Why do private real estate investors seek out individual deals, instead of allocating to a diversified private real estate fund? I believe the fact that most individuals are familiar with real estate – we own homes and use real estate for work and consumption, after all – and its tangible nature provides people with confidence in their ability to identify the best real estate investments.
This is a trap.
Just because an individual is familiar with owning a home does not mean they are qualified to identify real estate investments that will produce high returns. Investors should avoid this trap by investing in well-run real estate funds or by educating themselves on how to analyze a real estate opportunity.
At Origin, we utilize financial models to analyze every real estate investment that goes into our private real estate funds. An investor who would rather not invest in a fund should either educate themselves on financial modeling or work with consultants who do. Even then, proper analysis requires market knowledge on rents, the price of the property versus its peer set, demand drivers and limitations on their ability to increase rents, and the assumed price at sale.
The financial modeling process is complex, and therefore many investors tend to focus on only one lens when analyzing a deal. For example, they may hone in on competitive sales in the market or properties with similar rents. Another investor may look at price per square foot versus replacement cost. Still, other investors focus on the capitalization rate of the building at purchase and sale.
Each of these methods is useful, but none work as well as using all of them together.
Further, the property’s post-acquisition business plan should be analyzed and each input defended. Every real estate, venture capital or private equity model assumes growth of revenue and profits after purchase. How much growth does the model assume and how will be achieved? Defensible inputs are key.
At Origin, we don’t simply run the numbers when analyzing a potential real estate investment for our funds. We also conduct a stress test of each input into our financial models, and calculate its effect on our investment’s outcome. We know that we won’t be 100% accurate in our business plan assumptions, and want to quantify what under-performance on an input will do to the investment outcome. Fund investors should demand a similar level of diligence from their managers.
Stress Test #1: The Business Plan for Origin’s Ground-Up Development
To familiarize investors with what a stress test should look like, I’m sharing the math behind a recent stress test we conducted on an asset in Denver. Origin focuses on value-add investments, but we will invest in opportunistic investments such as ground-up development if we feel the returns are outsized, relative to the risk. I have chosen a development for this example because it provides more variables in the model to stress. I’ll explain each input and why we chose the level of stress testing for each model input in the potential investment.
|RENT SEN||DEV SENSITIVITIES||RENOVATION SENSITIVITIES||VACANCY SEN||HIGH STRESS*|
|0% Rent Growth||20 Month Construction||18 Month
|10% Lower Starting Rents||20% Hard Cost Overrun||5% Higher GVAC|
|Total Dev Cost||$33,692,503||$33,692,503||$34,026,089||$34,366,945||$33,692,503||$38,746,874||$33,692,503||$33,692,503|
|Gross Sale Price||$46,508,921||$40,564,797||$46,280,844||$46,557,518||$41,066,516||$46,508,921||$43,557,100||$34,673,455|
* 0% Rent Growth, 5% Lower Starting Rents, 10% Exp. on Cap Rate
Base Case Column – The base case column represents the estimated base case assumptions and returns for the deal in a five-year hold. The model assumes that Origin develops the project in 18 months, leases the apartment units in 15 months and sells the asset in year five at $306 per square foot, or a 5.75% capitalization rate on rents in year five.
0% Rent Growth Column – Private real estate models assume future earnings growth. The key question for investors to consider is the defensibility of that growth rate. In our model, we assume 2.5% average rent growth through the five-year hold. This is based on two submarket and property-level databases that we license, as well as our own market knowledge. The four-year trailing average for the sub-market is 7% rent growth and 95.6% occupancy of multi-family assets. In this particular case, the Denver airport is expanding and Amazon is building a fulfillment center close to our development, which we believe will create strong demand for local housing. Lastly, the Denver light rail has been completed, linking our property to the greater Denver community and airport, another demand driver. Given this information, Origin felt that a 0% rent growth was an accurate stress test for the model.
20 Month Construction Column – Origin is building garden construction on a flat land parcel. Utilities and roads are in place, and constructions drawings and permits are secured. The general contractor in the business plan has built an identical apartment community in Salt Lake last year and it took them 12 months. Origin’s base case business model assumes 18 months of construction and the stress test assumes an additional two months to complete construction. While this seems like a large cushion, it should be noted that construction times to complete have lengthened, due to the lack of sub-contractors available as the economy has strengthened.
18 Month Lease-Up Column – Our base case model assumes that it will take 15 months to fill the building with renters after construction is complete. We’ve seen similar buildings take 12-15 months to fill up with renters and we believe the demand will outpace supply in the next 24 months. The business plan assumes that we begin to deliver in month ten (January 2019) and continue to deliver into the fall of 2019. The stress test adds an additional 20% to the 15 month base case assumption, in the event it takes us longer to rent out units.
