Data InSight



Supply and Demand, More Than Oil, Explain Houston’s Warehouse/Distribution Market, with Key Differences Among Northwest, North, and Southeast Submarkets

 

February 2016 I Vol. 2, ISSUE 3 I Download PDF

Executive Summary

Though not weakened as much as the office and manufacturing markets, oil prices can influence Houston’s warehouse/distribution market. Specifically, our analyses show that demand (net absorption) for warehouse/distribution space decreases with declines in Texas rig count, but importantly the oil industry only explains about 21% of net absorption. Yet, the oil industry, as measured by Texas rig count, does not explain vacancy, indicating that current availability of the warehouse/distribution market is shaped by other factors. Here, we show that supply (deliveries, stock) and demand (net absorption) in recent years have shaped current vacancy and availability of warehouse/distribution space, with important differences among submarkets.

Houston has 398 million sq. ft. of warehouse/distribution space spread across 30 submarkets and 10 submarket clusters. About 62% of this stock inventory occurs among just the Northwest (115 million sq. ft.), North (68 million sq. ft.), and Southeast (64 million sq. ft.) Corridors, which differ greatly in their supply and demand. The Northwest market has remained relatively stable with new supply (deliveries) balanced or exceeded by demand (net absorption) from 2006 - 2015, resulting in vacancy at low levels of 3.7% - 5.4% with a modest increase to 6.7% in 2016 (Figure 1). The North Corridor, on the other hand, has experienced great growth (new deliveries) since 2011 that has not been matched by net absorption, leading to an overbuilt market whereby vacancy increased from 4.9% in 2012 to 9.3% in 2015 (Figure 1). In contrast, following an 11.6% peak in vacancy in 2009, the Southeast Corridor has strengthened each successive year with demand exceeding new deliveries, resulting in currently low 3 - 4% vacancy (Figure 1). Collectively, these analyses show that, while the oil industry can impact short-term net absorption for warehouse/distribution products, it is the long-term supply and demand that shapes Houston’s market as a whole.

Motivation

Houston’s office and manufacturing real estate industries are strongly tied to the oil industry (see Data InSight Vol. 1, Issues 1, 2, 5, 7, 10), but it is unclear how tightly coupled warehouse/distribution market is with oil. Moreover, substantial differences appear to occur among submarkets of Houston’s warehouse/distribution market. For example, Houston has 398 million sq. ft. of warehouse/distribution buildings spread across 30 submarkets and 10 submarket clusters, with 62% of this stock inventory occurring among just the Northwest (115 million sq. ft.), North (68 million sq. ft.), and Southeast (64 million sq. ft.) Corridors.

It is often assumed that recent slowing in some submarkets is the result of being tightly coupled with oil, yet other submarkets seem to be growing and balanced in their supply and demand. Here, we examine the extent to which Houston’s warehouse/distribution market is influenced by the oil industry, and why Northwest, North, and Southeast submarkets appear to be diverging in their supply and demand. We also examine how the availability of different sized buildings differs among these three submarkets, including those <25,000 sq. ft., 25,000 - 49,999 sq. ft., 50,000 - 99,999 sq. ft., and >100,000 sq. ft.

Oil Industry Contributes Slightly to the Warehouse/Distribution Market

Vacancy and net absorption are two basic measures of the supply and demand of CRE. Percent vacancy is the fraction of the stock inventory that is not occupied by tenants. As a measure of new demand, net absorption is the change in the occupied stock of inventory. Here, we test whether changes in the oil industry, as measured by Texas rig counts, is predictive of net absorption or vacancy of Houston’s warehouse/distribution market. Vacancy was not associated with

Texas rig count, but as we detail below Texas rig count did contribute to net absorption. Texas rig count is predictive of net absorption of warehouse/distribution space in Houston (Figure 2). Increases in rig counts lead to increases in demand (net absorption) for warehouse and distribution products. The coefficient of determination (r2) indicates how well the data fit the statistical model in Figure 2. In this case, r2 = 0.21, that is 21% of variation in net absorption is explained by Texas rig count, leaving a substantial 79% of variation in net absorption explained by other factors. This indicates that there is a lot more at play than just Houston’s oil industry in shaping demand for warehouse and distribution space.

The solid red line in Figure 2 is the linear regression model of the statistical relationship between net absorption and Texas rig count. Specifically, Log(y)=0.0012x+0.846, where log(y) is the logarithm of net absorption in millions of square feet, x is Texas rig count, m is the slope of the line, and b is the y-intercept. The dashed blue lines are the 80% prediction intervals (upper and lower bounds) for net absorption. That is, there is an 80% probability that net absorption will be in this range for a given level of Texas rig count. (Note, we can back transform the log of net absorption to get actual measures of net absorption through exponentiation [i.e., exp(log(y))].)

