程度上忽视了这样一个事实：大规模生产必须与大规模消费相伴而行。因此，问题的关键并不在生产端，而在消费端。
这一问题其实在马克思主义经济分析中已有所强调，但卡尔·马克思进一步认为，自由市场制度的基本原则使得这一问题无法得到解决。然而，以多数决为基础的民主制度, 实际上蕴含了解决这一问题的可能性，这一点在第二次世界大战之后的社会与经济发展中得到了体现.
从学术角度看，凯恩斯理论的核心在于利用货币的表象特性来设计各种刺激生产的方案。这一特征被后来凯恩斯主义的发展所捕捉并不断强化。任何依赖价格、工资以及其他名义变量“粘性”的刺激手段，本质上都带有某种“策略性”甚至“权宜性”的色彩，远远不足以彻底解决凯恩斯当时所面对的社会经济问题。然而，在温和衰退的时期，这类手段在一定程度上会产生短期效果。若将这种表象刺激当作推动经济增长的魔法杖，则是危险的，因为表象刺激不可避免地会引发通货膨胀。
但从学术角度看，政治家往往是“天真”的，而经济学家出于对现实影响力的追求，也往往愿意为短视的政治需求提供理论背书。于是，便出现了20世纪70年代现代经济体系中的严重通货膨胀。这一后果，与其归咎于凯恩斯本人，不如说应由当时的政治家与经济学家共同承担。
总体而言，凯恩斯理论在学术上可以归结为两大支柱：投资理论与货币和利率理论。在投资方面，他的推理逻辑如下：表象投资通过投资乘数被放大，从而带来名义收入的增长。名义收入的增加会导致两个结果。其一，当然是通货膨胀；其二，由于价格具有粘性，名义收入的上升会被企业解读为消费需求的增长信号，从而推动生产的扩大。于是，似乎可以“一石二鸟”：通过名义货币的增加，一方面改善就业状况，另一方面促进产出的增长。
“表象收入的增加会对生产产生正面影响”这一命题，是凯恩斯对经济学的重要贡献之一。但这一命题并不自动意味着，个人收入的增加会对整个经济体系产生总体上的正效应——而这恰恰是凯恩斯所默认的前提。要产生总体正效应，正面因素（生产的增加）必须能够抵消负面因素（通货膨胀）。那么，通货膨胀的负面效应究竟是什么？在新增的收入中，有多大比例转化为正面作用，又有多大比例转化为负面作用？这些关键问题，在凯恩斯的理论中并未得到触及。
接下来转向他关于货币与利率的理论。在这一部分，他的推理在某种程度上是“倒退”的，因为他（某种程度上并未自觉地）采用了已经破产的古典价值理论（李嘉图）。他认为，货币利率是其客观价值的主观反映，而这种客观价值体现在货币的完全流动性以及零持有成本上。
在凯恩斯所处的时代，主流理论认为：（1）货币是一种商品，利率是其价格；（2）像其他一切商品的价格一样，利率遵循供求规律，因此可以用均衡理论来分析。凯恩斯（正确地）意识到，（2）会像当时的就业理论一样存在问题。于是，他试图将（1）和（2）一并抛弃，建立自己的理论，但实际上却回退到了古典价值理论。
这样的理论并没有提供真正的洞见，其学术价值也相当有限。他所支持的一些政策设想，例如通过定期在货币上加盖印记以人为制造持有成本，只能被视为一种近乎荒唐的空想。 现代的凯恩斯主义者并未采纳他理论中的这一部分。
凯恩斯理论的总体负面影响，并不在于其结论，而在于其方法。他将注意力集中在现代经济体系的金融层面，却没有深入探究自由市场体系的基本动力机制。他的理论给人一种印象：只要通过对若干宏观层面的变量（其中许多是表象变量）之间关系的直接观察，就不仅可以理解，而且可以医治整个经济体系的“疾病”。这种倾向在现代经济学中因统计分析的广泛运用而被进一步强化。在某种程度上，正是由于他的影响，现代宏观经济学逐渐变成一个支离破碎的学科，各种相互竞争的观点可以无休止地争论，却缺乏对因果关系的整体把握。像卡尔·马克思和约瑟夫·熊彼特那样具有深刻洞察力的分析，在凯恩斯之后几乎消失殆尽。
附注：宏观经济与均衡理论
宏观经济学与均衡理论：在凯恩斯的分析中，一个显著缺失的因素是时间。在经济分析中忽视时间依赖性，在很大程度上源于古典均衡理论的影响——即认为经济分析的任务在于寻找一个供给与需求相等的均衡状态，而现实则围绕这一均衡状态波动。均衡原则在微观经济分析中，尤其是在短期问题上，取得了相当成功。然而，并没有充分理由假定，在宏观层面同样存在这样的均衡状态。
以古典就业理论为例。在需求方面，假定雇主愿意雇佣的劳动人数是单位工资的函数，即 D=D(w)；在供给方面，则假定愿意提供劳动的人数为 S=S(w)。显然，D(w) 是 w的递减函数，而 S(w) 是 w 的递增函数。所谓古典均衡，就是这两条曲线的交点——即供求相等的静态状态。
这一整套论述的核心在于：像经济体系中的其他事物一样，就业也会波动——时而上升，时而下降，但总体上围绕某一均衡状态运行。供求之间的平衡，被视为维持整个经济体系稳定的根本力量。然而，是否有理由认为 D(w) 与 S(w) 与时间无关？如果在经济不景气时期（例如衰退或大萧条），D(w)持续向下移动，又会怎样？一旦允许 D(w) 和 S(w) 随时间变化，那么前述分析便不过是一堆含混不清、毫无实质意义的空谈。
凯恩斯理论的一个根本缺陷在于，在他的经济模型中，缺少了约瑟夫·熊彼特意义上的企业家。在凯恩斯的体系中，企业家的关注点不是通过创新获取超额利润，而是精打细算地决定雇佣多少劳动力以实现其有限收益的最大化。如果现实世界真如他所描述的那样运作，那么资本主义早已走向消亡。
从本质上看，凯恩斯的宏观经济模型不过是古典均衡模型的延伸。之所以获得成功，并非因为它比古典理论更贴近现实，而在于它在理论上迎合了当时政治行动者的倾向。这一理论随后为经济学界带来的政治影响力，也同样为经济学家所乐于接受。
Keynes Theory--Academic Assessment
