US: The Best Way to Shrink

If you've got to have a ghastly downward revision that puts US 1Q GDP shrinking at an annualized 1%, this is probably the best sort you have any right to expect. Three things stand out: I) the central role of inventories; ii) the upward revisions for the core private spending categories; and iii) this sort of slowdown does little damage to the grounds for expecting a cyclical acceleration in investment spending. In addition,  monthly data suggests this capital goods cycle is already quietly underway.

Inventories. The bulk of the downward revision, enough to account for the entire contraction, comes from a single source –  inventory adjustments. The slower pace of private inventory build-ups stripped 1.62 percentage points from GDP growth, compared with 57bps in the advance estimate. By itself, this change is responsible for the entire GDP contraction.  Now the thing about inventory movements is that they regularly do have a significant impact on quarterly GDP results, but that their volatilities tend to be answered rather quickly by equal and opposite volatilities.   Since 2000, the average contribution of private inventory movements to quarterly annualized GDP growth has been a fall of just 1bp, but the standard deviation of such movements is . . . 1.48 percentage points.   We can and should expect private inventory movements to rebound, and quickly, at least in terms of its contribution to GDP growth.

It should also be said that the inventory fall recorded in the GDP data is surprisingly large: additions to wholesalers' inventories slowed 12.7% qoq ,  whilst additions to manufacturers' and trade inventories slowed 19.5%, whilst the GDP estimates record a 52% qoq slowdown in nominal terms, and 56% in real terms.

Revised Up: Investment & Consumption Spending. Meanwhile, the core private sector spending categories were revised up, narrowly. Non-residential investment spending was revised up to an annualized fall of 1.6% from an initial estimate of 2.1%, and residential investment was revised up to minus 5.1% from an initial minus 5.7%.  To put these falls into context, they stripped only 36bps from growth.  Private consumption spending was revised up to 3.1% from the initial 3%, and added 2.09% to growth.  

The third thing to emphasise is that this blip in 1Q's GDP has had virtually no implications for the US cycle: even factoring in the 1Q decline, my ROC directional indicator (which expresses nominal GDP as an income stream from a stock of fixed capital, which in turn is estimated by depreciating all fixed capital spending over a 10yr period), continues to be at the high end of the range of the last 30 years.  Even after factoring in the declines of 1Q, private gross fixed capital formation is still running at 5.5% a year in nominal terms, and 3.2% in real terms. There is no reason not to expect this to re-accelerate throughout the year.

The Capital Goods Cycle is Turning. In fact, April's monthly data for orders and shipments of capital goods (nondef ex-air), provided some evidence  that this acceleration is already underway. This was not immediately obvious, since orders fell 1.2% mom – which was worse than expected. But that fall disguised an underlying  recovery which looks distinctly like a turning point. In fact, both orders and shipments of capital goods nondef ex-air are currently rising more rapidly than they have since around 2011, once one strips out the early 2013 rebound from the orders slump in 2H12.  That's partly because of the size of the revisions: March's total for orders, for example, was revised up to +4.7% mom from a preliminary 2.2%; the total for shipments was upped to 2.1% from the original 1%. Previous months were also upgrade, albeit less dramatically. The upshot is that when you look at the nominal dollar numbers, the breakout from previous levels for both orders and shipments is quite unmistakeable. 

And it is backed up by a further indicator: constructing a book-to-bill ratio for these capital goods, you find a sharp reversal from the rather threatening decline seen between Sept and Feb. Right now, the 1.03x ratio is 0.6SDs above the long term average and has recovered back to levels commonly held pre-crisis. 

Regardless of this, it will be hard for the the annual GDP growth tally for 2014 to recover to the 2.7%-2.8% which seemed likely at the beginning of the year. Nevertheless, whether you forecast GDP by looking at the various expenditure categories, or whether you take a production function input-based approach, the swing factor in this year's US GDP (ie, the difference between 2.7-2.8% and c3%+) was always going to be a long-predicted and long-overdue capital goods cycle. And it seems likely that, once this weather-afflicted quarter is stripped out, a modest acceleration can still be expected.