By: Walter Bialas
The question has been coming up often these days about where office development stands in this cycle. A recent D Real Estate blog post articulated this question, but came up short of an answer. From our perspective, the office construction pipeline is beginning to feel pretty frothy. In the last few week we’ve updated our under construction / soon to break ground numbers a few times a week – and that number now totals 13.4M SF (up from 10.5M SF several weeks back). The real question here is that most of the additions are fully spec. We get it, we have reached that point in this cycle that if you do not break ground soon, you may be late to the party. There is also some limited anecdotal evidence that smaller, regional banks are playing in this space – so the higher level of due diligence brought by the big names may be lacking.
As we looked at the question, we realized that it is not all that easy to answer – and, impossible to set out in a single “chart”. So, this is the first of a three-part series that attempts to open a dialog that outlines where we stand today and, maybe, where we are headed.
The first step necessary here is a small piece of history. The chart below sets out our construction activity going back to the mid-1990s. We recognize that this is a long time, but went back this far to capture the last two real estate cycles here in Dallas. What is clear from this review is that our history gives us no clue as to where we are going. Quite simply, the last two cycles have been quite different and are not likely to be repeated.
Looking back to the beginning of the tech boom was the first time where Dallas delivered significant office space after the commercial real estate crash of the late 1980s. Combine that with the fact that the tech boom was a true “bubble”, where the demand for office space was artificially high (remember the days when a 10,000 SF tenant took 50,000 SF in anticipation of growth). As such, the office construction volumes seen over the 1997 to 2002 are not representative of a “normal” market. In addition, because so much product was underway, even though the crash happened in 2000 to 2001, we continued to deliver space well into the recession – simply because started projects had to be finished. This led to a run-up in vacancy from the an all-time low of 16.8% to 25% by 2003. Interestingly, while other many tech markets around the US saw horrific levels of negative absorption into the bust, Dallas was pretty tame with only 3 million SF of occupancy lost. The issue here was more about the almost 34M SF of spec deliveries over the boom.
This over-building positioned Dallas uniquely in the last up-cycle. Because we had to take our vacancy rate down, there was actually only modest construction from 2005 to 2009. As noted below, a mere 10.3M SF of spec hit the market during the period – or 2.7% of inventory. That compares to over 9% at the peak of the tech explosion.
We came out of the last cycle in decent shape. While other markets around the US saw a bleed, negative absorption here was less than 500K SF in 2009, even though vacancy did notch back up to 23.6% by 2010 (only 1.5-percentage points above its general low point for the cycle). Because we did not seriously over-build, when the economy rebounded, demand took vacancy down quickly and continued to push it lower. While vacancy has now begun to rise, it is still near an historic low.
Thus far into this cycle, we have had total deliveries of 8.9M SF, with spec coming in at 5.2M SF. While comparable to the peak level reached in the last cycle (2008), it is not all that much given 326,000 jobs added – and we still have a clock running on this economic expansion (remember, last time the buzzer sounded pretty much in 2008 or early 2009).
The key here is that our “under construction” number still looks good relative to the number of large- to medium-sized build-to-suits going on. The catch here is that not all of those build-to-suits are net-new to Dallas. In fact, except for Toyota, many will be pulling out of existing space scattered around the market. And, to that we must add in the back-filling that will be required for the anchor tenants in new buildings now underway. To make this chart easier to view, I’ve also attached it as a PDF.
In part 2, we look at how the newest spec buildings fared, both near-term and longer-term, going into the last downturn.