Keynes theory is in essence a nominal theory. It is rather shallow in the sense that his analysis does not touch the root of the problem that trouble the entire free capitalist system before and at that time. The basic characteristics of modern production system is that of a mass production of consumption goods. The free capitalist system focused on the issue of production but, to a degree, ignored the fact that mass production must go hand to hand with mass consumption. Therefore the problem was not on the production side but on the consumption side. This problem was actually emphasized in Marxism economic analysis, but Karl Marx went further to claim that the fundamental principle of the free market system makes a fix of this problem impossible. The super structure of this system: democracy based on majority decision, actually implied a possibility for a solution, as illustrated by the social and economic development after the second world war.
The essence of Keynesian theory, if academically pursued, is to use the nominal feature of the money to design schemes to stimulate production. This character was caught, and has been emphasized in modern development of the Keynesian theory. Any such deceptive schemes, relying on the stickiness of price, wages, and other nominal quantities, would be far from adequate in solving the problems Keynes faced. However, in the time of mild recession, it might help as far as it is novel. To adopt such nominal schemes as a magic wand to stimulate economic growth would be foolish, for nominal stimulation would inevitably lead to inflation. But politicians are foolish in academic sense, and economists, guided by their longings of worldly influence, are willing to lend their theoretic authority to the needs of the short-sighted politicians. There then came the super inflation episode of the modern economic system of the 1970s. The blame, in this case, should not befallen on Keynes, but on both the politicians and on economists of that time.
Overall, the academic contents of Keynes theory has two legs: his theory of investment and his theory on money and interest. On investment his reasoning is as follows. The effect of nominal investment is amplified in terms of nominal income (by the investment multiplier). Increased nominal income would lead to two things. First is of course inflation. However, because the stickiness of price, increment in nominal income would also be interpreted by private enterprise as an indication of increased consumption demand, therefore lead to increment in production. Hence we could hit two bird with one stone: Nominal increase of money would improve employment, and at the same time, increase production.
