The Option 2 is better solution among the options proposed in the Draft Regulatory Standards.
Option 1 does not seem to be a plausible solution as it would flag defaults for residual negligible due amounts of smaller clients (e.g. €25 past due for technical reasons in a €600 outstanding debt) and it would also mark as default relatively small hitches like a technical overdraft of €600 in a large corporate with €10 million exposure.
Option 1 would lead to a vast number of technical defaults being computed as true defaults thus distorting the meaning of the internal risk model results. Technical past due cases include typically:
- Neglected payments for small amounts;
- Payment processing errors.
In addition, the use of option 1 implies that the breach of an absolute threshold would trigger the default mark with no regard to the relative materiality.
Option 2 is the preferred one because it really leaves out virtually all cases of technical default, as intended, without representing any material departure from the general principle of computing all true defaults for regulatory purposes.
We recommend that the same approach is applicable by all national competent authority. Such solution will allow the international banks to limit their costs and create level playing field for banks.
The relative threshold should be placed at a little higher level for some kinds of credit exposition. In Poland we have the experience with credits for consumer or mortgage purpose and these credits are denominated in foreign currency. The problem of changes in the exchange rate may cause that thanks to the application of bad exchange rate may have the impact on treating the credit as past due, but this example should be treated as clear “technical” reason of past due.
We have the experience with the threshold in Polish regulation and in the practice of Polish banks. That is the reason we are convinced the time needed for implementation is not long. This proposed solution is favorable for banking society and this reason why we are in favour to implement short time.
However, we can imagine that banks using the IRB model will need more time to implement the new approach to the past due to their internal model. It is necessary to have more time to rebuild these models.
Due to the fact that implementation process will require at least the following:
- historical data adjustment to allow models’ calibration/validation
- new threshold implementation in the systems
- validation and recalibration of the models
- internal and external governance processes
- final implementation of the models in the systems and decision processes.
We express our opinion that full implementation requires not less than 2 years and in the case of institutions with high number of the risk models (PD, EAD, LGD), the time required may read more than 3 – 4 years after inclusion of business-as-usual activities and other changes on the horizon like deployment of IFRS 9.
The additional problem is how to calculate the period necessary for implementation of past due rule. In one solution the starting point can be the date of the adoption of this technical standard and in the second one the date of later setting the threshold by competent national authorities.
In general, the adjustments to data sets, the re-calibration of risk parameters and the implementation of model changes will mean new burden for banks. These costs would be exacerbated if the change is intended to be applied retrospectively and, in many cases, it could be unfeasible due to lack of necessary data.
The standard will generate the additional costs for banks in following areas:
- Data systems and IT adaptations.
- Human resources devoted to the change.
- Increased risk weighted assets: Option 1 would be much more punitive as it would not only treat as true defaults debts which are only technical defaults, but it would also push up the value of risk parameters thus inflating the risk weighted assets of the healthy portfolio.
The draft should not have big impact on credit strategy of banks.