How an Objective Risk Model Can Improve Investment Decisions
The human mind is wired to make quick, instinctive decisions, resulting in biases. We make many decisions each day and, yet, most of us are unaware of the biases that influence these decisions. Biases also exist within businesses, based on leadership perspectives and company processes.
For example, at Origin, we acquire multifamily and office assets where we believe we can enhance value through capital improvements, cost-cutting and management change. But despite us following a very strict process to evaluate potential deals, we realized we faced decision-making biases that influenced what assets we acquired. In this article, I’ll share how we modified our asset acquisition processes to overcome these biases and to make better investment decisions.
First, some background. At Origin, we have acquisition officers who live and work in each of our target regions to provide local knowledge and connectivity to off-market deals. Our acquisition officers and analysts first thoroughly evaluate a potential asset’s location, vintage, quality, demand drivers, supply threats, growth rates, and expected price to determine if the property will be a lucrative investment for Origin and our 600+ investors. We also have an asset management team who will ultimately be responsible for executing on the asset’s business plan. It’s critical that our acquisition and asset management teams collaborate, and both sign off on a potential investment’s business plan.
Most opportunities don’t make it through our initial screening process. We review thousands of deals to only acquire about six properties per year. Opportunities that pass this initial screening process are then analyzed further through site visits and competitive building analysis, using both external data and our own proprietary research. Our deal team will then present a potential deal memorandum and financial analysis to our investment committee, if an opportunity passes the second phase. Only a handful of deals make it to this stage of the process. To put it in perspective, we received over 1,000 deals in 2017 but only acquired three properties. Roughly 700 properties made it through the initial screening where we narrowed down the list to 150 and, of those properties, only 20 were presented to the entire investment committee.
One area that we examined to overcome our biases is regarding how we evaluate potential deals at our investment committee meetings. We realized that my partner, Michael Episcope, and I would frame the discussion by leading the conversation. So we now make sure the broader team can showcase their expertise before my partner and I weigh in. We also found that it was best practice to get additional feedback privately, as some team members found they could be most honest when remaining anonymous. Furthermore, we tied the compensation of all acquisition officers and asset managers to the overall performance of all deals to incentivize them to spend the necessary time to vet all deals and provide honest feedback.
In late 2017, Dave Welk, who leads our acquisition team, suggested that we create an objective risk model to further mitigate risk when evaluating opportunities. The model would incorporate historical data on deal performance based on the asset class, city, sub-market, idiosyncratic risk, building attributes, and the structural risk of the deal. It took months to build the objective risk model and even more time to evaluate the inputs and weights. We then tested the model on deals that we were in the process of underwriting, as well as the 30 deals that we had acquired over the past five years.
The results were extraordinary; the model accurately predicted the outcomes of our buildings and their business plans and, more importantly, would have prevented us from acquiring our two worst-performing assets. We were very surprised at this result, considering that we have such a talented team and a thorough process for deal analysis.
In retrospect, I believe we didn’t recognize two inherent biases skewing our judgment until we implemented the new model.
Off-Market Deals Aren’t Necessarily Better than Marketed Deals
The first bias that affected our judgment was the perception that off-market deals will ultimately earn more money for our investors than marketed deals. We thought that since there were fewer participants bidding on the asset we were going to be able to secure the property at a lower price, and eventually earn a higher return on the investment. It’s easy for an irrational mind to feel that an exclusive look at a deal is a better opportunity than a broadly marketed deal. However, we found this was not always the case and that all off-market deals must still undergo the same rigorous vetting as on-market deals.
Being aware of this bias in other times, however, sometimes works to our advantage. We have sold some of our deals off-market because in certain instances buyers overpay for assets when they feel they have an exclusive opportunity, or they overprice the asset and do not have the broader market to help them realize their error. For example, we sold a Chicago West Loop multifamily asset in 2015 to an off-market buyer at a price we felt could not be attained through a marketed process. 24 months later, the buyer sold the asset at a 10% loss in a market that continued to strengthen.
Recapitalizations Don’t Guarantee Favorable Returns
The second bias that clouded our judgment was the perception that joint ventures guarantee favorable returns. In these instances, our deal team recapitalized the asset at a price lower than the asset would achieve if sold and negotiated favorable profit sharing and legal rights. However, we can’t let these advantages cloud our focus. Even when we negotiate the most favorable terms, our attention should remain on the total risk/return of the deal, because if profits are less than projected, investors will only receive a large percentage of a small pie.
Our risk model will continue to improve, as we test it with more and more deals that are underwritten and back-tested. This is not to say that we don’t value subjective discussion and the expertise of our team — that remains critical. However, our risk model will be used to evaluate every deal because it is able to objectively overcome our judgment-clouding biases.