A Building Cycle EndsPosted on Tuesday, February 2nd, 2010 at 5:09am.
As our chart here shows (click to enlarge), the years 2000 and 2002 saw 65 new SFRs completed and sold in the Hill, Sand and Tree Sections combined.
Of course, all of that building has now come to a near-halt. Speculative construction is all but dead for the time being, with the bulk of new projects you may see under way being custom builds launched by the owners. Lot values are dipping as builders have exited the market.
Advertisements for the few remaining examples of new construction increasingly tout the end of the decade's building boom as adding some kind of urgency for buyers who want the cache of "new" in their home purchase. Want new? Better hurry.
One ad even says, "Nearly zero brand new homes will be available for years to come as most developers are out of business."
All those ads are right, of course, as the cliff-diving red line above shows. Many of the sales of new construction that made 2008 the recent peak year actually represented homes intended for sale in 2007, but which rolled over to the next year after the local RE market turned strange. Similarly, the 30 sales in 2009 did not consist mainly of homes intended for the 2009 market, but were holdovers from previous years.
We're down to 13 newly built homes on offer today, by our count.
Yes, it's the end of an era, or, really, a cycle – one that's intimately tied to the health of the economy and the local real estate market. Right now, spec construction is on ice, but some day it, too, will return.
We have more data to share, but let's briefly explain the methodology.
Short version: If a home was built new, sold by the end of the calendar year after its occupancy permit was issued, and was reported on the MLS, it's in these charts. The year that the sale closed is the year in which it's recorded on the graph. (Long-winded nerdy notes at the bottom of this post.)
Sales of new homes in the Trees were in the high 20s early in the decade, dropped off from 2003-05 and roared back in 2006-08.
Given the time lag for new construction from lot acquisition to a closed sale, this pace appears to mirror the end of a late-90s boom, financial uncertainty around the dot-com bust, and later the impact of the aggressive flow of easy money by 2003-04.
Sand Section new construction follows a roughly similar pattern, though, with fewer sales, the data jumps around a lot. Were we to include condos/THs in the Sand, we suspect it would have the greatest overall new development in this period, with perhaps a less jumpy trendline.
The Hill Section saw the fewest new homes overall – befitting its smaller overall footprint west of Sepulveda. Peak years for sales of new homes were 2000, 2002 and 2003, without the same kind of bounceback to high levels of spec construction that we see in the other areas.
Now, perhaps you're wondering, as we did when first seeing the data from the early 2000s: If the turn-of-the-century years were already humming along at such a peak, when did the previous building cycle begin?
The vital caveat here is that pre-2000 data in the MLS are generally less reliable. For illustrating a trend, they may suffice.
What those data show is a kind of malaise in new construction sales in 1995-97, with what looks like a sudden jump in 1998.
Keep in mind that local median home prices declined consistently for 4 years in the early-90s housing slump. (See MBC's "The 90s: What a Drag" for charts & discussion.) That would be a major factor in lower rates of new building.
From July 1990-July 1994, prices fell 27.7% in non-inflation-adjusted terms. It was not until that trend reversed that spec projects would have much hope of being profitable.
The jump in new construction sales in 1998 probably reflects "go" signals from 1995-96, after a couple years of gently rising prices, with builders beginning to grab lots again, and projects intended for marketing in 1997-98 and beyond. (Your blog author wasn't here then and welcomes input and perspective on the previous cycle from those who were.)
It's that late-90s cycle that was possibly overheating by 2000-2002, before a very big new injection of cash everywhere fostered a new building cycle locally. We may look back at it some day as MB's Mediterranean Period, though we're open to better monikers for the period just passed.
Methodology: The data in this post come from an a search of MLS data conducted by a local real estate agent who wishes to remain anonymous. (Your blog author is not an agent.)
This search picked up new construction that sold before the end of the calendar year after each home's stated completion year.
Included in the search is any home that sold within the same year it was completed and an occupancy permit issued, or within the next full calendar year. Hence, a home completed in March 2002 that sold in November 2003 would be in this mix. A home completed in December 2007 and sold in January 2008 would be in the mix. In both of these examples, the sale year is within one year of the completion year.
Excluded from this search is any new home sold in a calendar year more than one year after its completion. Hence, a home completed in October 2000 that did not sell until February 2002 would not be in the data. A spec home that tried to sell, but rented out instead, would not be included. A lease-option purchase of a new home would not be included if the sale did not close in time. We don't know how many new homes built during this period were excluded from the data above due to time lags of this nature.
Additionally, new homes built as custom projects for homeowners would not be included in any of these charts. Nor would spec homes sold off-market (pre-market) that were not entered into the MLS database.
Finally, the data above are subject to the garbage-in/garbage-out problem – we're entirely reliant on faithful and accurate reporting of data by real estate agents using the MLS. Occasionally a completion date gets changed to make "new" construction seem "fresher" than it really is. Or data is not entered correctly. That's why this data can only be illustrative, not final.
A superior method to report these data would be to pull permit data from the city, but that's a research-intensive operation that MBC isn't currently set up to conduct. We hope you enjoy, and find useful, what we have been able to put together.
comments powered by Disqus