Jan's inflation has regained some momentum, reversing the slight relief we got in Dec's results. We've now had Jan CPI flashes from both Germany and France, and they confirm it. Germany's CPI was up 4.9% yoy against a very steep base of comparison (because Jan 2021 was when German CO2 charges really escalated). The monthly rise of 0.4% mom is 2.2SDs above historic seasonal trends. Not only was energy up 20-.5% yoy, but food was also up 5%.
This year, food inflation is going to feature heavily around the world, in response to the jump in fertilizer costs in 2020.
France's CPI also rose 2.9% yoy, which was worse than expected, with a 0.3% mom rise which was 1.8SDs above what was expected. Like Germany: energy was up 19.7% yoy, and fresh food was up 3.6%.
Tomorrow we get the Eurozone CPI flash: at the moment, consensus is expecting 4.3% yoy, but from what we know now, it will come in a bit higher than that.
The second thing I want to draw your attention to is an addition I've made to my data universe. I've included the Chicago Fed's Brave-Butters-Kelley (BBK) set of coincident and leading indexes for the US for the first time. Data is something I take seriously, but it has a fundamental problem. In particular, the idea that there is an unaltered canon of measurements that make up 'the economic data' is misleading - because an economy is a thing that's changing all the time, and the things that get included as 'the data' can only ever reflect an economic structure that have already been and gone. That's the 'data problem'. One way to address that is to scan as widely as possible - it's why I look at 500 data pieces every month.
The BKK indexes do pretty much the same, taking into account 500 datastreams a month just on the US. This is almost a 'big data' view on the US economy, and one I very much approve of. In Dec the BBK coincident index was 0.66 SDs above trend (good); but the leading index was 0.31 from trend (bad). Interpreting that, it suggests GDP growth was running at an annualized 4.6% in Dec. But there are warnings here too: this running growth rate was down from the 8.3% signaled in Nov and 9.1% in Oct. Worse, the leading index has been consistently negative now since April last year, and during June-August came within a whisker of minus 1 each month.
The significance? Leading index readings below minus 1 tend to signal an elevated likelihood of recession ten months out: ie, between March and May this year. I think that 'biggish data' judgement is worth taking seriously.
