I wonder how many people have looked at industry standard operating expense studies and simply accepted them as the truth, the whole truth and nothing but the truth. Dare I say 99 percent? That would be me in the 1 percent exception with a small number of other knowledgeable persons who understand the limitations of these studies. Not that we don’t use them, mind you, but we know the things that they’re not telling you. By the time you’re done reading this blog post, you’ll be eligible to join our little club. There’s no entrance fee, so there’s no need to fish dive your pockets for spare change.
What is “Industry Standard”?
By industry standard, I’m referring to the studies that are accepted at face value and are used so often by real estate practitioners, investors and owners that they become household words in an industry. Examples include Dollars & Cents of Shopping Centers by the Urban Land Institute, Income/Expense Analysis for Conventional Apartments by the Institute of Real Estate Management, the Experience Exchange Report by the Building Owners and Managers association and even lesser-known studies like the International Marina Institute’s Financial Operational Benchmark Study for Marina Operators. These publications and others are excellent and I use them frequently, but like any tool, if used incorrectly the results may not be what you expect.
Why Am I Letting the Cat Out of the Bag?
As an appraiser who has had attorneys try to discredit me by hammering home the flaws in operating expense studies as they were I quoted in my reports, I’m merely putting in the public venue what I’ve already put on public record. I originally stumbled on this about 15 years ago and then again in 2002, so you could say that I’ve wanted to speak my mind for a long time. My intention is to educate, not to prosecute.
The Gray Between the Black and the White
Most of the major property type industry studies show you multiple columns of income and expense numbers. Let’s take Dollars & Cents. We have a column for Average, Median, Lower Decile and Upper Decile. Taking this study further, you’ll find two sets of Median, Lower Decile and Upper Decile – one for dollars per square foot of GLA (gross leasable area) and one for percentage of total receipts. That’s 7 columns! Can you guess the first big problem? Yep, it’s selecting the most appropriate column. I’m sorry to say that you won’t get good guidance in these studies of when to select a particular column – you’re presumed to be the expert, I guess.
I know what you’re thinking – which one should I tell you to choose? I’m going to discuss a real-world example below that answers this question… yet doesn’t answer it. Don’t shoot me – I’m just the messenger!
I’d bet that most people would say select the average column without giving it any more thought. Maybe that is the answer. Maybe not. If you say it is, make sure you understand how the average fits in with the data. For instance, I’ve randomly selected a page from one Dollars & Cents study that has an average utilities expense of $0.87 per square foot; however, the median (most frequently occurring value of a sample) is $0.65 per square foot. The Lower Decile is $0.15 and the Upper Decile is $2.84. Given this, would you think that the average might be skewed higher due to the number of sample items between the average and the Upper Decile? Hmmm… the numbers you see are black on white, yet I’ve introduced some gray into the equation. This is just one of many types of comparisons that can be made among the dozens of operating expense line items shown in these studies.
Speaking of averages, IREM’s operating expense studies don’t have averages, also known in statistics as the mean. They have median, high and low. For those of us that like averages… oh, well.
Plugging Some Holes
So looking at these industry standard studies, you’ll typically find some white space in columns where there should be numbers. You won’t find that for averages or medians, but you will find it in whatever other columns are reported. As you might guess, that means there was no data for the line item in that category. For instance, from my random page selection, landscaping expense has an average and a median, but no Lower Decile or Upper Decile. To me, that says that you can’t really use Lower Decile or Upper Decile because you have incomplete data. You’re forced into using average or median. Well then, it seems that there isn’t much use for Lower Decile or Upper Decile. It wouldn’t make sense to fill in numbers with averages, at least not by my way of thinking. You’d be better just using the average column completely. What these missing “cells” have in common, most likely, is an insufficient sample size. On my random selected page, every single cell that is blank has a sample size less than 10 observations. So there is the cutoff – the designers of the study feel that 10 is the magic number to report Lower Decile and Upper Decile.
Majoring in Mathletics
For you math athletes out there, you’re probably wondering why we don’t see some statistics. You want standard deviation. You want variance. All your old number friends from Statistics 101 want some playtime. I’m sorry to say we don’t see much of that. Clearly, the standard deviation would tell us how reliable a particular line item really is. With the number of observations reported for each line item in Dollars & Cents, you’d most likely see a correlation in the line items with the most observations having the smallest standard deviations, but without reporting this number, there is no way to know which line items meet the user’s test of reliability from those that do not.
Part 2 of this series will discuss sample size, numbers that don’t add up and the importance of using totals. Part 3 will take all these items into consideration with a real world example.
John Simpson, MAI