10% Lower Starting Rents Column – Rent growth in all models begins with a starting rent assumption. If that rent is lower than expected, the returns suffer. Origin first mitigates this risk in development by assuming that we’ll lease up the building’s units at rents that are currently in the market today. We call this an untrended rent model, and this provides us with the additional cushion of the market growth rate for one year– the time to construct the apartments. In the stress test, we start rents at 10% below where the market rent rate is today, 10-15 months in the future. Not only are we not growing rents during the build out as the market data suggests, but we actually assume that rents fall 10%. This historically has been very unlikely, in Denver in particular.
20% Hard Cost Overrun Column – Origin works with the builder to understand the estimated construction costs. The closer the plans are to final, the more accurate the cost assumptions should be. In our example, our model was based on a signed general maximum price (GMP) with a general contractor that only allows cost overruns in cases of approved change of scopes to the project. All other risk is assumed by the general contractor to build to the specified quality, budget and time frame. Further, the plan set to bid the GMP contract was 99% complete, giving Origin confidence that the scope of work and quality of finishes were clearly defined. The General Contractor recently built an identical product on time and on budget, so we are confident that the builder can achieve the same success on our site. Origin also builds in a 4% construction contingency in the project budget, in the event something unexpected comes up during the build. We use a 20% overrun in our stress test on construction cost overruns, which is defensible, given the steps we outlined above.
5% Higher GVAC (Vacancy at Sale Assumption) Column – We estimate out how occupied we think our building will be at the point of sale. Our databases showed that the current occupancy in our project’s sub-market is 95.5% and has been for the past four years. Future occupancy is projected in the sub-market to be 95%, based on supply and demand projections. Origin also owns other properties in the market and this level of demand for rentals is evident at our own assets. So we stress test this metric by increasing vacancy through our hold and sale to 10%, which represents a 100% increase in the market, from 5% to 10% vacant.
High Stress Column – The last column combines all of these stresses together. It is highly unlikely that they all would happen together, as many of them do not correlate. For example, if rents are declining it is unlikely that construction costs are also increasing, as one happens in a recession and other when the economy is growing.
However, if all stresses happen together, we see that our equity multiple will decline from 2.6 to 1.4. This means that $1 million invested into the project will still receive $1.4 million in gross returns from the investment. The price per square on exit in this stress test has declined from $306 per square foot to $228 per square foot, a 25% decrease in exit price. The combined stress test gives Origin confidence that we have an investment that can withstand significant headwinds and still be profitable.
Stress Test #2: The Multiple on Earnings for Origin’s Ground-Up Development
|CAP RATE SENSITIVITY|
|Sale Price PSF||$335||$320||$306||$294||$282|
|Sale Price Per Unit||$339,589||$324,153||$310,059||$297,140||$285,255|
Besides stress testing a project’s business plan, we also stress test our estimated multiple on earnings. The capitalization rate is important in any real estate deal, as it is used to compute a multiple on net operating income in the year of sale. This equation derives the sale price in the sale year.
Watch David Scherer explain the number one mistake made by novice real estate investors.
For investors, it is imperative that the sale price of assets in the base case model are defensible through both capitalization rate and price per square foot methodologies. If the price at sale exceeds replacement costs, the deal better have a good reason for it. Good reasons include the escalating cost to build, inability to build due to zoning, or some unique, irreplaceable aspect to the building that will create competitive advantages over decades of investment.
Origin assumes a 5.75% cap rate in our base case model. Our base case capitalization rate on exit already assumes a 10% increase in the rate. We do this on a straight line basis and increase the rate 2% per year. We believe it is best practice to be conservative, as this is a market driven rate that Origin has no control over, based on investor appetite for risk, the risk free interest rate, inflation expectations, and amount of capital seeking to acquire assets. We can execute the construction, lease-up, and management of the building, but we have little control over capital markets at sale.
The above matrix shows our returns in our base case at 5.75%, which assumes a 10% increase in capitalization rates at sale. We then create an upside and downside scenario around this midpoint. The estimates produce a 3.0x gross multiple for investors if capitalization rates remain where they are today throughout the hold. Importantly, if cap rates expand by 20%, investors would still receive a 2.3x gross multiple, over a five-year hold. This shows us that we can be confident this project will produce a good return for our investors, even in the downside scenario.
Don’t fall into the trap of blindly pursuing individual real estate deals. At the very least, before investing in a deal, an investor should be familiar with how to utilize financial models to understand the risks and returns of an investment, and also the market knowledge to analyze each input assumption and its defensibility. Investors can take it a step further by stress testing the financial assumptions in the models, as I walked through in this article. Or, outsource real estate investments to a qualified real estate fund manager who will conduct all the due diligence needed.