How Does Houston’s Current Slowdown in the Industrial Market
Compare to that of the 2009 Great Recession?
May 2016 I Vol. 2, ISSUE 5 I Download PDF
In this issue we examine how the current slowdown in Houston’s industrial market compares to that of the 2009 Great Recession. Specifically, we compare changes in availability and leasing activity between these two downturns, including differences among flex, manufacturing, and warehouse/ distribution buildings. By indexing the data to the first quarter (Q1) of 2009 (right yy-axis in Figure 1) at which WTI had the lowest average monthly price during the Great Recession, we are more readily able to compare rates of growth (or decline) in availability and leasing activity, particularly given the different magnitudes of values of these variables among flex, manufacturing, and warehouse/distribution buildings.
The bottom in WTI prices is similar between 2009 and the current downturn, that is a 70-75% drop from peak to trough. Availability is currently 9.1% for all industrial products combined, which is about 1% less than the peak following the 2009 recession (Figure 1A). It took about 3-4 quarters following the WTI bottom in Q1 2009 for availability to increase and stabilize at 10%. We will likely not see availability stabilize for another 2-3 quarters given WTI’s bottom was likely in Q1 2016. Availability is much lower for flex space today compared to the downturn of 2009, but much greater for manufacturing than 2009. Leasing activity in the current slowdown has had more volatility than following the 2009 crash in oil prices (Figure 1B), more for manufacturing than flex and warehouse/distribution. Overall, the current state of Houston’s industrial market is within the bounds of the 2009 pullback, but some key differences occur among flex, manufacturing, and warehouse/distribution space that push these bounds.
Supply and demand are fundamental concepts in economics which shape the commercial real estate (CRE) industry. Demand is most commonly estimated by net absorption, but is also occasionally measured by gross absorption or leasing activity. Supply is measured by stock inventory, construction, new deliveries, and building permits. Stock inventory estimates realized supply, that is how much of a particular product a market offers, such as 275 million sq. ft. of office space in the Houston market. The other measures of supply are estimates of new supply anticipated to be added to stock inventory in the future. Note, vacancy — which is often used as a surrogate for supply — is actually a function of both supply and demand.
Recent updates by the U.S. Bureau of Census indicate that the top three fastest growing cities in the United States in 2015 were Houston, Austin, and Dallas/Forth Worth, with San Antonio being the 10th fastest. Texas dominated the U.S. in having 40% of the top 10 fastest growing cities. Here, we examine how the supply of office space in terms of stock inventory has increased over the past decade along side of population growth in these four cities
From 2006 - 2016, availability among flex, manufacturing, and warehouse/distribution buildings varied from a low of 2.6% for manufacturing in Q2 2008 to a high of 16.5% for flex space in Q2 2010. Similarly, leased space ranged from as little as 35,970 sq. ft. for manufacturing space in Q1 2007 to more than 7.4 million sq. ft. for warehouse/distribution space in Q4 2015. Comparing such data for variables of different magnitudes as these can be difficult for even the trained eye. For this reason, we indexed the data series as often occurs in economics. Indexing variables to a standardized value at a common point facilities their comparison by setting them equal to one another at that point. Indexed values can be calculated as x_i=(x_t/x_0)*100, where xi is the new indexed value, xt is the raw data value for given time period t, and x0 is the data value of the common standardized point. Indexing readily allows for more direct comparison of points through time and between variables differing in magnitudes, as they are all scaled relative to one another.
We indexed availability and leasing activity to the first quarter (Q1) of 2009, the point at which WTI had the lowest average monthly prices during the Great Recession (Figure 2). The left-hand y-axis shows WTI prices in dollars, while the right-hand yy-axis shows the values indexed to 100 in 2009 or $39 WTI price. In this way, changes in WTI prices (and indexed values in general) can be interpreted as percentages. For example, in February 2009, WTI prices were about 25% greater (i.e., 100-75=25 on the yy-axis) than they were in February 2016, whereas February 2009 prices are about equal to April 2016 prices (i.e., indexed values equal about 100). Likewise, changes in availability and leasing activity over the past 10 years and among the different building types can be interpreted as percentages. Below, we discuss shifts in the rates of growth of availability and leasing, and how they are similar and different between the two oil collapses and among industrial products.
