Housing Starts

Housing starts, industrial production and import prices. The only real market driving event is the massively overhyped 12:30pm MerkOzy press conference.

Housing starts and permits for June. As usual, the key driver will be Europe-based headlines. There is a tiny POMO closing at 11 am.

While this morning litany of economic news was modestly better than expected, it really did nothing to change the picture that the US is rapidly regressing into another recession. Initial claims came at 414K, better than expectations of 420K, but as always expect next week's number to be revised higher to 418K or so: last week's number was as always pushed up from 427K to 430K. Far more importantly, this is the 10th consecutive week in which the initial claims data prints over 400k. Bullish? Continuing claims was just worse than the consensus of 3,670K, at 3,675K, down from an upward (of course) revised 3,696K from 3,676K. The biggest change was attributed to New York state, where 4,060 fewer layoffs were seen in the construction, mfg and retail industries, followed by California with 2,510 fewer claims due to a "Shorter work week, as well as fewer layoffs in the service industry." So shorter work weeks are now economic positive. Lastly, on the claims front, the 99 week cliff is pushing ever more people from under the government subsidy wing as 115K people dropped from ongoing EUC and Extended Benefits. The EUC 2008 number is 3,293,507 compared to 4,798,009 a year ago: nearly 1.5 million people have now lost the weekly government check to sit around and look for jobs. Looking at housing starts, the seasonally adjusted annualized number for May was 560,000, just above the 541,000 from April, although below the 580,000 from May 2010. Consensus expected 545K new home starts for the month, or a 4.2% increase from the unrevised April number of 523K. In other words, the starts data was both a miss and a beat, depending on what the baseline used is: revised or unrevised. On an unadjusted basis, there were 55.6K units in May started on, with multi-family units jumping to 13.1K, the highest since September 2010's 13.2k. Lastly, the Q1 current account balance, a largely delayed and irrelevant number, came at -119.3 billion, on expectations of -130 billion.

Since the triple dip in housing was recently circumvented courtesy of QE2, and was "transitory" in theory today's subpar housing starts and permits data is the beginning of the quadruple dip. And subpar it was: starts came at 523K on expectations of 569K, down from revised 585K previously. Permits were also ugly, missing expectations by a comparable account, printing at 551K, with consensus of 590K(and the previous revised this time lower from 594K to 574K). In starts, annualized single-family units dropped from 415k to 394k, with declines in Northeast and South, and increases in the Midwest and West. The actual, non-annualized number of starts was 46.8k, with 36.2k in single family units. Completions increased modestly from 532k to 554k. And the most interesting number was the number of houses under construction, which hit a fresh all time low on an annual, seasonally adjusted basis, or 418k. At this point it is probably passe to bring up the Cramer clip calling the housing bottom back in 2008.

Following last month's second lowest housing starts number ever of 479,000, March came at 549K, on expectations of 520K. Incidentally last month's number was revised to 512K from 479K. This likely accounts for what many have attributed to a weather related plunge in starts. The biggest swing factor in the March number was in units started at 5 units or more, which came at 117 or about 15% higher than February's 102. On the other side of the starts number was the housing completions, which at 509K was the lowest number ever, even worse than January's 510. Of note is the completions number in one unit structures which came at what is probably a record low 374k. But since nobody cares about the actual final outcome, and merely about the first dig, Housing Permits came at 594K on expectations of 540K. Once again this number was rescued by multi-unit housing, as the structures with 5 units or more came in at 173k compared to 135k previously, and was the highest number in over a year. Anything to fudge the number.

