The post-COVID period has been tough for macro time series models. The decision by the @NYFedResearch to suspend their nowcasting model is the latest sign. I offer some thoughts based on my recent work with I. Petrella and @td_econ (1/N) The standard model used successfully for nowcasting in the last decades is a linear, Gaussian, Dynamic Factor model with constant parameters. During the great moderation, this was a fine approximation to the process of economic activity… (2/N)

My take: in Macro, careful analysis of the data is FEASIBLE and PAYS OFF, so it should be done before blindly applying off-the-shelf methods. Many new papers want to have cool buzzwords in their title but ignore the lessons from the previous decades & make basic mistakes....

https://twitter.com/JuanAntolinDiaz/status/1067370148754087937

There is tons of decades-old research highlighting the importance of careful data treatment in time series (trend, seasonality, outliers), which is simply IGNORED, with embarrassing results!

My co-author Thomas Drechsel (@td_econ) is on the academic Job Market this year with a paper on "Earnings-Based Borrowing Constraints
and Macroeconomic Fluctuations". Here is a short summary, be sure to check his profile! personal.lse.ac.uk/drechsel/ (thread) Looking at direct micro evidence on firm borrowing, Thomas points out that lenders typically restrict debt based on firm's earnings flows. Covenants, for example, are pervasive legal conditions written into loans that state that debt-to-EBIDTA cannot exceed certain limits.

The Financial-Entertainment Complex these days seems fixated with the idea of a looming recession (in 2020, apparently). Does this make sense? NO! Here is what MODERN MACRO has to say...
\begin{rant} First, much of the empirical evidence points to economic activity being essentially unpredictable beyond 1-2 quarters ahead. Nowcasting and ST forecasting has actually improved a lot. Beyond that, the best we can hope for is a good density forecast around the right LR mean (2/N)

Many private-sector economists use the concept of "credit-impulse" as if it was totally obvious that this is a reasonable and valid indicator. But this concept does not belong to the language of economics! I would like to make a few points. 1/ It was coined in 2008 by Michael Biggs, then an economist at Deutsche Bank. He found that a suitably transformed measure of credit flows had a strong contemporaneous correlation with GDP growth. He also wrote an (unpublished) paper on the issue. 2/ dnb.nl/binaries/Worki…

Hemos escrito una pequeña nota con los resultados de un "test de estrés" macroeconómico, simulando el impacto de una recesión en Italia sobre el crecimiento en otros países de la zona euro y EEUU: bit.ly/2HhuO92 Nos sale que la economía de Alemania sería la peor parada, Francia y España algo menos y EEUU solo se vería ligeramente afectada.

We have written a short note with the results of a macroeconomic "stress-test" that hits Italian economic activity with a 5th percentile outcome, measuring the propagation to other economies and stock markets. bit.ly/2HhuO92 A recession in Italy would transmit to the Euro Area, mostly through Germany, who appears to be the most exposed to Italian growth. Spain and France appear to be less vulnerable.