This is an open question. In "non-modern" banks, decision makers are generally assisted by simple dash boards (a set of indicators) that tell them about "the health" of the company. I am convinced that this approach is not efficient. There should be a lot of performance measures that can be modeled as single objective or multi-objective optimization problems.
Dear Nabil Belgasmi
I think the profit of the Bank can be influenced by some uncertain
factors such as market price. It is well known that market activity always involves different sources of risk such as future prices. To control the risk of financial losses during liquidation, I suggest to employ the Expected Downside Risk (EDR) or Cvar methods. EDR quantifies the risk in terms of values of loss distribution. Also you can use robust optimization for this condition.
What about using MCDA methods like Analytical Network Process to group the criteria and then imagine or use the best in class bank and the under examination bank as alternative choices. This way the "health" of the bank will be examined and it will also produce an ordering of the criteria used so you will have a clear view of which are the good points and which points should be improved.
Dear Nabil,
I think that for banks which are public companies in the sense that are listed to the stock exchange (and not state controlled companies), the best modeling practice is the shareholders value maximization subject to the balance sheet identity. Given that all the other stakeholders like employees, suppliers, savers etc are covered by fixed contracts in the short term horizon, the shareholder value max might be considered as a sufficient simplification. In a more dynamic long term analysis where all contracts, including loans and deposits, are under reprising, I think that a risk metric model (like EDR ) is more appropriate.
Isn't it the profit what is tried to be made optimal (i.e. maximal) in every bank?
This one generates a multitude of subproblems, starting from maintenance/rental expeditures, personel's wages, investments in computer/network equipment, long and/or short-time strategy on the stock/bonds market, credits conditions, and so on and on.
But even the profit itself may be defined differently, as daily, monthly, quarterly, yearly or in still another way. Do not expect a single best answer, it most likely doesn't exist.
Thanks Marek.
Let's say: what is/are the most important performance measure(s) that should be optimized in a bank (from an operational research point of view).
HI,
You may have a look to this paper:
Board structure and corporate risk taking in the UK financial sector
International Review of financial Analysis, March 2017, vol 50 Pages 101-110
Saeed Akbar, Buthiena Kharabsheh, Jannine Poletti-Hughes, Syed Zulfiqar Ali Shah
Best regards