December 27, 2010

AP’s New-Home Industry Meme (‘Worst in 47 Years’) Is Demonstrably False; It’s the Worst Since World War II

In its reports about the U.S. homebuilding industry and new home sales, the Associated Press has gotten lazy and/or deliberately deceptive. In doing so, it is giving readers, listeners and viewers at its subscribing outlets a completely incorrect impression that the industry and market are getting off the mat after recently being in their worst shape, in their words, “in 47 years.” After identifying offending examples, I will demonstrate that industry activity and sales during 2010 have been almost undoubtedly at their lowest levels since World War II.

The following items, all from Thursday, demonstrate AP’s concerted attempt to limit the damage to “47 years” ago.

Christopher Rugaber, in a sad, predictably pathetic attempt (“Economy brightens as consumers spend, layoffs slow”) to describe how the economy is allegedly turning around in upbeat terms, wrote: “… people bought new homes at a seasonally adjusted annual rate of 290,000 units, the government said. … barely above the weakest pace in 47 years.

Earlier on Thursday, the AP’s Martin Crutsinger, as is becoming a habit (see the three demonstrated and yet unretracted falsehoods discussed at NewsBusters and BizzyBlog in September), falsely opened a falsely headlined story (“New home sales rise 5.5 percent in November”) by writing that “More people purchased new homes in November, though not enough to signal better times are ahead for the battered housing industry.”

No they didn’t. For the record, as seen at the last page of the latest related Census Bureau report (large PDF here), the total of 21,000 homes actually sold in November represents the lowest monthly figure in the 47 years and 11 months such records haven been kept. That number isn’t “more” than anything. Thus, Crutsinger’s statement is false, and cannot possibly be spun into being true. Chris Rugaber’s phrasing is technically accurate, Marty; your statement is just plain wrong.

Along the way, Crutsinger also wrote: “the increase follows a dismal October sales pace that nearly matched the lowest level in 47 years.”

A Google News search on [housing "in 47 years"] (typed exactly as indicated between brackets, sorted by date) reveals that the AP seems to be the meme’s “inspiration.” A third unbylined AP item about homebuilder share prices is at Bloomberg, and yet another story from the wire service about mortgage rates is at

USA Today’s Julie Schmit, in her odious ode to fact-free optimism (covered last night at NewsBusters; at BizzyBlog), is among those who have swallowed the AP meme: “New home sales data for November, out today, are also expected to show slight improvement after an October that made the third-worst showing in 47 years.” You’d never know from Schmit’s choice of words that the other two “worsts” both happened in 2010.

Conducting a searches through previous months this year at the indispensable, one finds additional items from November 24 (“Economic data reveal a hint of consumer merriness”), June 15 (“Homebuilders less confident in recovery”), and April 24 (“New US home sales rise 27 per cent in unexpected surge from record low”), all of which make an unclarified “in 47 years” reference.

One June 23 AP item found at Breitbart (“Housing market struggles without government aid”) got it narrowly right:

New-home sales for May came in at a seasonally adjusted annual sales pace of 300,000, the Commerce Department said Wednesday. That was the slowest in the 47 years records have been kept.

The “(since) records have been kept” qualifier is crucial. Without it, news consumers informed that the situation is the “worst in 47 years” will naturally believe that 1963 was dreadful, and that 2010 hasn’t been as bad as that year.

It simply isn’t so:

  • The first page of the Census Bureau’s Monthly Sales Report tells us that the worst individual month for sales in 1963 was 31,000. Nine of the 11 months reported in 2010 thus far have come in below that figure.
  • The highest month for sales in 1963 was July’s 62,000. 2010′s highest thus far: April’s 41,000, which was artificially juiced by the expiration of the $8,000 homebuyer’s tax credit.
  • The Bureau’s report on annual new home sales shows a total of 560,000 for 1963, the first available year. At 298,000 through November, 2010 will be lucky to come in at 320,000, almost 43% below 1963.

Further, it is fairly easy to infer that every year back to 1946 had higher new home sales than 2010′s projected total — even before adjusting for population, which would turn the proof into an utter rout.

