索罗斯：我为什么（部分）同意哈耶克

一般上都认为弗里德里克·哈耶克（Friedrich Hayek）是市场学派的信徒——市场学派相信，只要政府不进行干预市场市场会保证最优的资源分配。这是一个正规化，数学模型化的理论，而这一理论有两大支柱：有效市场假说和理性预期理论。

这通常叫作“芝加哥学派”，它统治了美国的经济学教学。我则称之为：市场至上学说。

我自己一个不同的解释——跟有效市场假说与理性预期理论正相反，学说的两根支柱是易犯错性理论和自反性理论。

我坚信这两者与哈耶克的观点都是一致的。

然而我们不可能都是正确的。如果我正确的话，那么市场至上主义就是错的。这就意味着我必须能够拿出哈耶克的观点中不自洽的证据来，而这正是我要做的。

二十世纪四十年代，我在伦敦经济学院读书时，就在学校的期刊《经济学报》上看到了卡尔·波普尔（Karl Popper)与哈耶克的方法论之争。

我把自己看作是波普尔的信徒。然而这一次我却站在了哈耶克一边。他强烈地抨击了他所谓的“唯科学主义”——即机械模仿牛顿物理学。而波普尔反对这一观点，而赞成所谓的“科学大统一学说”——即对所有的科学分支都可以采用相同的方法和标准。

我被自己对波普尔的兴趣带入了这场争论。我阅读了波普尔的著作《开放社会及其敌人》，在书中他辩称：永恒真理是人类的智慧所不能及的，而任何宣称掌握了永恒真理的思想体系都必将是错误的。因此，他辩称，要在社会中推行之这种“思想体系”只可能采用强权。

这让我看到了共产主义政权和纳粹的相似性。我在两者都经历过的匈牙利生活过，这给我留下了深刻的印象。

这介绍给了我波普尔的科学方法论。波普尔宣称：科学理论永远不可能被证实，而只能被证伪。所以其有效性只是暂时的——理论都必须永远面对证伪的试验。这就避免了需要证明引科学理论放之四海而皆准所引发的所有问题，也确立了试验的重要性。只有能被证伪的理论才能被称为科学。

在我沉醉于波普尔理论的优雅之时，我还学习了经济学基础。我惊讶于完备知识能否的矛盾：它是完全竞争理论的前提，而波普尔的理论又断言它不可企及。要解决这一矛盾，就需要认识到：经济学无法达到牛顿物理学的标准。

这就是为什么我同意哈耶克——他警告那些照葫芦画瓢模仿自然科学的人，并且不同意波普尔的意见——他主张"统一方法主义"。

哈耶克辩称：经济学个体基于自己对现实的诠释，而非现实本身进行决策。而诠释和现实并不是一回事。

这就是我所谓的“易犯错性”。哈耶克也认识到，基于对现实不完备的理解的决策，必定会导致意想不到的后果。然而据这一点，哈耶克和我却做出了截然相反的论断。
哈耶克用这一点来颂扬市场中的“看不见的手”，这只手就是经济学个体追求自身利益所带来的意想不到的后果。而我，则用它来解释金融市场所内禀的不稳定性。
我的自反性理论认为：经济学个体的思考有两个职能：一方面，他们试图理解现实——这是认知职能；而另一方面，他们又试图影响现实——这就是参与，或曰操控职能。

这两种职能从两个相反的方面把现实和参与者认知联系起来。只要两者互相独立，那结果就是确定的。但当两者同时作用的时候他们就会互相影响。这种情形不仅仅在金融市场，而且在许多社会现象中都可见到。

我把这种相互作用叫作自反性。自反性在参与者的理解以及事件的实际进程中都引入了无法量化的不确定性。

这种双向联系以反馈环路的方式工作。反馈可正可负：正反馈增强当前的趋势以及偏差——并导致金融资产的定价出现错误；而负反馈纠正偏差。走到一个极端的话就会出现平衡，而另外一个极端就是金融泡沫——定价的偏差太离谱以至于无法持续的时候就会产生泡沫。盛极则必衰。

在现实世界中，正负反馈兼而有之，而极端的情况即使能出现也会比较罕见。因此，“有效市场假说”中假设的“平衡态”实际上是极端状态——与现实基本无关。

弗兰克·奈特（Frank Knight）是第一个发现金融市场中无法量化的内禀不确定性的人。凯恩斯（John Maynard Keynes）及其追随者们则详细说明了其内涵。

古典经济学家们则相反，他们认为通过他们发明的方法，就能消除自反性所带来的不确定性。哈耶克就是其中之一。

《经济学报》上的方法论之争的背景，是一场更大的关于政府在经济中的职能的政治争议。哈耶克是正方，凯恩斯以及其他社会主义计划经济学家是反方。

然而哈耶克让他的政治倾向引导了这场方法论之争。这就是他不一致的原因。他在《学报》上攻击科学主义，而二战后当共产主义的威胁尖锐起来的时候，他一改对方法论的疑虑，转而成为市场至上理论的信徒——只在他的诺贝尔奖致辞中稍微驳斥了一下过度使用量化方法而已。

