You are a bank: what to do to reduce risks?

    Imagine that you are a bank. You, on the one hand, need to prove to the regulator that you are reliable (that is, to do something to be that way), and on the other, to minimize costs. At the same time, it is impossible to cut obviously important things for safety and insurance against phakaps, so one of the few successful ways to reduce your expenses is to use terabytes of collected data to optimize processes.

    This is part of risk management, which has become especially relevant recently.

    So, you have 4 main types of risks:
    1. Credit risks when the borrower simply does not return the funds even despite the presented iron.
    2. Market risks (risks associated with changes in asset prices).
    3. Liquidity risks (that you will not be able to fulfill your payment obligations on time, for example, if you immediately get a huge number of panicking investors screaming “turn everything back and forth”).
    4. Operational risks - everything else that does not fall into clauses 1-3. This may be the hijacking of a collection vehicle, misuse of funds and so on. Smart, you already have a huge database with an understanding of when and what was lost.

    Since you are already a fully formed bank, you were able to adjust for yourself and use all the world practices for risk management from points 1-3: these are complex, but still quite formalized processes. It’s worse with operational risks, because the question is not very clear. Now it’s quite obvious that the task of managing operational risks requires quite specific procedures for collecting and analyzing data, that is, a mass of methodological and organizational measures.

    The situation is still interesting in that all future risk management will need to be described to the regulator. For example, now the Central Bank requires you to simply lay a certain percentage of profits in insurance against such risks. Roughly speaking, it looks like this: the more you earn, the more you risk, and the more you need to reserve funds. There is a second possibility (which will appear soon) - it is connected with the fact that you can reduce the capital burden to cover operational losses if you justify that everything is really in order for you.

    How does it work on fingers?

    Imagine a hypothetical situation of dialogue with the Central Bank.

    - How much do you earn?
    - 100 000 rubles.
    Central Bank:
    - Then you need to reserve 10,000 rubles, because you can make mistakes on them.
    You recall your friend Petya (suppose he is also a bank) and specify:
    - Is it true that Petya-gouging has the same reserve?
    “Yes,” says the Central Bank, the same, “10,000 rubles.”
    “But I have superb work experience, I constantly go to trainings, don’t drink, don’t smoke and do my best to become better.” But Petya drinks, scores for work, and they are ready to fire him tomorrow. It turns out that you treat me like Petya?
    - Yes. And we pose the same risk to you as for the case when Petya screwed up. He is the most irresponsible in the team, and the others in terms of scale are equal in it.
    - Well, if so, I will also drink and skip work. I have no incentive to be good.
    - Ok, ok, you are great, you know how to manage your risks. Is there a sound methodology, just to make sure?
    You get a hell of a bunch of papers:
    - Here I have a technique, according to it I have to reserve 8 thousand, not 10.
    Central Bank:
    - Okay, persuaded.

    At the same time, Pete can be “rolled in” 20 if he shows his “hammer on everything” methodology, so he decides to stay on the old “like everyone else” (like Petya) and pay 10.

    As a result, for the “good”, the transition to an advanced method leads to lower costs. Given that this does not change the bank's vulnerability to risks (after all, the system was already used), this is simply the release of resources.

    So what?

    According to a number of data, the Central Bank will soon give banks the opportunity to calculate risks according to their internal models. There must be two things for this:
    1. Base of operational losses for 3 years.
    2. The bank should show that its methodology for collecting this data + the methodology for assessing operational risks is adequate from the point of view of Central Bank experts.

    Does it really work?

    Yes, we considered for several banks (unfortunately, I can’t do it in more detail), and it turns out that on average, operational risk is reduced by about 20%. This means that you can free 20% of the amount that was previously frozen under the reserve in this direction. ROI - from 30 to 200%. Usually, the first year is spent for the system to “fight off” in terms of implementation costs.

    Why is this being done now, and not when the Central Bank will introduce changes?

    Because the bank must in a certain way keep records of operating losses. Still need to protect the method of collecting information and protect the methodology for assessing the magnitude of operational risk. Now we are automating such processes. To do this, we introduce systems tailored to the methodology of a particular bank - or we help to put in place such a methodology (used in advanced countries and satisfying the requirements of Basel II).

    Not IT specialists work with the delivered system in the bank itself. This comes to the IT department as part of the task of setting up services so that there are fewer crashes. The system is managed by the operational risk department.

    There are banks that have neither a methodology nor a systematic collection of data on operational losses. There are banks where operational losses are collected in an XLS file (or paper) and then evaluated by specialists. There are cases when the data is collected correctly, but not synchronized in a large branch network or synchronized crookedly, by hand. In such cases, we put everything from scratch, there are experts with the necessary experience. That is, we introduce a modern risk management methodology, if it is not, and put everything that is important for the subsequent protection of the regulator.

    Right now there is no incentive to deal with risks, but in the future there will most likely be no options. And by this moment you need to be prepared.

    Another feature

    My practice has shown that when working out the issue, facts of losses are often revealed, about which no one thought at all. Sometimes a number of simple recommendations reduce risks at times, sometimes you can and should do additional research on other issues. As a result, this leads not only to cost optimization, but also to actual risk reduction.


    It may seem that a correct risk assessment is needed only under the new standards of the regulator. In fact, already now many banks are moving to it in order to reduce possible losses and make forecasting more transparent. At the most commonplace level, this means that when a decision is made by a bank, the hidden risks are many times less, which, of course, ultimately affects the bank’s earnings as a whole.

    Also popular now: