Why CRE Industry Data Is Easy to Misread

A recent opinion piece in Bloomberg by Justin Fox, author of The Myth of the Rational Market, discussed the shrinking amout of office space per worker that companies were providing. As a GlobeSt.com analysis suggested, there were a number of factors that created significant caveats in the data. One of them was that using a CoStar analysis as part of the mix might be a problem because the data sets, definitions, and methodologies might not match up with the other sources.

CoStar contacted GlobeSt.com and offered to discuss some of the issues that face working with data in CRE. National Director of US Office Analytics Phil Mobley says that one underlying difficulty is having to intersect the office space market with the workplace world where things actually happen.

“There are a couple of things that are clear,” Mobley says. “Office-using occupations are occupying less space per worker.” But then he clarifies the “very strict sense” in which CoStar uses the term occupying. An organization has physical control of a space. That doesn’t necessarily address the number of people working in that space. “When we say they are leasing more or less space per worker, we mean per [absolute count of ] workers. We don’t mean per worker that is physically attending.”

It’s the appearance of a standard problem in using data on any level. Sometimes definitions differ but use the same word. If many of a company’s employees work from home or are in the office only part of the time, then the actual space per worker in the office could be significantly different.

“For the most part in the workplace world, offices probably don’t feel dense because we have lower attendance on any day,” Mobley says. Office-using organizations will take opportunities “to rationalize how much they formally occupy based on how much they need in terms of number of people and the number of people at any given time.”

“With words like occupancy and utilization, you have to be very disciplined,” he adds. “You see the stories and reports of companies continuing to adjust their workplace policies, their attendance policies. Our take on this is that the trend is really about efficiency of asset usage.” It’s related to concerns about where the economy might be going. But while space is being downsized, hiring has slowed. Companies still struggle with attendance policies.

Part of understanding the picture is looking at a particular subsector of business size and property type. “If you’re a company that needs multiple floors in a high-rise, the amount of vacancy in a 50,000 square foot building is not going to impact your decision because it’s not going to impact your data set.”

Another complication is that the measures being used to describe usage, according to Mobley, all work on approximations. “You can’t use those to predict what is going to happen in leasing markets,” he says. “You can find relationships, absolutely. But companies are making rational leasing decisions based on their own specific needs.”

In short, it’s easy to look at mixes of data, put them together incorrectly, ascribe more accuracy than exists, and to draw mistaken conclusions.