“Wherever there is human judgment there is potential for bias.”
That’s a quote from an excellent chapter titled “Are You Smarter Than A Television Pundit” in The Signal and The Noise. Before he moves onto baseball, Silver does a nice job differentiating between quantitative versus qualitative opinions in the partisan media.
“So you will need to adopt some different habits from the pundits you see in TV. You will need to learn how to express – and quantify – the uncertainty in your predictions. You will need to update your forecast as facts and circumstances change.” (pg 73)
If that sounds familiar, it should. This is all encompassing in Hedgeye’s founding principles of establishing Transparency, Accountability, and Trust through independent research. Journalists are not analysts. And only the great analysts in our profession embrace the uncertainty of there being a high probability of being wrong. It’s ok to say it like that. We’re not on TV.
Back to the Global Macro Grind…
China’s 7.4% GDP report for Q3 of 2012 provides a great example of where we were wrong this morning. In our Monday research meeting we discussed what we thought was a heightening probability that Chinese GDP surprised on the upside. It didn’t.
Now if you turn on the radio or TV this morning, you’ll probably see something very different than what I just wrote (or what I have been writing on Twitter). Since most of these sources are journalistically driven, they obviously don’t have forecasting models. Instead, they anchor on other people’s content (the sell-side’s), which at times can be even worse than a journalist’s opinion.
“China beat”, “China has bottomed”, “China is not Spain” – scanning the Old Media’s headlines will get you spew like that. Whereas I we’ll just show you the data within our analytical framework:
- China’s Q3 2012 GDP of 7.4% slowed sequentially (quarter-over-quarter) from 7.6% in Q2
- China’s Q3 2012 GDP of 7.4% slowed -19% year-over-year versus 9.1% in Q3 of 2012
- China’s Q3 2012 GDP of 7.4% slowed more than the low-end scenario in our forecasting model of 7.6%
In other words, across our risk management durations (TRADE, TREND, and TAIL):
- TAIL (3 years or less) – China continues to slow, and surprise both its government and the world on the downside
- TREND (3 months or more) – China continues to slow, at an accelerating rate, sequentially
- TRADE (3 weeks or less) – China’s stock market just moved to immediate-term TRADE overbought on the news
Maybe China has “bottomed.” But I have no high-probability edge on that and neither do you. Or, let me say that more democratically – if you can send me a model that shows me why and how China just bottomed, I’m happy to look at how you’ve analyzed the Chinese government’s made-up numbers. I’m even happier to change my mind.
Made-up, or Madoff? Yes, both American and Chinese guys make up the numbers. And this makes it all the more difficult to make a macro forecast that something has “bottomed” or “topped” with a straight face. It might get you on TV however.
Having had to learn from all the mistakes I have made the hard way, the best risk managed opinion I can give you is that both tops and bottoms are processes, not points.
In forecasting “Principle #1: Think Probabilistically” –Nate Silver
“Instead of spitting out just one number and claiming to know exactly what will happen, I instead articulate a range of possible outcomes.” (page 61)
I don’t love everything Silver thinks, but I do love that. That’s the closest thing I have read in the last year to what we call our Risk Range. That’s what you see at the bottom of every Early Look - our immediate-term range of probable upside/downside (risk) – and I think most people would say that’s the most accurate and repeatable forecast we give you every morning.
So skip my rants and go to the bottom of the note - save yourself some time.
One more point on China - to get Commodities right, we think you need to get the Dollar, Supply, and Demand right. If you think China has “bottomed”, you’re going to have a very different long-term forecast than ours right now. The risks are rising that there’s a decade long-cycle of price and demand topping.
Everything that happens in Macro that matters most happens on the margin. And if the new long-term range of Chinese GDP growth is 4-8% instead of what it’s been (8-12% for the last decade), that could matter, big time.
If and when we get that wrong, we’ll write about it transparently and accountably that morning.
Our immediate-term risk range for Gold, Oil (Brent), US Dollar, EUR/USD, UST 10yr Yield, Shanghai Composite, and the SP500 are now $1, $112.60-115.05, $79.01-79.69, $1.29-1.31, $1.73-1.83%, 2066-2139, and 1, respectively.
Best of luck out there today,
Keith R. McCullough
Chief Executive Officer