Use of group IRB models in EU subsidiary and representativeness

U Offline

A banking group are looking to develop IRB models for the corporate exposures booked in the
local subsidiary, which is overseen by the local regulator

My gut instinct is that they should really be looking to use Group models, but maybe with a couple of tweaks (e.g. UK PRA and ECB seem to have a different perspective on question of sub ratings and parental support)

Have we seen any / many forms developing entirely different models for large corporates?

I appreciate mid-market may be slightly different issue where portfolio materiality may matter more

Any insights would be great

U Offline

I’ve previously seen the following at other ECB-supervised banks:

Client having a slightly different cut with “global corporates” (rather than large, although similar definition) which are rated via a group model
For local exposures if the parental support is there it inherits this (potentially some notching), and if not its treated based on the local counterparts size / profile (which also on its own might make it into the “global corporates” portfolio, or if smaller is treated via a local model).

I have seen different models at a SEA bank, where the same counterparty was rated differently depending on the rating entity but they were also working on consolidating them to eliminate this inconsistency. Reason for this (besides historically being separate portfolios) were different data availabilities, i.e., both different quality and availability of financial information as well as
other factors (e.g. credit bureaus) translating into different model factors, which were locally requested when the facility was opened, so did indeed depend on the country of booking. But I cannot see this being relevant in the EU/UK context

Representativeness in this context I have seen being addressed via add-on factors, i.e. factors which only take non-average values for that relevant portfolio segment – but this still results in one overall model score with a joint calibration

U Offline

I'll just flag that a Canadian bank who had decided to extend several of their existing "global" models to their US subsidiary based on similar logic, and while the logic makes sense broadly, they ran into significant issues with their US subsidiary’s regulator about the way they did it

I would primarily attribute the root causes of those issues to:

Operating model for how those decisions were taken and where they were reviewed and challenged – the Group development and validation teams led the substantive assessment of “fit-for-use”, and while there were some US stakeholders involved, the ones most involved were not very senior in stature, and often deferred to the expertise of Group. But those US stakeholders then couldn’t / didn’t do a good job of credibly defending those decisions to their regulators, leading the regulators to question if US senior management had been sufficiently involved in determining that these models were appropriate for the US portfolio
The documentation they produced justifying and validating the use of these models in the US was a narrow “fit for use assessment”, which taken as a standalone artifact fell far short of the comprehensive model documentation and validation expectations of SR 11-7. This led regulators to question the “effective challenge” provided by US model risk management and more broadly by US senior management