China in Focus

Industry Articles
GCiS Commentary
Interesting Links

China Demographics

Poor-casts (poor forecasts) for China in 2012: part 1

December 21, 2011

Once again it’s the time of year when, feeling festive, we indulge in a little glühwein and try and throw a few darts of clairvoyance at 2012. In the context of China this means we do so in the equivalent of a dimly-lit room with a moving dart board. Possibly with darts being thrown at us.

Before I step up to the oche I want to understand why so much of what is predicted for the country turns out to be wrong. China has been described as the graveyard of prediction. Obviously, it’s a vast county, running on multiple speeds and with scant systemic cohesion, but could the problems also be with those who make them? After searching around, I’ve grouped bad predictions into five themes.

1. Playing to your bias (es)

“Tiger never does anything that would make him look ridiculous.”

- Golf Digest, asserting that Tiger Woods would make a fine role model for President Obama, 2010

These appeal to our sense of what is generally agreed to be true but we have no way of knowing. It was generally agreed that Tiger Woods was a squeaky clean model for the modern game of golf. The truth is that no one knew enough about him personally to make that call, but as it played to our biases, it was a very reasonable statement, for a while.

In the China business context, where we are both aware of malfeasance and doubly skeptical of assumptions, you’d think that this wouldn’t be the source of many “poor-casts”. It turns out that this is actually the biggest area. We assume we have a deeper level of knowledge about China than we actually have, using these cracked building blocks to make the case for a prediction. The predictor will start with commonly accepted facts: the economy has been growing rapidly for 30 years, people have been getting wealthy, investment has poured into the country, and inflation is high – to name a few – then, right when we are nodding our heads most vigorously, the logic skips like a broken record to what the forecaster wants you to hear: “it means the housing bubble is going to burst in 2012.” Non-sequiter.

2. Prediction that your job depends on

“People won’t want to play these electronic games for more than a week, not once we start selling pinball machines for the home.”

- Gus Bally, Arcade Inc., 1979

There were lots of quotes to chose from, many of them so-called famous last words. The captain of the Titanic, an alarming number of tech industry captains (including the above), and apparently anyone who ran a Wall Street firm before 2008. They’re not making forecasts so much as defending a current course of (in)-action.

This is the second most widely abused type, and, in our experience it occasionally causes sparks with a client. Say a sales manager wants a certain budget and to get that, he or she must convince the folks upstairs that China’s going to grow to a certain size, or that all his or her major competitors are investing in new factories. A third party, like GCiS, is brought in to make an estimate based on research and analysis. Occasionally our findings will contradict what the manager is hoping for and our modeling is poked and prodded (which is why we have to keep it really, really transparent) until a market scenario is mapped out by the client. And no, we don’t have fiery boardroom showdowns with the sales manager. We are typically (and understandably) kept away from them.

3. The snapshot fallacy

By 1985, air pollution will have reduced the amount of sunlight reaching Earth by one half….”

- Life, 1970

So far we have just been looking at qualitative fallacies. But it is equally possible to make poor predictions purely on numbers. As the economist Ronald Coase said, “Torture data enough, and it [sic] will confess anything”. Growth rates are the most frequent vehicle used for this because they describe flow instead of stock. But growth rates are never linear. They are a polynomial functions, any given differentiable point of which will show a snapshot of a reality shaped by the relative influence of the variables in play at that point in time. In other words, we forget that variables vary.

This gives rise to extrapolation ad absurdum. China’s market for automobiles grew at 25%-30% for many years, leading many people to think that this growth would continue unabated. In the super high growth years we were told that the automotive market would be three times the size of America’s in a few short years. A whole litany of reasons were used to justify this, although essentially people were extrapolating first, then making up the reasons why. A 25% growth rate, according to the rule of 7s, means that a market doubles every 42 months, which the automotive market did for a couple of cycles, then it became the biggest new car market in the world by volume, then it started to slow.

Will the slowdown (see my predictions in part 2) bring an end to this type of fallacy? Sadly not. Dinner conversations with panda-bulls claiming perpetually high growth, will be replaced by dinner conversation with panda-bears predicting imminent Armageddon (some may turn out to be the same people).

4. Apples & Oranges

China [doesn’t]... have the modern welfare state and China’s growing. And so what I would do is look at the programs that LBJ gave us with the Great Society and they’d be gone.

- Michelle Bachmann, Republican presidential hopeful

This is more of a forward-looking statement than a prediction, but it captures the “A is like this, so B is going to be like that” fallacy. Like the first group, it is useful when the predictor and / or audience understands one side of the coin much better than the other (although in Mrs Bachmann’s case, it could be neither). It is also frequently seen when the predictor has made a successful call on apples, but views oranges a bit like this.

A large number of these predictions are seen in the context of a China housing market bubble, especially when people talk about the causes of the bubble being the same. The US and China are completely different cases. In the case of the US, it boils down to mispriced interest rates. In China, it is due to a lack of alternative asset classes.

5. Out of your depth

“X-rays will prove to be a hoax.

- Lord Kelvin

There are a plethora of quotes here as well, along the lines of “so-and-so called the market turn of x and so here’s his opinion on something completely unrelated…” but even these paled against the Kelvin one above, because he was an otherwise distinguished expert, commenting on something he didn’t understand.

Tying in with the apples and oranges group, someone who has a good track record in their area of expertise, and, enticed with the whiff of laurel wreaths, is nudged back into the limelight in a different arena. Predictors in this group may also covet the attention of controversial remarks. When Jim Chanos, who predicted the Enron collapse, said that China – that is, the entire country – is the next bubble to burst, the whole world took notice. What it is still waiting for is sufficient detail on how this is going to transpire. This is not to say that Mr Chanos will be wrong, just that his prediction lacks the rigor of the call that brought him fame.

Next time
In part 2, I’ll be making my predictions for China in 2012, trying to stay out of these traps.


About GCiS China Strategic Research

GCiS ( is a China-based market research and advisory firm focused on business to business markets. Since 1997, GCiS has been working with leading multinationals in sectors ranging from technology to industrial markets, medical, chemicals, resources, building and constructions and a few others.


more commentary




Copyright © 1997-2016 GCiS
ICP 05040291
All rights reserved