Stagflation, bitchez. PPI in February doubled to 1.6% on expectations of 0.7%, compared to 0.8% previously, and 5.6% Y/Y! This is the largest increase in finished goods prices since a 1.9-percent advance in June 2009. At the earlier stages of processing, prices received by manufacturers of intermediate goods moved up 2.0 percent, and the crude goods index climbed 3.4 percent. On an unadjusted basis, prices for finished goods advanced 5.6 percent for the 12 months ended February 2011, the largest 12-month increase since a 5.9-percent rise in March 2010." And the stunner: "The index for finished consumer foods surged 3.9 percent in  February, the largest increase since a 4.2-percent climb in November 1974. About seventy percent of the February rise can be traced to higher prices for fresh and dry vegetables, which jumped 48.7 percent. Advances in the indexes for meats and dairy products also were major factors in the increase in the finished consumer foods index." And while inflation is everywhere, or at least for items people need, the housing market is now official dead. February housing starts came at 479K on expectations of 566K, a massive 22.5% collapse from revised January data, and the second lowest ever, better only to April 2009's 477K. Overall: complete stagflationary disaster, and one which means the Fed will use any excuse for QE3, inflation be damned.

More weak economic data today, as Housing starts were reported at 529K on expectations of 550K, another sequential decline from the prior revised 553K. The silver lining was in the housing permit number which was 635K on expectations of 554K (compared to a prior revised 544K). Yet as the note from GS below explains the only reason for the surge in permits is due to a jump in applications ahead of the implementation of new building codes in 2011. As Hatzius notes: "If building code changes are the main explanation for the rise in permits, we should see a substantial drop back in multifamily permits next month."

The October Consumer Price Index came at 0.2%, on expectations of a 0.3% rise, compared to a previous read of 0.1%. The CPI ex-Food and Energy was unchanged, also below expectations of a 0.1% rise. And somehow, despite what is obviously a massive campaign for "value deflation", the BLS once again saw a barely notable increase in food prices (and an outright decline in some): "The food index rose 0.1 percent in October after a 0.3 percent increase in September. The index for food away from home rose 0.1 percent while the food at home index was unchanged.... The fruits and vegetables group posted the largest decline, falling 0.7 percent, while the index for nonalcoholic beverages fell 0.5 percent. " Elsewhere housing starts and building permits both missed expectations by a wide margin, coming in at 519 (vs exp of 598K), and 550K (vs exp of 568). Starts plunged from a revised 588K the month prior. One wonders how this contraction for the builders will be spun.

August Housing Starts come at 598K, on expectations of 550K, as the bounce along the bottom is now nothing but noise. None of this is relevant as the most recent (July) existing home inventory number hit 12.5 months from 8.9 months prior, and even with that in mind, the starts number is the largest since April 2010. Which merely means that even more spare capacity will be added. Of course, with the GMAC Mortgage scandal front and center, this whole statistic may soon be quite irrelevant should foreclosures grind to a halt. Oh, and this being a US Census number, the prior number was obviously revised lower, from 546K to 541K. No surprise there.

Even as the PPI data came out as expected, both on a MoM (0.2% vs exp 0.2%), and YoY basis (4.2% vs exp 4.2%), from a previous reading of -0.5%, and thus serving as no market moving indicator in either direction, housing starts of 546k came in well below expectations of 560k. And in keeping with tradition, the US government once again revised the prior period data, to make today's print seem like an improvement: the previous reading of 549k was revised to 537k. As the Census Bureau reported, "Privately-owned housing starts in July were at a seasonally adjusted annual rate of 546,000. This is 1.7 percent (±9.7%)* above the revised June estimate of 537,000, but is 7.0 percent (±7.5%)* below the July 2009 rate of 587,000." Completing the trifecta of economic data, Housing Permits also missed expectations of 580k, coming in at 565k. Far less relevantly, we also find that "privately-owned housing completions in July were at a seasonally adjusted annual rate of 587,000. This is 32.8 percent (±6.8%) below the revised June estimate of 874,000 and is 25.4 percent (±7.3%) below the July 2009 rate of 787,000." In other words, houses really are not being built.

No surprises in the housing starts and permits data, which continued the disappointing recent macroeconomic trend. June starts came in at 549K on expectations of 575K, even as May data was revised down from 593K to 578K. Of course, with roughly 12 million vacant units still to clear out, 8 of which in shadow inventory, it is a little odd this number has not been at zero for about the second year running. Futures, somehow, ramp on this disappointing number.

June held good news for Housing, unless one looks at the full story.