BLOG

06 December 2023

Is Risk Management the same as Compliance? No, but risk management is increasingly (at least in this part of the world), becoming just compliance.  Reasons:

1. Regulatory and supervisory environment that favours prudency and conservatism.

2. Domain dependence risk management - siloed.

3. Lack of exposure and training.

4. Delegate to model culture.

On 1, the environment in this part of the world simply means it is easier to just follow instructions less the full power of a letter from beyond the walls of the bank spells the end of any risk manager's career.

On 2, despite decades of "Enterprise Risk Management", many risk managers remain domain dependent. That is, they operate and think in siloes, unable to transfer their understanding of risk outside of their specialty or domains. 

On 3, there was a time when risk managers "cut their teeth" by taking risks. For example, started in the credit evaluation shop. Being a branch manager. Being a trader. Being a treasury manager, etc. Today, most risk managers are based on their past work exclusively in risk.

On 4, increasing use of models and data meant instead of relying on analysis, risk managers simply delegate decisions to models - hence the substantial resources spent on model risk management as compared to fixing data or just practicing risk management.

Perhaps it is the evolution of risk management that risk management is slowly or has become compliance. Ideally risk management should be an exercise where the risk manager has some "skin in the game" - the goal isn't to say zero loss, or no risk, but to take acceptable risks based on robust analysis. 

#riskmanagement #riskculture #riskstrategy

03 July 2023

Climate change means different things to different people. Different people have different objective functions to optimise when it comes to climate change. Therefore, decisions on climate change also differs between people. Regardless of each person's views or opinions on the matter, the key remains to make a decision on how to proceed. 

Examples:

• Climate change is the E of ESG - compliance to standards and disclosure centric.

• Climate change is not important - no need to change course.

• Wait for others to act - no need to change course.

• It is the next person's problem - no need to change course.

• Profit is still king - tell me what I need to change.

Each of the 5 examples are valid decisions as they are made based on the objective functions agreed with their stakeholders. If the climate change discussion is solely on the risks, there will be resistance to change – fear tactics create resistance, fear is also exhausting. If more people focus on the opportunities, there will be interest to change – money piques or arouses interests. So to encourage decisions and actions, one ought to be honest about the risks, but pique interests with the opportunities to make more wealth. Self-preservation which seems to be the dominant narrative is only effective when people are desperate.

#climatechange #climaterisk

07 April 2023

Transition plans are something most organisations' avoid creating. A strategy yes, but a plan no. Where a strategy can be written in strategic terms (or broad terms), a plan requires details. Details in the form of targets, decisions, tasks, timeline, and numbers. It binds the author (e.g., senior management and the board of directors) to a course of action for which their performance will be assessed. In an era of of "not doing anything" is better than "making a mistake", no one wants to write a plan for fear of being wrong. 

To make it easier, organisations' can start with a more limited number of decisions, focusing on the strategic requirements (e.g., avoid mentioning tasks). As the organisation develops the capability and capacity to understand climate risk, and incorporating feedback from stakeholders, the organisation would update / enhance / improve the plan with more granular information (e.g., tactical requirements).  The organisation would also caveat the plan as subject to periodic review (like any business plan), and constantly being updated to reflect operating conditions.

A recommended information set to include in any plan:

1. Scenarios - The operating environment / condition.

2. Risk and Opportunities - The impact (positive and negative) from the scenarios.

3. Strategy - The description of the approach, direction, visions or intent.

4. Goals and Objectives - Targets / measures / performance indicators / outputs / outcomes.

5. Decisions - Decisions to be taken to the strategic goals (i.e. the intent).

6. Tasks - The specific activities or actions to be taken based on the decisions made.

7. Impact - Financial and non-financial impact from the strategy, decisions and tasks.

#climatechange #climaterisk

02 March 2023

Climate vs. Weather, different concepts that are often used interchangeably. In risk management, they are technically different. While both can change, changes in the weather is the season-to-season, or moment-to-moment changes, while climate is the long-term persistent drift from normal values. 

Short-term changes in the weather is not by itself an indication of climate change. An earlier winter this year does not imply the climate has changed if last year's winter arrived later and next year's will be arriving on time - such that the 3 season average moves around a stable point.  What is important however, is how the earlier, later, and on-time weather affects the business. In the context of banking risk management, this relates more to Physical Risk - how would changing winter start dates affect credit, market, operational and liquidity risks? 

Suppose the short-term changes includes a gradual shift away from the normal (or average). That is there is a dominant trend in the season-to-season, moment-to-moment change in the weather. For example, we might say the summers of today are cooler than my childhood 20 years ago. We will typically add context to the comparison by describing how it has affected the summer experiences, such as we used to stay indoors to avoid the heat, now we stay outdoors more. This is the essence of Transition Risk - how would the long-term change in the weather (i.e., climate change) affect the things we do, the decisions we make.

