Does storytelling have a role to play in financial and economic analysis? Many approaches to forecasting either narrowly assume that the future will simply be a repetition of past events or are based on a biased view of how the future will unfold. Scenario analysis helps business decision makers plan strategy by imagining and presenting alternative future scenarios, discussing the variables that drive those scenarios and allowing for debate regarding how a business would be affected by different future conditions.
Nostradamus’ Business Legacy
While most people look upon people who claim to be able to see future events with a healthy degree of skepticism, almost everything we do in life is based on some view of what the future will be like. We take the bus because we believe that it will bring us to our destination, we deposit money in a bank because we believe we will be able to withdraw it when we need to, we make an investment because we believe it will provide a certain rate of return.
Although it is not the job of many people to create complicated financial or economic models that try to predict future global scenarios, essentially every person in business tries to forecast whether the demand for his product will go up and down, what will happen to prices, what will occur in the labor market and what its competitors will do. These are very important calculations to make in order to be able to run a business, seize opportunities and steer clear of risks.
Back to the Future
Given how central making predictions is to running a company, it is sensible for businesses to critically examine their own approach to forecasting, consider alternatives and adopt what is most useful.
At least in the field of financial analysis, a lot of forecasting is based on assuming that the future will be similar to the past which is often true due to the facts that human biology and nature does not radically change, laws of economics can be bent but are very difficult to break and a number of factors combine to reduce the infinite number of choices that humans could make into generally predictable patterns.
Despite this, while general statements can often reasonably be made about broad business economic probabilities, these statements are often of limited use in describing the specific set of circumstances that will affect a particular business. This is a vitally important gap, because it is precisely in the space between comfortable mathematical generalities and messy realities that businesses face the conditions that will cause them to thrive or fail.
One of the most common types of forecasting is regression analysis, which in simple terms calculates the relationship between two or more variables and then projects that relationship onto the future. For example, if one assumes that every middle class family will buy a new car and ten new middle class families will be formed, one can forecast that ten new cars will be purchased. Of course in forecasting practice multiple variables can be used and these variables may interact in ways that are more complicated than the car demand example.
While approaches such as regression analysis sound sophisticated and can be quite useful, they often suffer from the reality gap drawback highlighted above. This is due to a number of reasons, including selected variables may represent a small portion of the factors that affect the issue to be forecasted, the relationship between the variables may change and businesses may not be able to recover from particular moments of distress (such as a capital shortage or the loss of client) that deviate from the smooth future path suggested by a regression or other forecasting equation.
An Overview of Scenario Analysis
One way to strengthen forward-looking decision making is through a technique known as scenario analysis. A fundamental premise of scenario analysis is that since that it is not possible to predict the future with certainty, an alternative approach is to try to imagine different possible future scenarios, set forth the factors that would lead to the development of those scenarios and analyze how likely it is that the scenarios will actually occur.
The development of these scenarios requires a combination of a great deal of imagination, lateral thinking, detailed mapping of cause and effect relationships and quantitative analysis. To prepare a forecast of real estate rents, for example, we might start not by forecasting a specific rent but rather setting forth a defined range of possible rents. This band could include current rents, rents that are +/- 5% of current rents and rents that are +/- 10% of current rents.
Once we established those rents, the next step in scenario analysis is to imagine what types of scenarios could occur to produce those different rent levels. There is no one single way as to how this could be done, but a reasonable first step would be make some assumptions about how demand and supply imbalances affect rent levels. Once we defined this, we could try to consider what degrees of imbalance would produce the different possible rent outcomes that we have forecasted.
Next, we would need to establish different demand and supply side scenarios. On the demand side, for example, we might consider such factors as population growth, immigration rates, GDP growth, family size, levels of savings and disposable income. On the supply side, we might consider currently existing real estate stock and vacancy levels, new projects under construction, available land for construction, zoning restrictions, time required to receive construction approval, availability of construction finance and the size and competitiveness of the construction industry.
Once these scenarios were established, we would then have to consider how likely the scenarios were. If a demand-side scenario that led to a point at the higher end of the rent forecast was heavily based on an assumption that GDP would grow sharply, this scenario would like need to be discarded if the country’s economy was heavily based on commodity exports and commodity prices were sharply falling.
Similarly, if a short supply scenario was based on the assumption of very limited space for construction, this would need to seriously reexamined if the government announced plans to invest in a major transportation project that would make commuting from suburban areas to the city much more convenient.
Additional Benefits of Scenario Analysis
Scenario analysis has several benefits. In addition to helping with forecasting, scenario analysis helps create discussion within a firm as to what might occur and how future events could affect a company. Doing this effectively generally requires input from people who have different perspectives and different types of expertise. Creating this type of robust internal dialogue about what will happen in the future can help firms avoid the trap where only a single view of the future is accepted. This type of view can cause firms to be blindsided and have difficulty reacting when unexpected events occur.
Conclusion
Making decisions based on what will occur in the future will continue to remain one of business’ most crucial and challenging aspects. Approaches such as scenario analysis can provide a very useful complement to traditional forecasting approaches and tools.