Sie sind auf Seite 1von 2

The views expressed in this paper are the views of the author and do not necessarily reflect the

views or policies of the Asian


Development Bank Institute (ADBI), the Asian Development Bank (ADB), its Board of Directors, or the governments they
represent. ADBI does not guarantee the accuracy of the data included in this paper and accepts no responsibility for any
consequences of their use. Terminology used may not necessarily be consistent with ADB official terms.

COMMENTS ON SECTION II: Session II. Fintech and Financial Inclusion for Small and Medium-Sized
Enterprises (SMEs)

Credit Risk Database (CRD) for SME Financial Inclusion


Lan Hoang Nguyen, Credit Risk Database, Japan

The Paper:

The historical background:

- The SME credit crunch: land collateral lost much of their value, not to depend on collateral
anymore, so switch to risk-based valuation for lending
- Calls for nation-wide financial database for SME
- Hence, the Credit Risk Database funded by METI and BOJ

Contribution to SME Financial Inclusion

- SME sector
- Banking sector
- SME Guarantee System

Application of Machine Learning to Credit Risk Management

Comments:

Historical background: as the economy recovers, no more going back to collateral-based lending

Contribution to SME financial inclusion:

- On the accumulated data, what percent is it of total SMEs in the country?


- Standardized financial data of SMEs: who is in charge of regularly collecting them?
- Are they audited financial inclusion?
- Are they the same FS submitted to the revenue office?
- Will they be provided to member institutions only?
- Application: Credit scoring model: Subscription-based or transaction-based
- Improvement in guarantee policy for SME: from undifferentiated to differentiated guarantee fee
based on CRD scoring

Application of Manchine Learning to credit risk management:

- FS-based to complex sources of data: ability to collect and process high frequency data
- Includes SMEs with underdeveloped FS and start-ups
- Timely and targeted monitoring
- Important take-aways: useful for those developing credit guarantee scheme for SMEs and
application of machine learning to bank transaction data

General comments:

- Who is in-charge?
- How much is the cost and does revenue cover cost
- Seems to be a program to improve credit guarantee scheme which substitutes for land
collateral-based lending, but in developing economies, SMEs have even no land nor access to
credit
- At the end of the day, it is improving the guarantee scheme as a collateral substitute, but how
does an SME build a credit history without having land and access to credit guarantee facility?
- What happened to those that defaulted? How can they rebuild their credit standing to re-access
credit from banks?
- Comparison in performance: bank own scoring model vs. CRD model?
- Credit guarantee: individual vs. portfolio guarantee

Das könnte Ihnen auch gefallen