The easiest years to prove are 1959-1962. Though the Bureau didn’t start reporting new home sales until 1963, it did begin reporting on monthly housing starts (large PDF here) and building permits (large PDF here) in 1959. What follows are the annual figures for single-family starts and single-family permits for 1959-1963, followed by annual estimates for 2010 based on the eleven months reported so far, along with the two years for which sales data where available (note that I excluded two-family units, many of which are often owner-occupied; adding an estimate for such situations would make the case even stronger):


It’s no leap of faith to conclude that home sales in 1959-1962 were roughly at or slightly below that of 1963, except for 1959, which was probably much better. In all four years, annual new home sales were clearly way higher than the roughly 320,000 we’re going to see by the end of 2010.

For those keeping score, I have already proven that 2010′s new home sales are the worst in over 50 years. The “in 47 years” stuff is out the window. Even “the 47 years records have been kept” term is really not true, because, unless someone wants to show how hundreds of thousands of homes each year got started but were never finished, related records going back four more years show that sales in 1959-1962 were also much higher.

That home sales in each year from 1946-1958 were higher than 2010′s projected total can be virtually proven as follows:

  • The number of households in the U.S. grew from 34.86 million in the 1940 Census to 42.86 million in 1950. Because of World War II, most of the growth in households occurred during the last half of the decade. Conservatively estimating that six million additional households came to be during those 5 years, and that half of that increase could only have been absorbed by the building of new homes, it should be obvious that far more than 400,000 single family homes a year were built and sold, let alone 2010′s pathetic tentative total of 320,000.
  • The number of households in the U.S. grew from 42.86 million in 1950 to 52.04 million in 1959. That’s an increase of almost 9.2 million. Again, it should be obvious that 400,000 or more new homes would have been built and sold each year during the 1950s, even before considering the corroboration in the next two items.
  • In case anyone thinks there was a meaningful interruption in homebuilding between 1946-1959, relevant Gross Domestic Product data tells us that after a 319% increase in 1946, private residential investment (i.e., new dwellings) far more than doubled in real terms during the next 13 years, with no huge decreases that would have represented a dip to anywhere below 400,000 in annual sales.
  • Finally, in case anyone thinks that the investment involved might have been heavily weighted towards apartments and multi-family dwellings, note that the percentage of home ownership in the U.S. rose from 55.0% in 1950 to 61.9% in 1960.The only way to accommodate that type of increase would have been to build and sell millions of new homes.

C’mon, guys. No one can reasonably believe that 2010 hasn’t been the worst on record for new home sales and the homebuilding industry in general — and, as noted earlier, that’s before any kind of per-capita or per-household comparison. The “wonders” of the 2010 version of Obamanomics, particularly its utterly mismanaged Home Affordable Mortgage Program (HAMP), have taken us to depths never before seen.

So here’s the message to the folks at the Associated Press, its subscribers, and other establishment press outlets: From this point forward, unless you can prove otherwise, in discussing this year’s housing market and comparing its results to previous data, you would be more accurate to characterize 2010 as “the worst since World War II” (you know that if a Republican were in the Oval Office, you would be using this language). If you instead wish to refer to a period during which records have  been kept and reported, as demonstrated earlier, you should refer to something along the lines of “the over 50 years related records have been kept.”

Now that sufficient proof has been provided, any establishment media attempt to water down the assertions in the previous paragraph will from this point on represent a deliberate exercise in news consumer deception. In the AP’s case, any such attempt will represent a willful violation of the wire service’s pledge in its “Statement of News Values and Principles” not to “knowingly introduce false information into material intended for publication or broadcast.”

The establishment press now has no excuses (I know you guys read these posts; don’t try to pretend you don’t). Unless you can prove otherwise, 2010 was the worst new homebuilding and new home market since World War II, period. Stop your stupid spinning, start reporting and start doing decent analysis for a change.

Cross-posted at



  1. [...] That’s not so. 2010 was 43% worse than 1963, and worse than every full year after Japan blessedly surrendered to us–even before adjusting for population. [...]

    Pingback by BizzyBlog — January 2, 2011 @ 8:44 am

  2. [...] couple of weeks ago, I demonstrated that the homebuilding industry is at its lowest level since World War II — even before [...]

    Pingback by BizzyBlog — January 12, 2011 @ 9:21 am

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.