For banking risk management, the approach to climate risk would require an understanding of the differences between 1) Weather and Climate, 2) Time horizon between Physical and Transition Risk, and 3) The decisions required to manage the risks. What happens in the next 30-years requires an understanding of how things will develop, and how the development affects the bank's prospects. What happens next year requires an understanding of what events can happen, and how the events affect the bank's performance.

#climatechange #climaterisk

14 February 2023

Artificial Intelligence (AI) is the new hit thing after the release of Chatgpt. What can be confusing is, while Chatgpt demonstrates Artificial Intelligence, AI as a concept means replicate or mimic human intelligence. AI is actually not a new concept. Videogames have used AI for decades to replicate the experience of interacting with people in a virtual world. Within the videogaming world, AI has evolved from simple (or low level) intelligence with limited vocabulary and responses (e.g. pathfinding and NPC interaction), to more complex (or higher level) intelligence where "memories" of the players previous actions influence the vocabulary and responses available. 

The point being, AI can be applied to a variety of applications. My personal favourite is in the space of model risk management. Anyone that has worked in the model space will know the entire process is inefficient. Over the years and decades, model development and validation has become a standardised process. Models now follow a conveyer belt process in their development and validation - develop, validate, deploy, use, monitor, validate, develop - repeat cycle.

So why not apply AI to automate the process? Low level AI would simply be data preparation and running the battery of tests, with the method and rules specified by a human. Higher level AI (with the appropriate computing power) will seek out the method and rules based on the constraints specified by a human.

The role of the human would be to review, discuss and decide on the results. To apply the principle of proportionality to decide if a model if fit for use, fit for purpose. To balance the needs of the business with the needs of risk management (and prudential requirements). Over time this role would likely be replaced by higher level AI since AI can be programmed to eliminate self-preservation feelings when required to make a decision - to further improve efficiency and effectiveness.

26 January 2023

As we end the first month of 2023, we see changes. Estimated guesses of what would drive 2023, in no particular order of importance:

1. Hotter than usual European summer and warmer than usual winter - more conducive gas demand and prices.

2. Reduced drought conditions in CONUS - based on the wet 2022 winter in the United States.

3. US and Europe recession - doesn't matter if its deep or mild. 

4. China supports Asia Pacific growth (just like in 2008) - lockdowns which were the main growth disruptor in 2022, no longer the default policy, removing a big headwind.

5. Ukraine-Russia war will see a settlement - no one knows what the end-state looks like, but its not sustainable for all parties.

6. Political gridlock in the United States as nothing bi-partisan gets done - dysfunctional political landscape plus campaigning for the 2024 presidential election starts. 

06 January 2023

In the previous post, it was whether banks can be agents of change. In this post, we ask, whether banks should be agents of change. The answer is, it is up to the financials and shareholders'. Unless a bank is a non-profit, its sole purpose of existing as a business enterprise is to be profitable. The acceptable levels of profit depends on the shareholders' (i.e., the owners of the bank). 

What ultimately drives change is the market. Without demand for Green Finance, there is no reason for banks to supply. Banks are not the ones that are building coal power plants, oil refineries, or converting forests to farmlands. In some markets, those activities are the main sources of demand for credit (e.g. commodity based economies). 

So long as there is borrower, there will be a willing lender. The only thing banks need to do is measure the risks posed by climate change and price the risks accordingly - the basic framework that also applies to pricing sovereign risk, country transfer risk, default risk, etc. 

If banks commit to supplying Green Finance when there is no demand, it goes back to whether the financials make sense and whether the shareholders' are on-board. So can banks be agents of change? It can, so long as the financials make sense and there is buy-in from shareholders'. But it shouldn't be asked to be one if the market dynamics do not exist.

12 December 2022

The first in a 2 part post.  Basically, the banking system has been asked or tasked to act as agents of change. Most climate risk management requirements stipulate the need for banks to 1) Understand its customers transition plans and strategies, 2) Consider those plans and strategies in its credit approvals (and pricing), 3) Assist (or facilitate) the customers transition. 

Is this a reasonable request to ask of banks since it looks like a request to provide advisory services to its own customers. While it is not uncommon for banks to advise on deals, I would argue it is uncommon for banks to guide customers in their long-term strategy. 

Perhaps the approach is to inform customers on impending regulatory changes, so they can respond accordingly. Rather than the "be a friend and partner" relationship expected by the various regulations.