因为他反对共产主义，一套证明市场参与者各自追求自身利益就能最优地分配资源的科学理论实在让他无法抗拒。然而，这一理论又是太好了，难以成真。

人们用自己不完善的知识指导行动，他们的决定会导致一些他们意想不到的后果。这就让社会事件比自然现象更不可预测。因此，哈耶克最开始反对科学主义是正确的。

在《学报》争论时期，波普尔的观点介于哈耶克和社会主义计划经济学家之间。他同哈耶克一样反对共产主义对个人自由的威胁，然而他则主张他所谓的“零敲碎打的社会工程”，而不是无为而治。

在这一点上我同意波普尔。但波普尔和哈耶克分歧并不大，我受他们两者同时影响，而又发现他们两者都有错误。

在此说明哈耶克的前后不一以及他的政治倾向，并非是为了贬低他，而是为了增进我们对金融市场以及其他社会现象的理解。我们都有某种偏见，而在自反性作用下，我们的偏见就是塑造历史的主力军。

因为知识完备不可达到，那么，差别就在于我们对现实的理解有多接近现实本身。清楚了“有效市场假说”和“理性预期理论”都已进了故纸堆，我们就前进了一大步。

正如早先一样，现在又流行起辩论起国家在经济中的作用来了。不过，政治辩论的水平已经大不如前。大选双方曾经展开启发性的唇枪舌战，而现在却互相几乎不理不睬。就因为这个，我接到凯脱学会的邀请时欣喜过望。

正如我所见，辩论双方横看成岭侧成峰，都声称看到了庐山真面目。极右派带头发难，宣称一切都是政府的错，而所谓的左派，目前　则被迫为规范私有经济以及开展政府服务辩护。

尽管我常被看作是左派，而且我也必然不会没有政治倾向，我还是认识到另外的一方也对了一半：他们宣称政府浪费资源，效率低线，应当更尽职尽责。

然而我依然坚持另一半的真理：也就是金融市场生来就不稳定，必须被规范。

最重要的是，我非常担心那些把部分真理鼓吹成真理的人。无论他们是左派还是右派，都会对我们的开放性社会造成威胁。
我相信，无论是哈耶克还是波普尔，都会有着同样的担忧。关心保护个人自由的人们应该联合起来，恢复过去曾让我们的民主更好地发挥作用的政治辩论水平。
（本文摘录自作者２０１１年４月２８日在凯脱学会（Cato Institute）的演讲）
Friedrich Hayek is generally regarded as the apostle of a brand of economics which holds that the market will assure the optimal allocation of resources — as long as the government doesn’t interfere. It is a formalized and mathematical theory, whose two main pillars are the efficient market hypothesis and the theory of rational expectations.

This is usually called the Chicago School, and it dominates the teaching of economics in the United States. I call it market fundamentalism.

I have an alternative interpretation — diametrically opposed to the efficient market hypothesis and rational expectations. It is built on the twin pillars of fallibility and reflexivity.

I firmly believe these principles are in accordance with Hayek’s ideas.

But we can’t both be right. If I am right, market fundamentalism is wrong. That means I must be able to show some inconsistency in Hayek’s ideas, which is what I propose to do.

Let’s start with Hayek’s influence on the twin pillars of my interpretation. I was a student at the London School of Economics in the late 1940s and read the great methodological controversy between Karl Popper and Hayek in Economica, the school’s periodical.

I considered myself a disciple of Popper. But here I was on Hayek’s side. He inveighed against what he called “scientism” — meaning the slavish imitation of Newtonian physics. Popper took the opposite position. He argued in favor of what he called the doctrine of the unity of science — that the same methods and criteria apply to all scientific disciplines.

I was drawn to this controversy by my interest in Popper. I had read his book, “Open Society and its Enemies,” in which he argued that the inconvertible truth is beyond the reach of the human intellect, and ideologies that claim to hold this truth are bound to be false. Therefore, he argued, they can be imposed on society only by repressive methods.

This helped me see the similarity between the Nazi and communist regimes. Having lived through both in Hungary, it made a great impression.

This led me to Popper’s theory of scientific method. Popper claimed that scientific theories can never be verified — they can only be falsified. So their validity is provisional — they must forever remain open to falsification by testing. This avoids all the problems of needing to prove scientific theories beyond any doubt and establishes the importance of testing. Only theories that can be falsified qualify as scientific.

