


Economist, paid for doing what I love by Moody's Analytics as Head of Eurozone forecasting. Views my own. @crisisstudent.bsky.social


P.S.: Of course that I forgot to add that there a long-form piece I wrote on this: kamilkovar.substack.com/p/wi…

What was surprising was how geographically broad-based the increase in energy prices was. It wasn’t just Italy; we saw large jumps in gas prices in France, Spain, and Belgium. 7/

Finally, the energy detailed revealed what we suspected: the main driver were ga prices, seconded by electricity prices. We have moved away from a narrow energy shock focused on fuel alone… 6/

The good news is the peak effect of the summer heat waves shock on vegetables might be finally upon us, so food prices should show more moderate prints in the coming months. 5/

We also got explanation for hot-ish food price print. Largely, this was function of fruit, helped by vegetables. But few other categories also showed higher increases than was the trend lately. 4/

The other category are airline tickets, which have finally showed some hotness this year. As usual, pretty much only when people gave up expecting this… 😊 Don’t expect this to stay here, especially with fuel prices going down. 3/

What I mean by this? Most of the surprise came from volatile components which can easily reverse some of the increase. Here packaged holidays and accommodation suddenly shot up above normal trend in last two months. 2/

The final eurozone inflation report provided some clarity on the sources of strength in the October report. Main takeaway: the relative hot service print has pretty benign drivers. 1/
Euro zone inflation had another bad month in October, with only German fuel tax cut saving us from even worse numbers. Headline came in at 4.01% (vs 3.86% exp, 3.78% prev), and core also accelerated to 2.55% (vs 2.46% exp, 2.51% prev). To boot, composition was very nice. 1/7

Tagging authors, in case they are interested in this analysis. @DAcemogluMIT @davidautor

Bottom line: This is a very interesting paper that definitively shifted my priors on the topic, but just not to the degree the authors would wish. While I think automation will offset part of the ageing, based on experience so far, I think it will be only a small part. 23/23

Moreover, combining these two changes (GDP per worker and OECD) leads to even more negative results. 22/

So I was curious how things look in such a subsample – say in OECD countries? Turns out that the results do not carry over into this subsample: the coefficient either becomes smaller or changes sign if you exclude outlier in the form of Korea. 21/

The second, more important source of doubt relates to choice of countries. It felt like all countries are the wrong group here, and that we should rather focus on advanced economies, given that the primary mechanism is supposed to go through technology. 20/

And if you rerun their regression with GDP per actual worker, the coefficient drops substantially. 19/

And if ageing leads to higher participation rates, as seems to be the case, then it will make it look as though lower birthrates (i.e., ageing) lead to higher economic growth via productivity. 18/