Property Classification: It’s Important to Get Right

  in   NextGen Data

“Beauty is in the eye of the beholder.” Margaret Wolfe Hungerford was believed to have first written these famous words in the mid-19th century. The idea is simple yet profound. It suggests that beauty is created by the observer. For example, some people might look at this painting by Mark Rothko and not appreciate the elegant artistry. But back in May of 2015 someone liked the Rothko piece so much they paid $46.5 million for it. Obviously the buyer beheld a beautiful and transcendent masterpiece. But others might look at the eight foot tall painting and see nothing more than a blue and yellow canvas. The point is, while majority consensus can often agree, beauty is still universally subjective. And the opinion is particularly biased through the eyes of the owner.

The same concept is true in real estate. Apartment owners take pride in the assets they own, and they should. Owners want rental rates to be competitively priced to their peers. But to benchmark accurately, the biases toward the subject property have to be replaced with true statistical measurement.

What are you getting at Brandon? I am talking about the property classification system that’s ingrained in commercial real estate (CRE). CRE has traditionally been split into asset classes based on quality. While the consensus among the CRE community is often similar, the science of classifying has remained largely subjective.

Here are some typical attributes of Class A (“Top quality”) assets:

  • Highest quality
  • Newest, modern architecture
  • Tallest
  • Highest price per square foot
  • Latest technology
  • Prime location, very accessible
  • Luxurious amenities and finishes
  • Professionally managed
  • And of course, highest rents

Anything less than the best is usually “downgraded” to a Class B or C. But the wide degree of subjectivity can lead to differing opinions of classification. While there is validity to using many of these attributes to classify, inherent biases can still arise.

One of the most common approaches to classification comes from the broker or operator side. This approach is largely arbitrary. Beauty is in the eye of the beholder.

Another common approach is classifying by strictly using building age. While this approach can be a useful general proxy for classification, it discounts characteristics like building quality, location, rental rates and other factors that are relevant to apartment grade. If a vintage asset undergoes extensive renovation, its rents could be far greater than other comparably aged properties in the area.

One more common approach is classifying based on price per square foot. The flaw here is the bias toward unit size. Unit mix is not taken into consideration. For example, if an asset has a high proportion of small one-bedroom units, the high price per square foot would skew the property’s class into a higher tranche. Likewise, a unit mix comprised primarily of large floorplans would have a lower price per square foot and downgrade the asset’s class.


MPF Research believes the greatest common denominator is a combination of variables. Our analytics team has built a method to classify objectively using a combination of the asset’s unit mix, rent per square foot and a proprietary statistical model. The model creates a fair market value for each floor plan at the subject property and weights based on unit mix to derive a property level fair market rent. The model then benchmarks the fair market rent and classifies based on the deviation from the market-aggregated rent. This approach removes the inherent biases and arbitrary clustering of properties with different floorplan mixes and unit sizes.

Why is classifying objectively important?

  • Truer benchmarking
  • Enables owner to pinpoint specific market strengths
  • Shows asset’s relative investment grade over time
  • More accurate forecasts that are specific to asset-class
  • Better development site selection

Said another way, a true apples to apples comparison using one common denominator will help investors make better business decisions. Furthermore, true classification becomes more powerful with GIS overlays with metrics like household income, capital markets data, employment growth and others. This adds another layer for actionable decisions.

MPF Research took things a few steps further by benchmarking to the subject’s market, submarket and ZIP code. This allows the subject’s property grade to change based on a micro or macro benchmark scale. For example, a 1990s high-rise in downtown Washington, DC might be graded as a B+ at the submarket level because of all the expensive new deliveries there. Zooming out, however, the subject’s high fair market rents relative to the market could push the grade up to an A. Conversely, let’s consider a 2000s mid-rise in the Denver suburb Littleton. The property might have an A- quality grade compared to its submarket or ZIP code. But fair market rents in Denver’s urban core and the pricey Highlands Ranch submarket could downgrade the subject’s classification to a B when benchmarked at the market level.

Asset classification is important. And until recently, classifying apartments was subjective and overly simplistic, according to the “beauty” that was perceived. MPF Research through its Investment Analytics platform takes a statistical approach to classifying and clustering similar assets and benchmarks them at the market, submarket and ZIP code level. At the end of the day, the market should determine the asset class, not the owner or broker.