While I was admiring the elegance of Popper’s theory, I was also studying elementary economics. I was struck by a contradiction between the theory of perfect competition, which postulated perfect knowledge, with Popper’s theory, which asserted that perfect knowledge was unattainable. The contradiction could be resolved by recognizing that economic theory cannot meet the standards of Newtonian physics.

That is why I sided with Hayek — who warned against the slavish imitation of natural science and took issue with Popper — who asserted the doctrine of unity of method.

Hayek argued that economic agents base their decisions on their interpretation of reality, not on reality — and the two are never the same.

That is what I call fallibility. Hayek also recognized that decisions based on an imperfect understanding of reality are bound to have unintended consequences. But Hayek and I drew diametrically opposed inferences from this insight.

Hayek used it to extol the virtues of the invisible hand of the marketplace, which was the unintended consequence of economic agents pursuing their self-interest. I used it to demonstrate the inherent instability of financial markets.

In my theory of reflexivity I assert that the thinking of economic agents serves two functions. On the one hand, they try to understand reality; that is the cognitive function. On the other, they try to make an impact on the situation. That is the participating, or manipulative, function.

The two functions connect reality and the participants’ perception of reality in opposite directions. As long as the two functions work independently of each other they produce determinate results. When they operate simultaneously they interfere with each other. That is the case not only in the financial markets but also in many other social situations.

I call the interference reflexivity. Reflexivity introduces an element of unquantifiable uncertainty into both the participants’ understanding and the actual course of events.

This two-way connection works as a feedback loop. The feedback is either positive or negative. Positive feedback reinforces both the prevailing trend and the prevailing bias — and leads to a mispricing of financial assets. Negative feedback corrects the bias. At one extreme lies equilibrium, at the other are the financial “bubbles.” These occur when the mispricing goes too far and becomes unsustainable — boom is then followed by bust.

In the real world, positive and negative feedback are intermingled and the two extremes are rarely, if ever, reached. Thus the equilibrium postulated by the efficient market hypothesis turns out to be an extreme — with little relevance to reality.

Frank Knight was the first to identify the unquantifiable uncertainty inherent in financial markets. John Maynard Keynes and his followers elaborated his insight.

Classical economists, by contrast, sought to eliminate the uncertainty connected with reflexivity from their subject matter. Hayek was one of them.

The methodological debate in Economica took place in the context of the larger political controversy over the role of the state in the economy. Hayek was on one side, Keynes and socialist planners on the other.

But Hayek subordinated his methodological arguments to his political bias. That is the source of his inconsistency. In the Economica, he attacked scientism. But after World War II, when the communist threat became more acute, he overcame his methodological qualms and became the apostle of market fundamentalism — with only a mild rebuke for the excessive use of quantitative methods in his Nobel Prize acceptance speech.

Because he was fighting communism, a scientific theory that proved that market participants pursuing their self-interest assure the optimum allocation of resources was too convenient for him to reject. But it was also too good to be true.

Human beings act on the basis of their imperfect understanding — and their decisions have unintended consequences. That makes human affairs less predictable than natural phenomenon. So Hayek was right in originally opposing scientism.

At the time of the Economica articles, Popper was between Hayek and the socialist planners. He was just as opposed as Hayek to communism’s threat to individual liberty, but he advocated what he called piecemeal social engineering rather than laissez-faire.

Here I sided with Popper. But Popper and Hayek were not that far apart. I was influenced by both — and I also found fault with both.

By identifying Hayek’s inconsistency and political bias, I do not mean to demean him — but to improve our understanding of financial markets and other social phenomena. We are all biased in one way or another and, with the help of reflexivity, our misconceptions play a major role in shaping the course of history.

Because perfection is unattainable, it makes all the difference how close we come to understanding reality. Recognizing that the efficient market hypothesis and the theory of rational expectations are both a dead end would be a major step forward.

As in that earlier time, the political controversy on the role of the state in the economy is raging today. But the standards of political discourse have greatly deteriorated. The two sides used to engage in illuminating arguments; now they hardly talk. That is why I was so pleased to accept this invitation to the Cato Institute.

As I see it, the two sides in the current dispute have each got hold of one half of the truth. which they proclaim to be the whole truth. It was the hard right that took the initiative by arguing that the government is the cause of all our difficulties; and the so-called left, in so far as it exists, has been forced to defend the need for regulating the private sector and providing government services.

Though I am often painted as the representative of the far left — and I am certainly not free of political bias — I recognize that the other side is half right in claiming that the government is wasteful and inefficient and ought to function better.

But I also continue to cling to the other half of the truth — namely that financial markets are inherently unstable and need to be regulated.

Above all, I am profoundly worried that those who proclaim half truths as the whole truth, whether they are from the left or the right, are endangering our open society.

Both Hayek and Popper, I believe, would share that concern. Those of us concerned with the protection of individual liberty ought to work together to restore the standards of political discourse that used to enable our democracy to function better.