The proposition that nominal increment of income has a positive impact on production is Keynes's unique contribution to economics theory, but this proposition does not automatically imply that nominal increment of income would have an overall positive effect on the economic system, which was assumed by Keynes. To have an overall positive impact, positives (increased production) must off set the negatives (inflation). What is the negatives of inflation? how large is the likely portion of the increased nominal income working for the positive and what is the likely portion works for the negatives? These questions were not even touched in Keynes theory.
We now move to his theory on money and interest. Here his reasoning is kind of backward in the sense that he adopted (somewhat unconsciously) the method of the already bankrupted classical theory of values. Money interest, he reasoned, is a subjective reflection of its objective value, that is, its complete liquidity and its zero carrying cost. The prevailing theory in Keynes time was that (1) Money is a commodity and interest is its price; and (2) like the price of every other commodity, it obeys the law of supply and demand, therefore can be analyzed using the equilibrium theory. Keynes (correctly perceived) that point (2) would have the same problem as its parallel in the prevailing employment theory. He decided to throw both (1) and (2) out of window to construct a theory of his own, basically retreated to the method of classical value theory. Such theory is not in anyway enlightening, and it is of very little academic value. The program he endorsed, such as periodically stamp money to create an artificial carrying cost could only be regarded as fantastic. This is perhaps why no modern time Keynesian has actually picked up this part of his theory.
The overall negative impact of Keynes theory is not his conclusion but its method. He focused on the finance aspect of the modern economic system without dig deeper into the fundamental dynamics of the free market system. His theory gave the impression that, by candid observation on the relations between certain macro-level quantities, many of which are nominal in nature, one could not only understand but also cure the sickness of the system. This tendency is made worse by the extensive use of statistical analysis in modern economics. Partly because of his influence, Macro economics theory of modern time become a piecemeal subject, in which competing opinions can argue endlessly without an overall understanding on cause and consequences. Profound analysis, in the likes of the ones offered by Karl Marx and Joseph Schumpeter, disappeared all together after Keynes.
Additional notes:
Macro Economics and Equilibrium Theory: One thing that is markably missing in Keynes analysis is the time. To ignore the time dependency in economic analysis is by large due to the classical equilibrium theory: that economic analysis is that of finding an equilibrium state, at which supply meets demands. The reality would oscillates around such equilibrium state. This principle of equlibrium had been very succesful in micro econimc analysis, in particular when the issue is over a short period of time. There is, however, no base to assume there is an equlibrium state at macro level.
Let us take the classical theory of employment as an example. On the demand side, one assumes a function of total number of people the employer decided to hire as a function of unit wage, that is $D=D(w)$, and on the supply side we assume the number of labor who is forthcoming as $S=S(w)$. $D(w)$ is obviously a decreasing function of $w$ and $S(w)$ an increase function of $w$. The classical equilibrium is the point of intersection of the two curves: The stationary state where supply meet demands.
The entire point of this narrative is to claim that, like everything else in our economic system, employment does oscillate: sometimes it goes up, sometimes it goes down, but over all it stays around an equilibrium state. The balance of supply and demand is the fundamental force that maintains the stability of everything in our economic system. However, is there a reason that $D(w)$ and $S(w)$ are independent of time? What if $D(w)$ moves constantly downward in bad times, like in a recession or in the great depression? If one allows both $D(w)$ and $S(w)$ move as times goes, then the analysis outlined in the last paragraph would mean nothing more than muddled mumble jumble.
A fundamental fault of the theory of John Keynes was that, in his model of the economic system, the entrepreneurs of Joseph Schumpeter were absent. The focus of Keynes's entrepreneurs was not to acquire windfall pro t through innovation but to calculate carefully how many people they were going to hire to maximize their petty gain. If the world was like his economic system then Capitalism would be long dead. Keynes's macro-economic model was in essence an extension of the classical equilibrium models. The ultimate reason for the success of his equilibrium was not that it was more real than those of the classical theory, but that it confirmed in theory the proclivity of the man of political action of his time. The political influence it rendered to the academic world of economics afterwards was also greatly appreciated by the world of economists.
