Every analytical model is built on assumptions which guide the final outcome. The challenge in modeling real estate returns, particularly in the hotel space, is the unpredictable nature of random variables on the property’s performance.
- Is a new hotel going to open next door?
- Could we achieve an occupancy of 80%?
- What if it peaks at 75%?
Using the appropriate framework to model hotel returns will not provide certainty but will certainly help us better understand the risk and return profile of a proposed investment.
This is where Monte Carlo Simulations are particularly useful
Monte Carlo Simulations can be used to tackle a range of problems in virtually every field from finance and engineering to particle physics. Essentially, a Monte Carlo Simulation runs several scenarios based on pre-determined factors which are driven by their assigned probability. Easily set up in excel or similar applications, the model produces a range of potential outcomes which can collectively be used to understand not only the likely outcome but also what happens in tail-end situations.
While they have been used for years to price derivatives, forecast stock prices or to model scenarios, they are rarely used in real estate and I have yet to come across someone who has used them to evaluate hotel investments. This is a surprise to me as they can be an especially useful tool in the investment process.
Applications in hospitality
Starting with a simple example, using variables such as Occupancy and ADR, the model can be set up to return the IRR for a thousand individual simulations. Using different iterations of the two KPIs based on the selected probability, a range of outcomes is produced. This example is illustrated below.
This graph shows the investment IRR based on 1’000 different combinations of ADR and Occupancy driven by the probability of various events occurring. In this example, an IRR between 13% and 17% is achieved 70% of the time.
In other words, the result is an analysis, which provides one with a more complete picture of the risks and returns of a proposed hotel investment. This is especially true compared to using a deterministic method (i.e. your best guess at hotel occupancy in 5 years’ time). The deterministic approach, however, is the one most commonly applied by investors today.
The basics of a hotel model
The simple example above focuses on ADR and Occupancy but there is no reason why one cannot layer in variables such as growth rate, food and beverage revenues, departmental profit conversion, payroll or even capex.
While setting this up, it is important to keep in mind the cost structure of hotel businesses with the difference between fixed and variable costs. For example, insurance costs are relatively static while housekeeping costs fluctuate with activity and should, as a result, be built in to the model appropriately.
Monte Carlo Simulations: a better solution
While producing a performance forecast for a specific investment, rather than just replacing an uncertain variable with a single number, a Monte Carlo Simulation might prove to be a better solution for understanding investments. With countless applications ranging from hotel revenue management to investment appraisal, Monte Carlo Simulations allow for a better assessment of future outcomes and the potential risks of any given venture.