Commercial Real Estate (CRE) has been slow to adopt data-driven risk modelling, relying instead on spreadsheet-based models and guessed forecasts for key metrics such as interest rates, inflation, rental rates, vacancies, and capital values. While this tradition has served a relationship-driven market, institutional investors such as pension funds, banks, and insurers are increasingly demanding more than confident forecasts. They want measurable insight into how likely returns are to meet their objectives and a clear understanding of downside risks.
The Problem with Single-Point Forecasts
The traditional DCF (discounted cash flow) model delivers a single expected return, or at most, three cases (base, upside, downside). It assumes that risk can be captured by adjusting one or two inputs, such as vacancy or rental growth, and applying a simple percentage reduction for a "worst case."
“When sorrows come, they come not single spies, but in battalions.”
The flaw? Real-world risks do not arrive one at a time - they come in correlated waves. A shopping mall downturn does not just reduce rent: it triggers vacancies, higher re-letting costs, falling market rents, lower-quality tenants, and shorter leases, compounding the hit to cash flow. A single-point forecast cannot capture this chain reaction, nor can it reveal the shape and spread of possible outcomes.
Risk is About Probability, Not Prediction
In most fields that involve uncertainty - including aviation, nuclear safety, weather forecasting, and drug trials - risk is quantified using probabilistic models, not fixed-point predictions. CRE remains an outlier, adding ever more detail to single forecasts instead of embracing simulation-based approaches that reveal the full distribution of possible outcomes.
From a Single Number to the Shape of Risk
By running thousands of simulations with realistic, correlated assumptions about tenant defaults, market rents, expenses, and interest rates, we do not get one answer - we get a distribution of outcomes.
- Normal distribution: upside and downside risks are balanced.
- Wide distribution: higher volatility, potential for outsized gains but also bigger losses.
- Negative skew: higher likelihood of returns falling short.
- Positive skew: greater probability of outperforming the mean.
These shapes provide actionable insight, showing which assets contribute to portfolio stability and which add concentration risk.
Portfolio Diversification Requires Distribution Data
Single-point DCF models cannot show how one investment interacts with others in a portfolio under stress. Simulation-based models can quantify correlation of outcomes, revealing whether an asset truly diversifies risk or simply amplifies exposure to the same market shocks.
The Takeaway
Institutional investors gain far more from understanding the shape of potential returns than from overanalysing a single cash flow forecast. The question should shift from "What will I earn?" to "What is the probability I meet my objectives, and what happens if I do not?"
CRE must catch up with other areas of finance in embracing probabilistic modelling. The tools and data exist. What is missing is the willingness to let go of the false certainty that single-point DCF forecasts provide and replace it with insight into volatility, correlation, and diversification.