我們還能相信預測嗎？

Full disclosure: for much of my professional life I have been an economic forecaster. So I felt targeted when Michael Gove, a leading Brexiter, said during the EU referendum campaign that “the people of this country have had enough of experts with organisations with acronyms saying that they know what is best and getting it consistently wrong”.

首先，我需要先全面披露一下我的身分：在我職業(yè)生涯的大部分時間里，我一直是一個經濟預測者。因此，當英國退歐派的領軍人物邁克爾•戈夫(Michael Gove，見上圖)在退歐公投運動中說“這個國家的民眾已經受夠了那些來自名字為縮寫的組織、聲稱他們知道什么是最好、但卻不斷搞錯的專家”，我感覺自己成了靶子。

In a single blow, total denigration of experts, economic forecasting and forecasters. But surely there are some points to be placed on the other side of the ledger?

這是將專家、經濟預測和預測者“一網打盡”的一擊。但我們肯定還有一些砝碼能放在天平的另一邊吧?

First, it is a feature of the human condition that we are interested in what the future will bring. We rely on forecasts in daily life more than we might think. Before we leave home in the morning we listen to a weather forecast to decide what clothes to wear and a traffic forecast to decide what route to take to work. We make our own forecasts. By opting not to take a raincoat and taking the car, say, I am forecasting, albeit implicitly, that it will not rain and that the traffic will be manageable.

首先，人的一個特性就是會對未來會怎么樣感興趣。在日常生活中，我們比想象中還要依賴預測。早上離家之前，我們會聽天氣預報來決定穿什么衣服，聽交通預測來決定走什么路線去上班。我們也會做出自己的預測。比如，我選擇不帶雨衣并且開車上班，那么盡管沒有說出來，我預測今天不會下雨，交通情況也還能接受。

Furthermore, the reason that there are so many economic and financial forecasts is that people not only want them and need them, they also pay for them. The best economic forecasts tend to be the ones produced by organisations with the necessary computing power, and forecasting teams who provide the underlying data and spell out their assumptions. Users are thereby able to understand how the forecast was arrived at.

再者，有這么多經濟和金融預測的原因，是人們不僅想要、需要這些預測，而且他們也為一些預測付費。最優(yōu)的經濟預測往往來自那些擁有必要的計算能力的組織，以及提供基礎數據并詳細闡明其假設的預測團隊，這讓用戶能夠理解預測是怎么來的。

What about accuracy? Almost all forecasts are “wrong” to some degree. A forecast that consumer price inflation will be 3 per cent in a year’s time can be judged to have been “wrong” if the final figure comes out at 3.1 per cent. But that does not make the forecast useless. What matters is whether the forecast materially helps, or misleads, the user.

準確度如何呢?幾乎所有預測都在某種程度上是“錯誤”的。一項預測可能預言某年的消費者價格通脹將為3%，如果最終結果是3.1%，該預測也可能被判定為“錯誤”。這并不意味著預測是無用的。重要的是，預測在實質上是幫助還是誤導了用戶。

Weather forecasters have long appreciated this point. A farmer planning to spray his crop needs only to know whether it will rain, not how heavily. Similarly with economic and financial forecasts, the real issue is not one of arithmetic error; it is whether forecasts help or hinder the making of decisions. And here, inevitably, considerable responsibility rests with the user: it is unwise, irresponsible even, to use forecasts as a basis for making decisions they cannot support.

天氣預報員一直非常認同這一觀點。一個計劃給莊稼澆水的農民只需要知道是否會下雨，并不必知道雨會下多大。經濟和金融預測的情況也與此類似，真正的問題不是運算上的誤差，而是預測對決策提供了幫助還是造成了阻礙。在這一點上，用戶其實不可避免地要負相當一部分責任：一些預測并不足以作為決策的依據，如果非要根據這些預測進行決策，那就是不明智、甚至不負責任的。

Why discredit experts, then? In today’s “post-truth”, “post-fact” world, critics increasingly play the man rather than the ball. Where the economics profession is concerned, the tactic is as follows: first, make the principal test of their competence whether or not they can forecast accurately; then claim that they cannot do this; and finally conclude that there is no reason to take their analysis seriously.

那么，為何要貶低專家呢?在今天的“后真相”、“后事實”世界里，批評者日益對人不對事。針對干經濟學這一行的，他們是這樣做的：首先，對經濟學家是否有能力準確預測進行檢驗;然后主張他們做不到這一點;最后得出結論，人們沒有理由認真對待他們的分析。

John Van Reenen, a professor of applied economics at MIT, puts it thus: “[It is] as if the medical profession’s failure to predict the Aids epidemic means that you should ignore your doctor’s advice to give up smoking. No, we cannot predict the date you will die of lung cancer, but if you smoke we can be pretty sure your health will suffer.”

美國麻省理工學院(MIT)應用經濟學教授約翰•范里寧(John van Reenen)因而有此一說：“這就好比因為醫(yī)學界未能預測艾滋病會流行的緣故，就說明你應該無視醫(yī)生讓你戒煙的建議。是的，我們無法預測你會在哪一天死于肺癌，但如果你抽煙，我們很肯定你的健康將受損。”

And so to the final sleight of hand. Discrediting rational, fact-based analysis and those who use it to forecast, creates a vacuum; and with that comes the following, implicit, injunction: “Because these experts do not know what they are talking about, anyone’s view is equally valid.”

最后還有個花招。讓人們懷疑基于事實的理性分析以及那些運用這些分析進行預測的人，從而制造出一種真空;同時傳達一種含蓄的教導：“因為這些專家自己都不知所云，所以任何人的觀點都同樣有道理。”

This is a step too far. I am tempted to say to Mr Gove that when my wife developed cancer, we went to an expert — an oncologist. When I want to know tomorrow’s weather, I turn to a weather forecaster. And when I want to know what is most likely to happen to the economy, I turn to teams of professional forecasters who work for organisations with acronyms. The fact that they are unlikely to be exactly right does not give me any reason to suppose that Mr Gove will be.

這一步就有點過頭了。我很想告訴戈夫，當我的妻子罹患癌癥后，我們向一位專家——一位腫瘤醫(yī)生求助。當我需要了解明天的天氣時，我會查天氣預報員播報的信息。當我想要了解經濟態(tài)勢時，我會向那些名字是縮寫的組織旗下的專業(yè)預測團隊求助。他們不太可能完全正確，但我不會據此認為戈夫的話是正確的。

The writer, partner in Llewellyn Consulting, was head of forecasting at the OECD and global chief economist at Lehman Brothers

本文作者是Llewellyn Consulting合伙人，曾任經合組織(OECD)預測主管、雷曼兄弟(Lehman Brothers)全球首席經濟學家