What The Street Isn't Telling You About China

The thing is this: more than at any time in the recent past, the economic signals coming out of China are  contradictory, incomplete, wildly and inexplicably volatile and ultimately incoherent.   I've been watching China's data for - oh god - decades now, and I don't honestly think it's ever been as weird as this.  I cannot confidently tell you what China's economy is going to look like in six or nine months time, let alone in two or three year's time. But I'll tentatively say this: that unless things change swiftly and radically, the much-heralded and desired shift to an endogenous growth model (consumption allied with innovation-inspired productivity gains) and away from the exogenous growth model (massive investment, financial repression and US-devil-take-the-surplus) isn't going to happen. Indeed, if you have to straightjacket the current mess of data into a coherent pattern, it's one in which the investment & exports model is back big-time, mandated and reinforced by the choice of monetary policy instruments. (I'll explain how monetary policy choices are forcing this in a later post.)

Let's talk first of the gaps in the data. Here are just a few of the things we don't know about the 1Q. We don't know whether the government was running a surplus or a deficit - that's right, we really don't know the fiscal position. We also don't really know anything about retail consumption apart from the national total (up 16.3% in 1Q) because we have no by-province breakdown.  We think we know what's happening now to  property prices and sales, but we've no good way to interpret them because all the indexes have been abandoned or rebased.  We don't really know what's happening to liquidity, because PBOC has (wisely) started tracking a broader measure of credit, but (unfortunately) without also giving us some historic perspective. And, of course, we don't really know what's happening to inventory.

We do know what's happening to investment spending, because at least we have both national and provincial data. But of all the data, the fixed asset investment data is the ropiest in China because, like local authorities in the UK, it's wildly seasonal in order to satisfy various budget mandates.  What we do know, however, is that  the seemingly smoothly-growing national investment spend conceals unprecedentedly wild geographical swings. I regularly break down China's economy into seven different regions, and when you do that, you'll find that investment spending in the NW in March was 5.75 standard deviations above its historic seasonalized trends; spending in the Industrial NE was 4.5 standard deviations above; in the Yangtze Delta it was 3.9 standard deviations above.

Really?

But then March's data generally - particularly the industrial sector data - doesn't look right. Yes, January to March is the Dark Side of the Lunar New Year as far as data is concerned. But this year shouldn't have been too anomalous, since the holidays began on 3 Feb this year, compared to 14 Feb last year. But in fact, the seasonal anomalies were the most extreme in recent history:  exports fell 35.8% MoM in February and jumped 57.3% MoM in March - whereas normally (including adjusting for CNY) you'd expect a contraction of around 12.9% in Feb, followed by a 27.5% jump in March.  The anomaly was a full 2 standard deviations above where one would expect it.  Does anyone know why? I don't.

I also can't explain where the export growth came from.  In Guangdong, the pattern conformed pretty much to seasonal type, with March's exports up just 0.6 SDs from seasonalised trends. But in the Yangtze Delta . . . good lord, exports were 2.8 SDs higher than you'd expect them in March; in the Industrial NE they were 2.6 SDs above seasonalized trends, and in the Central Provinces they were 2.7 SDs.

How come Guangdong missed out on the fun?

Now I know what you're thinking: it must be the labour shortage. But I have news for you: I can track national employment totals monthly for 39 industries - it comes to about 80 million employees. And when I do that I find that in  February employment 1.6% lower YoY - ie,  these industries have shed about 1.3 million employees over the last year, mainly right at the end of last year. The figures tell me this: you can forget about the 'labour shortage' in the textiles, garments, paper, chemicals, plastic products, metal products, machinery  and nonmetal minerals industries, because they're all shedding workers fast.  Lucky the auto sector is still hiring (but what's this, auto sales up only 2.6% in Feb, whilst sector employment is up 5.2%?

Must be seasonal . . . touch wood.

And so on. We have tangled wreckage where we'd like to have data. Yet the question of whether China is overheating or heading for a hard landing has to be addressed. And, presumably, responded to by fiscal and monetary policies.  And that's what we'll look at next. As a sneak preview, though, I feel it fair to warn you that the likely outcomes of current policy settings are rather different from the ones you have been encouraged to contemplate and invest in.