This was fine as a thumbnail sketch, but there were weaknesses in this concept. The employment sectors used by the State of Nevada for a given business do not always relate to what that business is using its occupied space for. A manufacturing company, for example, might take industrial space in which to actually manufacture things, but it might also take office space for its executives and accountants and such. Whether the company is in industrial or office space, it is still a manufacturing company, and its employees might – might! – be included in the manufacturing category, whether they are in industrial or office space.
I decided a few months ago to try to figure out, as nearly as possible, what percentage of jobs in these different employment sectors were tied to the different types (and subtypes) of commercial real estate. To do this, I downloaded random groups of companies from Sales Genie. This data included the SIC code for the company in question, its address and the number of employees at that location. I then cross-referenced the addresses to the properties in our own database, and thus determined the product subtype occupied by each business. It was then easy to discover how many employees from each SIC code were stationed in each subtype (and class, in the case of office properties). Granted, time restraints forced me to rely on random samplings of businesses, but I plan to revisit this process each year to further refine my data.
For an example, let’s look at industrial space. In the past, “industrial employment” was a sum of the following employment sectors: Construction, Manufacturing, Transportation & Warehousing and Wholesale. Now, I know that industrial properties take in a share of virtually all sectors of employment. Because of this, I know that industrial employment is actually larger than previously thought, and the drop in employment between 2010 and 2011 less severe.
The above table shows the change in estimated industrial employment between the old method of calculation versus the new method of calculation.
As a result, I think I have a much better idea now of how changes of employment impact the different property types, and therefore have an improved aid in predicting future demand.