Discussion of "How Does BigTech Credit
A¤ect Monetary Policy Transmission?"
By Yiping Huang, Xiang Li, Han Qiu, and Changhua Yu
Advanced Analytics: New M ethods and Applications for M acroeconomic Policy
Bank of England, European Central Bank, and K ing’s College London
Jonathan Benchimol (Bank of Israel)
July 22, 2022
This presen tation do es not necessarily re‡ect the views of the Bank of Israel.
1 / 9
IThis study analyses the transmission of monetary policy
through Big Tech and bank (business) credit.
IUnique dataset covering the entire borrowing history of
sampled …rms from Big Tech and traditional banks in China.
ICompare the responses of extensive and intensive margins
of both lenders to monetary policy changes.
IThe Big Tech lender’s extensive margin is found responsive to
monetary policy changes, while the intensive margin is not.
IPossible mechanisms: advantages in data abundance,
screening, and monitoring of the Big Tech credit.
IWhy is this an important area of research?
IBig Tech credit is expected to reach $1 trillion by 2023 (BIS).
=)new challenge for monetary policy.
ICompletes de Fiore, Gambacorta, and Manea (2022): Big
techs and the credit channel of monetary transmission.
IFive comments for a very cool paper.
2 / 9
IThe policy rate (changes) is used in the estimation.
IInterpreting the corresponding coe¢ cient as the e¤ect of
monetary policy on the dependent variable is problematic
IInstead, use monetary policy shocks identi…ed at high
frequency + panel LP-IV approach.
IChen, Higgins, Waggoner, and Zha (2016): Identifying China’s
Monetary Policy Shocks.
IWhat about monetary policy communication (shocks), not
necessarily about rates?
ICausal e¤ects of data abundance, screening, and monitoring
of the Big Tech credit, and their interactions, are weakly
3 / 9
IMonetary policy in China is hardly comparable to other
1. The monetary framework remains complex with unclear/non
transparent aspects and multiple instruments of monetary
2. Unique institutional framework for macroeconomic policy
making, where the State Council is the ultimate
decision-maker =)monetary and …scal policies are highly
IComplicates generalizations, more focus on China.
IDas and Song (2022): Monetary Policy Transmission and
Policy Coordination in China.
4 / 9
3. Policy Uncertainty
INew (unsupervised) players =)alternative to bank services.
INew challenge for the monetary policy conduct.
IMore importantly, this threatens …nancial stability and
challenges banking supervision.
IStates and central banks may regulate more.
=)Would your results hold following regulation changes?
IMore discussion is needed.
IWhy? Threat to identi…cation.
IRegulation shocks could a¤ect non symmetrically the relative
lending behavior of traditional banks and Big Tech lenders.
5 / 9
IThe interaction ABshould be estimated with Aand B.
IIf Bis non-time varying, you still have to estimate ABwith
A, i.e., if FE could capture B.
IDo individual-speci…c e¤ects correlate with the independent
IIf no, you should use random e¤ects.
INo Durbin–Wu–Hausman test in the paper.
IAmount of credit
IWhy not use dlog instead of log?
IFrom Section 3.3 to the end, no more control variables.
IWhat about additional control variables?
6 / 9
IHow did you get these numbers? More explanations are
I"the average probability of new lending relationship is 3.4%
(see Table 1)",
I"when monetary policy eases by one standard deviation, the
BigTech lender issues more credits than banks to the MSMEs
in the city by 41.73%"
IWhat about o- ine …rms externalizing their online?
I"we split the full sample of …rms into a subsample of online
…rms that sell products on the digital platform and a
subsample of o- ine …rms that do not conduct e-commerce"
Iwhat about (o- ine) …rms online on/with another platform?
(e.g., Uber Eats)
IWhat about …xed vs. variable rate loans?
7 / 9
INew title suggestion
IBig Tech Business Credit and Monetary Policy Transmission
ISame Nfor bank, bank unsecured and secured credit use?
I+60 years old owners
IInteresting to know what is happening to +60 owners.
IRemoving +60 owners may create a selection bias.
ICompare to online banks.
IMonetary policy changes look seasonal.
IMore strikingly, they look non smoothed.
8 / 9