Credit risk management fundamentals

Extensive literature has been written about the “true” level of non performing loans (NPL’s) in the Vietnamese banking system over the last 18 months.

The ongoing commentary has been both absorbing and humorous. The first red flag was raised by ratings agency Fitch in August, 2011. At the time, reported system NPLs for Vietnamese banks was about 2.7 per cent. But, Fitch estimated that 13 per cent was a more realistic number.

The State Bank governor acknowledged late last year that the real number was 8.82 per cent. We are in the middle of reporting season and the highest NPL number so far publicly reported by a Vietnamese bank is less than 4 per cent.  

The real number is subject to much debate. The recent purchase by Japanese super bank, Bank of Tokyo- Mitsubishi UFJ of 20 per cent of Vietinbank values it at nearly $4 billion, but with a market capitalisation of just $2.5 billion just prior to the transaction. It seems that the clouded NPL issue has not significantly affected valuations of some Vietnamese banks and the Vietnam long-term growth story remains at the front of minds for international investors.

The single biggest question in the banking sector is how to deal with the current NPLs and I’m not going to comment any further on resolution of this issue in this article. The second biggest question is how to make sure that the NPL issue doesn’t repeat itself in the future, and this is the focus of the remainder of this article. Nearly all articles focus on an NPL cure or resolution while few have focused on prevention.

Prevention is better than cure, especially as it relates to bad credit decisions, and this is the historical issue in Vietnam. It is easy to blame the fall in property prices and the downturn in the global and local economies as the main factors that have contributed to the current high NPL issue. These are outside the control of any and all banks, so it makes more sense to highlight the contributing factors that are in the control of the banks.  Some of the many reasons for the current high level of NPLs are listed below:

• Poorly designed internal ratings methodology leading to poor credit decisions
• Dysfunctional credit processes and structures that have been easy to game
• Credit ratings are not updated after the initial assessment
• Weak internal control framework and Internal audit functions
• Overstated collateral values and lack of independent and ongoing valuations
• Ratings models that assessed different borrower classes using the same methodology
• Due diligence on customer financial statements was weak or non-existent
• Lack of early warning systems at banks to highlight triggers to problem loans

Local banks have been urged to work in a transparent manner to attack the NPL issue

Banks globally use internal credit ratings models to assess a customer’s creditworthiness, ability to repay and to provide indicative pricing of a loan. Banks in developed markets generally apply a 90/10 rule to corporate lending assessment which means that 90 per cent of the credit assessment is based on the corporate borrower’s ability to repay (by using quantitative factors) and only 10 per cent is dedicated to qualitative factors such as the borrowers corporate governance, internal control environment and strength of its management.

Models are structured in this way because at the end of the day, the bank wants its money back and while internal control and strong corporate governance are desirable qualities for a borrower, it is the strength of the company’s financial position  that will ultimately ensure repayment of the loan. The simple rule is, the more reliance placed on qualitative factors in the credit assessment process, the more risky the assessment process and the less likely the bank will get its money back.

In Vietnam historically, a reliance on the quality of the data in financial statements provided by borrowers has been low. Models were tailored for Vietnam that assess a corporate borrower 35/65 (quantitative/qualitative), and while that might be suitable at a point in time, it is not appropriate now and is one of the main reasons that banks made unsavory credit decisions by applying that model. The other important factor specific to Vietnam is that the same model was often used across different borrower classes (for example SME and corporate) and this is clearly inappropriate.

But this is the past, and banks now have an opportunity to revise ratings methodology and improve processes and controls that are line with the current economic situation. Vietnamese banks are now seeking a new model, a “through-the-cycle” model that works throughout the economic cycle. The current methodologies are not often prudent in the current economic climate, and in fact, in hindsight, were not always prudent in the past.

In 2012, KPMG Vietnam had the opportunity to work with many banks in Vietnam in assessing the appropriateness of their credit process, their internal ratings methodology and also their overall approach to credit risk management. We understand current issues with these banks credit risk management and have suggested appropriate changes to the methodologies being used by these banks.

Newly released Circular 02/2013/TT-NHNN (simply called “Circular 02”) is the new State Bank regulation governing the way banks provide for credit loss. The circular requires, amongst other things, banks formally develop an internal ratings methodology for each borrower class, have it approved by the board, to integrate it into the banks systems and to advise the State Bank of the approach taken. This is clearly a step in the right direction to sound credit risk management and ultimately bank boards need to be responsible for methodologies behind the assessment of credit. It is the most important function of any Bank as it is where most of the risk lies.  

Banks that improve their credit processes and methodologies are likely to attract more capital from interested investors. This is especially relevant given the muted increase in the amount foreign shareholders can hold in domestic banks. Additionally, banks that improve their credit processes will be in a stronger position when the anticipated sector consolidation really starts to unfold. It is not sustainable to have 40 local banks given the size of the market, and only the strongest will survive.

The views expressed by the author here do not necessarily represent the views and opinions of KPMG.