I observed in my last article (
As in any endeavor, the right tools can make the task easier and the outcome better. With respect to financing decisions, the industry has adopted several approaches, each with severe shortcomings:
- The default (i.e., fixed rate bonds): this approach is fast but likely not optimal when there are big differences among the alternatives (and there are).
- Common sense or intuition: also fast but subject to cognitive biases like overconfidence, loss aversion and confirmation bias
- A matrix of alternatives including their costs, risks and benefits: can lead to analysis paralysis from information overload and poor outcomes due to vague risk/reward definitions
- Monte Carlo simulation: this requires heroic assumptions about the future paths of interest rates, volatilities and correlations, input into an opaque model that cannot be easily replicated or confirmed
There is a better way. BMO has designed the Risk First Framework (RFF) to be a decision-making process that avoids many of the pitfalls of the other strategies while promising advantages in confidence, speed, consensus-building, robustness and transparency. It includes six steps, diagrammed below:
The six step Risk First Framework. Source: BMO.
The first step in the RFF, naturally, is to understand the key risks. These can be aggregated into broad categories such as 1) risk of the issuer's interest cost rising versus benchmarks; 2) systemic interest rate risk; 3) mark-to-market risk; 4) acceleration risk and 5) commitment risk, i.e., the potential cost of restructuring fixed rate bonds.
The next step is to quantify the impact of each risk in a worst-case scenario. For example, to help measure systemic interest rate risk, one can estimate interest expense assuming that interest rates increase three percent. One also needs to define a base case scenario, which could have rates increasing slowly over time.
Now that the key risks, stress tests and base case have been defined, one must determine how much of each risk the organization can tolerate. For each risk/stress test pair, one selects a number representing the worst acceptable case or "risk budget". An equally critical task is to set a goal. In most cases, we want to minimize interest cost, or interest cost net of investment income, under the base case scenario.
Next, we compile an arbitrarily long list of potential financing strategies. We include choices along dimensions such as floating vs. fixed, short vs. long term, puttable vs. non-puttable, taxable vs. tax-exempt and public vs. direct bank placement; each potentially hedged with a floating-to-fixed swap, basis swap or fixed-to-floating swap, if appropriate.
A relatively simple spreadsheet can now be constructed to a.) measure the outcomes for each strategy, under each stress test and b.) calculate whether any of the risk budgets were exceeded. We eliminate the too-risky strategies.
The final step is to rank the risk-tolerable strategies by increasing order of interest cost in the base case scenario. The best strategy is at the top of the list. In many cases, it is possible to improve on this strategy via optimization and feedback from stakeholders regarding the assumptions and results.
The RFF is a transparent, repeatable and robust approach that avoids the pitfalls of standard practice. When stakeholders such as board members speak the same language of feasible strategies, goals, risk budgets and framework for making decisions, finance staff can be empowered to make subsequent decisions more quickly and with greater confidence.
At the end of the day, the outcome of virtually every decision of consequence is contingent on not just good execution but also the unpredictable responses of others, both inside and outside the organization. Perhaps the greatest benefit of the RFF for risk-averse financial managers is in the future. It may be that a different decision might have yielded a better outcome with the benefit of hindsight. However, even in this case, it will be evident that the organization had performed robust analysis and made its decision in a way consistent with its mission, values, policies, constraints and knowledge at the time.
