Spreadsheet Vs. Reality
One of the more underappreciated dangers of real estate investing is that it is incredibly easy to be swayed by the underwriting model. Models and spreadsheets can make a bad deal look great and a great deal look mediocre. And while we know intuitively that models can be flawed, the allure of a precisely forecasted return in an uncertain world is very hard to ignore.
The truth is, the acquisition model is an educated guess at what the future holds, and it’s unlikely it will be 100% accurate. It may be too aggressive or too conservative, and if you are a long-term investor, you might not know the final outcome for years.
And yet the model is an important part of the decision-making process. It forces investors to think clearly about their expectations and assumptions, confirm existing income and expenses, and compare projected returns to other alternatives. It is a tool for determining acceptable pricing and appropriate leverage and identifying potential risks. And it is especially useful for understanding what key drivers will most likely determine whether the investment is a success or disappointment.
Known>Unknown
What is important to remember is that models are static projections made at a single point in time. They are built using known information combined with educated guesses about future conditions based on current facts. The more known inputs, the more accurate the projection. The more reliant on assumptions, the wider the range of possible outcomes.
Where it gets dangerous is when projected outcomes based on assumptions are treated the same as projected outcomes based on facts.
Think about a speculative office development project, where early on everything from final construction costs and timing to future lease rates, leasing costs, financing costs, and exit cap rates are assumptions. The range of potential outcomes is huge because there are so many unknowns at the outset.
On the other hand, the model for an existing, fully leased property with long term contractual income should be a reliable predictor of future cash flows, with any return variability coming from sale pricing or other capital markets factors. Yet investors might be tempted to treat the projections for these projects equally because both models spit out a number that serves to anchor expectations.
I’ve seen sophisticated investors shocked when parking income at an office property multiple years into the ownership period was below what was projected at acquisition, even though the parking projections were dependent upon assumptions around leaseup timing for a large vacancy as well as overall commuter volume in the surrounding area. A disappointing outcome for sure, but not out of the realm of possibility.
A much more impactful surprise – this time to the upside - came when the sale price of a client’s asset blew through even the most aggressive projections anyone could have made when an auction process got heated. In this case, they were the beneficiaries of a bidding war between multiple groups who had raised funds for this particular type of asset in this particular submarket and needed to show their investors they could execute. While owners might dream about situations like this, no one would include it in their underwriting with a straight face.
When weighing an investment’s potential returns, projections based on facts can be given a smaller margin for error than those based on assumptions.
The Map Is Not the Territory (Farnam Street)
We’ve all seen news stories of drivers who ended up in a lake because they were blindly following the GPS and not what was in front of them. In those cases, the map might show a road, but the territory says otherwise.
If the model is the map, the actual property and market conditions are the territory. Just because the underwriting assumes a certain rental rate for a lease renewal in year five doesn’t mean that rate will be achievable when the time comes. Market rents may be lower or have higher associated tenant improvements or leasing costs. Ideally, the rents are higher.
Depending on their incentives, asset managers might be tempted to turn down genuine lease proposals (at market rental rates) because they were below what was underwritten at acquisition years prior. They would rather have a space sit vacant for months than accept reality or admit the projection was wrong. While there can be good reasons for doing so (expecting a rebound, an undesirable user, or a pending sale), an unwillingness to face facts isn’t one of them.
Owners and investors need to recognize when the circumstances on the ground have changed, and the original model is no longer relevant. Just as the model and the underlying assumptions are updated to reflect new information, so should our expectations for future performance.
Simple Beats Complex
For some acquisitions teams, the allure of the complex, detailed model is strong. A multi-tab spreadsheet full of macros and pivot tables that incorporates every conceivable variable is a thing of beauty (at least in the eye of the beholder).
An old investing saying comes to mind: “If it doesn’t work on a napkin, it’s not going to work in a spreadsheet.”
In reality, a few key terms will drive the outcome of every investment, for good or bad. If the renewal rental rate for a major tenant is significantly higher or lower than anticipated, it doesn’t matter if the lease is signed one or two months earlier or later than predicted. If the success of a development project depends on obtaining a difficult rezoning approval in a reasonable timeframe, it doesn’t matter if the exit cap rate on sale is off by 25 basis points.
If the deal doesn’t work conceptually, the fanciest spreadsheet in the world won’t make it better. Knowing that layering assumptions on top of assumptions compounds the potential for errors (especially over time), the goal should be to eliminate as many assumptions as possible while focusing on the one or two inputs that will be the key drivers of return.
Sanity Check
Keeping these concepts in mind, when reviewing the projected returns on an investment it’s helpful to ask yourself a few key questions.
Do I believe that the assumptions in this model are reasonable?
Are they too aggressive, or on the contrary too conservative?
Do I trust the source?
How many inputs are based on assumptions versus facts?
What are the one or two key inputs that will determine success or failure?
If the assumptions are wrong, what does that mean for the investment?
To the downside – do I lose money or are my returns disappointing but acceptable?
To the upside – are the returns significantly higher than projected?
Does the person preparing the model have an incentive to overpromise?
The best underwriting models will be directionally right and precisely wrong. If we can eliminate as many assumptions as we can in favor of facts, update our expectations when new information becomes available, and focus on the key factors that will drive outcomes, we can step out of the spreadsheet and into reality.