The slope of the line, m = 0.0012, describes how y changes as x increases, that is an increase by 1 unit of the x variable increases the y variable by how much. For statistical reasons, the y variable of net absorption was log transformed (log millions sq. ft.). Because the y-axis is in log units, the slope of 0.0012 can be multiplied by 100 to estimate a percent change in net absorption for a given change in Texas rig count. Specifically, an increase of one rig results in a 0.12% percent change in the average annual net absorption. Or, a change of 100 new rigs equates with a 12% change in annual net absorption.

Vacancy Shaped by Past Deliveries and Net Absorption

In commercial real estate, vacancy is shaped by both supply and demand. In turn, supply is dictated by both existing stock and past deliveries, as well as current deliveries. Demand is measured by net absorption, the change in occupied stock and inventory. In this way, vacancy can greatly increase (or decrease) depending on the net absorption of new deliveries that add to existing vacant or occupied stock. Also, deliveries year over year under low net absorption can lead to increases in vacancy. Here, we examine how changes in supply and demand have differentially shaped the vacancy of warehouse/distribution space in recent years for the Northwest, North, and Southeast Corridors (Figure 1).

The Northwest market has had low vacancy since 2006, with only a modest uptick in 2016 with the large deliveries in 2015 (Figure 1a). From 2005 - 2008, the Northwest corridor had a modest surge in new deliveries each year, which was balanced by annual net absorption that matched annual deliveries. This kept vacancy at 4 - 6%. From 2012 - 2015 there was another surge in construction and new deliveries which was also largely balanced by net absorption, with the exception of 2015, leading to a modest uptick in vacancy to 6.7% in 2016. Overall, the Northwest market has remained relatively stable with new supply (deliveries) matched by demand (net absorption) from 2006 - 2014, resulting in low vacancies of 3.7 - 6.7% (Figure 1).

With balanced supply and demand in the Northwest market, construction and new deliveries have spread to and increased in the North corridor (Figure 1). Specifically, following few deliveries, modest net absorption, and falling vacancies from 2009 - 2012, substantial increases in new deliveries occurred from 2012 - 2015 which have not been balanced by net absorption. This has led vacancy to climb from 4.9 to 9.3%. Thus, the North Corridor has experienced great growth (new deliveries) that has not yet been matched by new demand, leading to a modestly overbuilt market with high vacancy in 2016.

The Southeast corridor, on the other hand, appears to be under-supplied with reasonably strong demand. After substantial construction and new deliveries from 2005 - 2009, vacancy in the Southeast market climbed to a high 11.6%. From 2010 - 2014, the Southeast market saw few new deliveries, balanced by net absorption, leading each successive year to decline in vacancy to a low 3.4% in 2015. The Southeast market remains strong, whereby supply and demand are in a balanced growth stage.

Availability among Submarkets

With three major industrial submarkets differing so greatly in their past and current supply and demand, we further examine how the availability of various sizes of buildings differ among these three submarkets, including those <25,000 sq. ft. (Figure 3a), 25,000 - 49,999 sq. ft. (Figure 3b), 50,000 - 99,999 sq. ft. (Figure 3c), and >100,000 sq. ft. (Figure 3d). Figures 3a-d show availability as a percent of stock inventory of that building size category. The Southeast and Northwest markets have similar availabilities for the different building size classes, while the North market has substantially greater availability, and in the case of >100,000 sq. ft., excessive availability nearing 20%.

Methodology

Commercial real estate data on industrial space were obtained from CoStar in early March 2016. The statistical analyses and data visualization were performed using the R software and programming language: R Core Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/. We used linear regression to examine the predictive effects of annual changes in Texas rig count (from Baker Hughes) on annual total net absorption (direct plus sublease) from 1999 - 2015, with log transformed absorption to improve normality. Assumptions of linear regression that could render a biased statistical model were tested. None of the assumptions were violated, including statistical outliers, unequal variance, heteroscadascity, and serially correlated residuals (nonwhite noise error).

J. Nathaniel Holland, Ph.D., Chief Research and Data Scientist

Dr. J. Nathaniel Holland is a research scientist with 20 years of experience in using the scientific method to extract information from complex multi-dimensional data. He joined NAI Partners in 2014 as Chief Research and Data Scientist. At NAI Partners, Nat leverages his sharp intellectual curiosity with his skills in statistical modeling to guide data-driven business decisions in commercial real estate. Like many data scientists in the private sector, Nat joined NAI Partners following a career in academia. Prior to taking up data analytics at NAI Partners, he held professorial and research positions at Rice University, University of Houston, and the University of Arizona between the years of 2001 and 2014. Nat is the author of more than 50 scientific publications, and he has been an invited expert speaker for more than 60 presentations. Trained as a quantitative ecologist, he holds a Ph.D. from the University of Miami, a M.S. from the University of Georgia, and a B.S. from Ferrum College.

Case Studies

Research