When We First Saw the Grizzly, It Was Half a Mile Away Across a Meadow...

Nouriel "Grizzly" Roubini (as Andrew Samwick calls him) looks back at his successful forecast of Q3 GDP growth:

RGE - Q3 GDP growth dismal at 1.6%; expect further slowdown in Q4 and recession by 2007: The first estimate of Q3 GDP growth is a dismal 1.6%, sharply lower than the 5.6% of Q1 and the 2.6% of Q2. In July - when I first predicted a US recession in 2007 - I forecasted that Q3 GDP growth would be 1.5% at the time when the market consensus was 3.1%. Given the onslaught of bad macro news in the fall professional forecasters started to cut their Q3 forecasts from 3.1% to 2.5%, down to 2.2% last week and 2% this morning. They were still wrong and overoptimistic as the actual first estimate came as 1.6%, only an epsilon higher than my July forecast of 1.5% (the same forecasters had gotten Q2 wrong too; my spring forecast for Q2 was 2.5% versus a consensus of 3.2%; the actual figure ended up being 2.6%).

The weakness in Q3 growth is widespread: real residential investment fell at an annualized rate of 17.4%, much worse than the 11.1% drop of Q2; the trade balance was a negative drag on growth as the trade deficit widened sharply in Q3 relative to Q2; inventory accumulation was slightly lower in Q3 than in Q2 thus being a small drag on growth as well; non durable consumption grew only at an annualized rate of 1.6%; and while durable consumption grew faster in Q3 than in Q2 you can expect significant slowdown in durable consumption in Q4 as the glut of autos and housing related durables consumption takes a hit on the economy. Even non-residential investment in structures that was growing at an annualized rate of 20.3% in Q2 slowed down its growth to 14% in Q3: you can expect a much sharper slowdown in such non-residential investment in Q4 and 2007 for reasons discussed below. Real investment in software and equipment – that had fallen in Q2 – recovered in Q3 to a 6.4% growth rate; but further weakness in the economy in Q4 and 2007 will lead to a significant slowdown in such investment in the quarters ahead.

What do these Q3 growth figures imply for Q4 and 2007 GDP growth? Expect today the usual spin with the soft-landing optimists – who were altogether wrong on Q2 growth and even more wrong on Q3 growth – having already started to spin the fairy tale of a Q4 rebound.

This Q4 rebound has, so far, no base or data behind it: residential investment will be falling at a faster rate in Q4 than in Q3 given recent data on building permits and housing starts; non-residential investment that was, until now, growing very fast will sharply decelerate in Q4 and much more in 2007: see the lead story in the WSJ today referring to a McGraw Hill Construction study forecasting a rapid fall in construction spending in 2007 (including non residential construction and specifically stores and shopping centers), the first decline of construction spending since 1991. This weakness in residential and non-residential construction will directly affect retail activity where employment has already started to fall. Expect in Q4 and 2007 actual fall in durable consumption (autos, housing related consumption such as furniture and home appliances and other big ticket items) as the housing slowdown, the fall in home prices and the negative wealth effects of falling prices and reset of ARMs take a toll on consumption, especially housing-related durable one. Given the ongoing sharp slowdown in the economy, real investment in software and equipment will be growing less in Q4 and 2007 than in Q3. And both inventories and trade are likely to remain a drag on growth in Q4 as inventory adjustment will continue (with demand growing less than production) while a strong dollar will further widen the trade deficit in spite of the economic slowdown.

The first leading indicator of economic activity for October – the Philly, Richmond and Chicago Fed reports – are all consistent with a further economic slowdown in Q4 relative to Q3. I thus keep my forecast that Q4 growth will be between 0% and 1% and that the economy will enter into an outright recession by Q1 of 2007 or, at the latest, Q2...

Comments

[Dean Baker]

Yes, this is very much on the money. Most forecasters seem to believe that consumption will not slow even when people stop getting jobs and can no longer borrow against their homes. That ain't the economics I learned.

An important and overlooked item supporting the housing crash story is the new Census Department data on vacancy rates (released yesterday). They hit yet another record high in the 3rd quarter, with the movement all on the side of ownership units.

[RedCharlie]

Carlos Torres, writing for Bloomberg, reports that Q3 would have been 0.9% without an "annualized 26 percent increase in motor vehicle production" which is pretty obviously a one-time fluke due to auto-industry cutbacks and inventory clearance. (hat tip to kos poster bonddad) Auto manufacturing in the US, like most manufacturing, is in decline, not expansion. The "economic fallout from the auto-industry cutbacks will instead come" in Q4.

Funny thing is, I heard on the NPR show "Marketplace" that while GDP growth has slowed dramatically "most economists" feel other indicators are more optimistic, specifically because they expect business investment to rise. What they did not explain was why they expect business investment to rise if most businesses expect a recession or slowdown in 2007. Kind of a chicken and egg problem here. To avoid a recession, we need business investment to rise, but for business to want to invest, we need to avoid a recession...

[wcw]

For-sale-only vacancies have been a puzzle to me. On the one hand, they seem to be in a long-term uptrend as percentage of owner-occupied housing. I had thought this might be related to steadily increasing ownership rates in some way, that perhaps the marginal homeowners move more often and drive up the rate of empty homes for sale. On the other hand, that rate was not only already at its high in Q2 since the series began in the '60s, in Q3 it just showed its fastest-ever year-over-year increase. http://www.bignose.org/blog/index.php?/archives/120-Q3-for-sale-only-vacancies.html

Torres was reporting Joe Carson/AllianceBernstein's take. It seems to make sense, though.

The most rational soft-landing case I have seen was Tim Duy's Fedwatch. Like many I can't see housing correcting without some fallout, but I am not completely certain we won't get lucky and muddle through.
http://economistsview.typepad.com/economistsview/2006/10/fed_watch_soft_.html

The homebuilders are dead money at best for years, though.

[Tom Marney]

FWIW, Roubini turned bearish on a dime in July; see http://calculatedrisk.blogspot.com/2006/07/roubini-landing-hard-as-it-gets.html and my post in the comments:

Need a laugh? Check out Daniel Gross' "The 'R' Word
Are we heading into a recession?," http://www.slate.com/id/2146882/, dated today:

Gloomy forecaster No. 2
Nouriel Roubini, economics professor at New York University's Stern School of Business and proprietor of mega-macroeconomics blog Roubini Global Economics Monitor.

Likelihood of recession: 50-50, but timing is unclear. As he noted: "I do not now predict with high probability a recession but now believe that its probability has significantly increased to around 50 percent."
====================

Gross linked to a Roubini article from July 24th. What a difference a few days and the right report can make.
Tom Marney | 07.31.06 - 10:21 pm | #