Availability measures the supply of industrial real estate as measured by vacant, occupied, direct and sublet space, typically expressed as a percent of stock inventory. Figure 3A shows how flex, manufacturing and warehouse/distribution space differ in their rates of percent availability. Flex space tends to maintain much greater percent availability, between 10-16% than manufacturing and warehouse/distribution space. In contrast, manufacturing maintains the lowest percent availability, between 2-6%, only recently jumping to 7%. Warehouse/distribution space tends to sit in the middle with percent availability between 6 - 11%.
Without a standardized benchmark, it is difficult to compare changes in percent availability among these buildings. Figure 3B depicts percent availability indexed to Q1 2009 equal to 100. With all three building types indexed to Q1 2009 = 100, we can interpret changes in time as percent differences in availability in their rates of growth and decline. For example, manufacturing has a 7.2% availability quarter to date for Q2 2016 (Figure 3A), which is 248 when indexed to 100 for Q1 2009 (Figure 3B). This equates with a 148% increase (248-100) in manufacturing availability relative to the Q1 2009 bottom of WTI oil prices (Figure 2, Figure 3B). Thus, while manufacturing has the lowest percent availabilities (2-7%) (Figure 3A), when indexed to a common standardized point of Q1 2009, we see that manufacturing shows the greatest volatility and largest increases in availability of 40 - 148% in this current downturn.
Such growth in availability of manufacturing space is yuge compared with changes in availability of flex and warehouse/distribution buildings. Since 2015, availability in flex space has increased from 9.7 to 11.0% (Figure 3A), but this only equates with indexed values of 64 to 72 relative to Q1 2009 (Figure 3B), which is actually a decrease in availability of 28 - 36% compared with WTI bottoms of 2009. Even though availability in flex space has increased a marginal 9.7 to 11.0% over the past six quarters or so, it is faring reasonably well compared to the last oil downturn and what is currently being seen in manufacturing. Likewise, availability in warehouse/distribution space since 2015 has increased from 8.6 to 9.6% (Figure 3A), equating with indexed values of 95 to 107 (Figure 3B), which is a decrease in availability of 5% to an increase of 7% compared with WTI bottoms of 2009. In sum, today’s availability for warehouse/distribution space following the recent decline in WTI prices is on par with 2009, whereas manufacturing availability is much greater and flex substantially less.
Leased space or leasing activity measures demand, including all direct leases, subleases, renewals, and pre-leasing of new deliveries. Figure 4A shows how flex, manufacturing and warehouse/distribution space differ in their leasing activity. Warehouse/distribution space tends to have much greater leasing activity, between 4-7 million sq. ft. per quarter, compared to manufacturing and flex space which tend to be less than 1.0 million sq. ft. per quarter.
Figure 4B depicts leasing activity indexed to Q1 2009 equal to 100. With all three building types indexed to Q1 2009 = 100, we can interpret changes in time as percentages and differences in the rate of leasing activity. In this way, we see that while manufacturing tends to have the lowest leasing activity per quarter in terms of sq. ft. of building space (Figure 4A), it actually has the highest volatility relative from quarter to quarter compared with flex and warehouse/distribution buildings (Figure 4B). Manufacturing consistently has the highest highs and the lowest lows among the three products, as seen in Figure 4B but not readily deciphered in Figure 4A due to scale. Leasing activity has been going steadily down for flex buildings over the past two years, but it has been up and down for manufacturing and warehouse/distribution buildings compared to Q1 2009.
Commercial real estate data on industrial space were obtained from CoStar in early May 2016. The statistical analyses and data visualization were performed using the R software and programming language:
R C ore Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/.
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.