David Rosenberg is chief economist with Gluskin Sheff + Associates Inc. and author of the daily economic newsletter Breakfast with DaveDeborah Baic/The Globe and Mail
Well, the first U.S. macro data point of the year was hardly stellar.
The cyclically sensitive manufacturing index from the Institute of Supply Management slipped to 48.2 in December from 48.6 in November – riding a six-month losing streak and retreating to its lowest level since July, 2009. The consensus was 49.0 so a fairly big miss.
The pricing index fell from 39.0 in October to 35.5 in November to 33.5 in December, which is the lowest reading since April, 2009.
The really big drag was in the employment subindex, which sagged to 48.1 from 51.3 – it was 47.6 in October when factory payrolls fell 1,000 but that month we had offsets from other sectors because total non-farm payrolls managed to come in at plus 211,000.
Keep in mind that manufacturing commands just a 10-per-cent share of the economy; much of the remaining 90 per cent that is not linked to energy or sensitive to the supercharged U.S. dollar is doing just fine, thank you very much.
If there were bright spots, new orders improved slightly to 49.2 from 48.9 in November, vendor performance stayed above 50 for the fifth month running (50.3 from 50.6), the production index rebounded to 49.8 from 49.2, the inventory data were constructive with the index up to 43.5 from 43.0 and the gauge of customer inventories rose to 51.5 from 50.5.
So inventories are not a drag, and it looks like net exports are no longer a drag either – ISM export orders improved to an eight-month high of 51 from 47.5 in both October and November, while imports slumped to 45.5 from 49.0, the lowest since May, 2009. The last time we had a 3.5-point up-move in exports and a 3.5-point down-move in imports was back in October, 2010, and that month we had a massive 6-per-cent collapse in the U.S. trade deficit (and that quarter we had real gross domestic product growth of 2.5per cent at an annual rate … hardly a disaster).
There will be no shortage of "oh me, oh my" recession calls out of this back-to-back sub-50 ISM manufacturing tallies, but outright economic downturns do not begin until the ISM breaks below 42.
And we have seen sub-50 on the manufacturing diffusion before, even in this cycle, and this did not come close to upsetting the apple cart.
Just to put the current sub-50 reading into context – we have thrown everything but the kitchen sink at this (including a rising U.S. dollar, a Greek debt crisis, emerging market strains, a slowing Chinese economy, to name a few), and yet this diffusion index has only weakened; it has far from collapsed.
So am I worried about a sub-50 reading on the ISM manufacturing index? Hardly.
Just look at the Asian meltdown and the super-sized dollar in 1998 – the ISM manufacturing was sub-50 each and every month in the second half of that year, and at the low point it hit 46.8.
No recession and no bear market.
Then go back to the Tequila Crisis in 1995 and along with the lagged impact of the strong U.S. dollar, the ISM manufacturing was sub-50 from August of that year right through to May, 1996 – at one point, touching 45.5.
Guess what? Again, no recession, no bear market.
Don't forget about the 1985-86 oil price implosion, in the context of a non-recessionary economy and secular bull equity market, where we had ISM manufacturing drop to a low of 47.1.
In none of these periods did it pay to run to the sidelines – just avoid the affected sectors (for example, large-capitalization export-oriented industrial equities).
Note too that the Institute for Supply Management states in its manufacturing report that "a [purchasing managers' index] in excess of 43.1, over a period of time, generally indicates an expansion of the overall economy."
So a sub-50 reading itself doesn't even suggest recession; we'll start waving a red flag if we see a persistent move lower, below that 43.1 level that has historically coincided with a contraction in the overall U.S. economy.
Let's look at the bigger picture: Since the ISM manufacturing index was initiated in 1948, there have now been 131 sub-50 readings in non-recessionary months – this means that the ISM manufacturing index has registered a sub-50 reading in nearly 20 per cent of all non-recessionary months over the past 67 years.
Over the 35 years from 1948 to 1982, the U.S. economy experienced 10 periods in which there were prolonged contractions (at least six months of sub-50 readings on the ISM manufacturing index) in factory activity and only two of these did not occur in recessions – June, 1951, to July, 1952, and January, 1967, to July, 1967.
These two periods were preceded by periods in which there was a sharp jump in production (for example, industrial production growth was running at a plus-20-per-cent year-over-year pace for the 12 months ahead of the contraction in the early 1950s and a plus-9-per-cent year-over-year rate in the lead-up to that 1967 slump) that resulted in a significant build in inventories that had to subsequently be pared but the rest of the economy largely remained strong.
Now in the three decades since 1983, we have seen seven periods in which there was a prolonged contraction in manufacturing activity but only three of these occurred during recessions.
Specifically, in the four periods in which manufacturing went into what could be construed as a recession but the American economy did not, real GDP growth averaged just under 4 per cent at an annual rate supported by real consumer spending of just under 5 per cent – we are increasingly seeing periods in which we have manufacturing activity slump but the broader economy not fall into a recession, which makes sense given that manufacturing plays a considerably less important role in the U.S. economy now than it did 60 years ago, accounting for just 12 per cent of total economic output now versus almost 30 per cent in the 1950s.
Notably, in two of these periods in which actual industrial production was effectively flat on a year-over-year basis (the February, 1985, to September, 1985, period and from October, 2002, to June, 2003), a key factor affecting the manufacturing sector was weakness in exports – in that 1985 period, real exports declined at a 1.9-per-cent annual rate while real exports slumped at a 2.7-per-cent annual rate in the 2002-03 period.
This is largely reminiscent of the dynamics that we are seeing now.
The U.S. manufacturing sector is coming under pressure from weak foreign demand that is being compounded by the strong U.S. dollar – the strong currency has the impact of reducing foreign demand for U.S.-produced goods while also reducing domestic demand since foreign produced capital goods are now comparatively cheaper.
So it should not be that surprising that we got a sub-50 reading for the ISM manufacturing index and this does not mean that we should worry about an impending recession, especially if real consumer spending growth continues to run around a plus 3 per cent year over year – which was been the case for five quarters in a row.
David Rosenberg is chief economist with Gluskin Sheff + Associates Inc. and author of the daily economic newsletter Breakfast with Dave.