ES -- DX/CL -- isee -- cboe put/call -- specialist/public short ratio -- trinq -- trin -- aaii bull ratio -- abx -- cmbx -- cdx -- vxo p&f -- SPX volatility curve -- VIX:VXO skew -- commodity screen -- cot -- conference board

Friday, August 01, 2008


worse than reported

gross domestic product for 4q07 was revised into negative territory yesterday, giving ammunition to those who believe that the united states has slipped into recession. certainly the jobs figures tend to corroborate that notion.

but 2q08 was reported as a difficult-to-reconcile 1.9% growth -- and a by-product of the figure was an extension of the interminable debate over whether we are or aren't. what gives?

in the end, the PCE deflator. per reuters, swedish economist stefan karlsson:

Looking at specific quarters, Q4 2007 official growth was revised down to -0.2% from +0.6% and Q1 2008 from 1.0% to 0.9%. However, the real numbers are in fact even worse than these numbers suggest, for two reasons.

First, there is the factor that I've mentioned several times before, namely the terms of trade factor. This was actually most dramatic during Q2 2008. That alleged 1.9% growth depended on the ludicrous assumption that inflation was just 1.1% at an annual rate. If you instead deflate the nominal GDP growth of 3% with the 4.3% increase in the gross domestic purchases deflator, then growth was -1.2%.

Second, income data suggest far weaker growth. ...

as with the much-maligned bureau of labor statistics' birth/death model or the u3-vs-u6 debate, there's no ad hoc political revisionism here. this is simply statistics at work, revealing some things and masking others. one can argue that some measures have been reconstituted with a political aim in mind -- the changed computation of CPI in an effort to be more parsimonious with social security payouts, for example -- but GDP methodology hasn't been altered here. no conspiracies, please.

but it is a statement that the economy is acting much worse than a 1.9% growth rate would imply, and one can easily see why by checking under the hood of the numbers.

it's also becoming clearer than june saw the effect of stimulus checks dry up and the economy take a significant turn for the worse. one might expect subsequent downward revisions for 1q and 2q (as they are more common in downturns), and a pretty ugly number for 3q come october if there is no further fiscal stimulus in 3q. the credit crunch is intensifying and firms are going bust at accelerating rates.

UPDATE: i commented at dash of insight as well.

i think it's important to realize that the GDP figure is not designed to capture dollar volume of transactions but physical volume of goods. as such, even with the application of the deflator it is not strictly 'inflation-adjusted'. jim hamilton explained here.

"The concept of 'real' GDP is intended to measure physical quantities rather than dollar amounts. So that if you paid a higher dollar price but purchased the same number of physical goods, real GDP would be unchanged," said professor James Hamilton of the University of California, San Diego.

so -- particularly when discussing the reality of prices paid and the squeeze being applied to consumers and producers alike as rising input costs and declining revenues destroy net income -- it's perhaps more realistic to take volume GDP and correct for inflation using CPI the gross domestic purchases deflator.

this is what stefan karlsson does here, also calculating real net national income.

these figures are much more consistent with how the economy appears to be behaving in terms of job losses (see here and here), with the corresponding decline in hours worked and therefore (in an environment of placid unit wage growth) income.

though i agree that the jump in weekly claims is probably being distorted to some extent by a transient effect, there's no mistaking the trend. moreover, continuing claims lag weekly claims by a week -- and they accelerated briskly upward as well in a move plainly not related to rule changes.

i don't know how great barry ritholtz's analysis is -- though i think he does have a valid point on the effect of birth/death slowing the sensitivity of NFP on trend changes even as it makes the model more accurate overall -- but i do have to agree with vermont above. all these indications are unambiguously recessionary. the question now isn't 'if' but 'how severe'.

UPDATE: more from menzie chinn, who shows that the GDP deflator is not adjusting for the change in terms of trade as foreign creditors find the united states a less-attractive credit. this is a far more cogent explanation than i could muster.

UPDATE: more still from jim hamilton.

Labels: ,

This page is powered by Blogger. Isn't yours?