“I am the object of criticism around the world. But I think that since I am being discussed, then I am on the right track.”
The “official” story is that North Korean Vice Premier Kim Yong Jin was executed for “not keeping his posture upright at a public event”.
Whatever the real reason is, it probably doesn’t matter – I’m sure it falls within the same broader bucket of absurdity.
As I was reading that Reuters report yesterday, I was reminded of the famous-in-econ-circles chart below.
It shows a night image of North & South Korea. The contrast in light output is meant to convey the relative income differences between the two countries.
Because they were once a single country, share a similar geography and have historically similar cultures and comparable resources, the two countries serve as a natural case study in growth policy.
Institutions and ideologies matter to real growth, real standards of living, and real lives, a lot.
Back to the Global Macro Grind ….
So, North vs South.
That’s pretty much what macro investing distills down to. Is the 2nd derivative of growth and inflation positive and headed north or negative and heading south.
Let’s review the recent fundamental data before drilling into some Jobs Day, labor market specifics:
- ISM: In confirmation of what may be the most passive aggressive top ever, the August ISM reading printed 49.4, down -3.2 pts sequentially and down for a 2nd month off the June “top” as current production, new orders, employment and backlogs all slide into contraction. Our view has been that while stabilization in manufacturing activity after a year+ of negative growth wouldn’t be particularly surprising, we didn’t have a fundamental catalyst for a sustainable inflection. In other words, while going from bad to less bad is good, it doesn’t really represent an investible Trend without a catalyst or expectation for a 48 --> 52 --> 58 --> 62 type progression.
- Auto Sales: Auto Sales fell -4.8% MoM in August, offering further confirmation that 2015 was the probable peak in demand. The companies themselves are publically acknowledging that 2015 was the top and we’re not going back to 18M in annualized sales. Take their word for it.
- Retail Sales: Auto sales represent ~20% of Headline Retail Sales so a -5% decline will sit as a -1% drag to reported growth when the data are reported on 9/15. Recall, Retail Sales were up just 0.0% last month despite auto sales being up +6.5% sequentially. The Retail Sales Control Group (GDP Input) and all the Ex-Auto subaggregates were negative month-over-month and decelerating on a year-over-year basis. Some industry other than “e-commerce” will have to step up in August.
If we’re scoring yesterday, that = 3 Souths, 0 Norths
Moving on to employment. Here’s a look at the North-to-South annual progression in the monthly NFP average:
- 2014: 250K/mo
- 2015: 228K/mo
- 2016: 186K/mo
From a rate of change perspective, payroll growth will continue to slow from here.
The why is relatively straightforward:
- Employment growth is hostage to the law of large numbers and as the base gets bigger, an accelerating growth rate implies an ever increasing number of jobs. On an NFP base of 145M, the numbers get unreasonable quickly (i.e. you start needing to add 700K … 900K … >1M workers net on monthly basis to maintain the growth curve).
- Diminished slack and a tighter labor market. As the expansion matures and labor supply tightens, there are simply less people to hire and more competition.
And as we reach full employment, jobs gains need only be large enough for the labor force to absorb net new entrants. In forecasting that figure and translating it into a monthly NFP estimate, there are four primary inputs:
- Population Growth
- Labor Force participation by age
- Natural Unemployment Rate
- A conversion factor
In other words, the estimate must account for the number of working age people (the net flow of new, young entrants and older retirees), the percentage of that working age population that will generally have or be seeking employment (the Labor Force Participation Rate) and the number of labor force participants that will be unemployed in some average sense over time (Unemployment Rate). Lastly, because the labor force participation and unemployment figures are estimated from the BLS Household Survey of employment and the NFP figures are derived from the Establishment Survey, the estimate requires an adjustment to account for systematic differences between the two surveys.
The output is moderately sensitive to the assumptions embedded in those inputs but, across a liberal range of assumptions, the estimates of trend employment generally fall between 60-120K per month (Janet thinks it’s <100K).
In other words, employment growth will continue migrating South for the Winter.
Which brings us to the simple macro reality we harp on every month:
If employment growth is slowing, there are only a couple ways in which (aggregate) income growth can remain flat or accelerate:
- An increase in the number of hours worked per week. It has been largely flat the last couple years. And/or …
- An acceleration in wage growth. Wage inflation has shown some fledgling mojo in recent months but wage growth needs to accelerate faster than employment growth slows, on an ongoing basis, to support any improvement in income growth.
And if income growth is slowing there are only a couple primary ways to maintain or accelerate consumption growth:
- Declining savings rate: A declining savings rate, while arguably a negative fundamental development, is supportive of consumption growth from an accounting perspective and the savings rate was down -20 bps year-over-year to 5.7% in July, providing a modest tailwind to spending growth. With the savings rate at or above 6.0% in 2H15, if we hold the current level it will be supportive of reported consumption growth over the coming months.
- Accelerate credit growth: With aggregate income growth slowing since its peak in early 2015, accelerating credit growth has helped buttress the deceleration in consumption growth. That support may be fading. Total Consumer Credit growth peaked in October of last year and has slowed in each of the last three months and revolving credit growth (i.e. credit cards) peaked in March and has also slowed in the subsequent 3 months.
I have more to say around the outlook for consumption but this morning’s verbosity is marching Northward and I want to leave you with a couple parting thoughts:
The trend in employment remains in somewhat of a Catch-22:
- If employment growth continues to slow and wage growth can’t manage a sustainable acceleration then income and consumption growth will continue to slow.
- If employment maintains its current pace and labor growth continues to run at a positive spread to output growth then Productivity, Corporate Margins and Profitability will all continue to head South.
Tactically, as Keith has highlighted, bad probably = good and good probably = bad in terms of this morning’s jobs number as the implication for equities and asset prices broadly follows some version of the following.
Strong employment report(s) = ↑ tightening expectations = Dollar ↑ = stuff priced in dollars ↓ = Deflation Risk ↑ = Yield Spread↓,Trade deficit ↑, etc
Does industrial/energy sector activity, investment or earnings inflect and inflate in that scenario? Did it in December’s iteration of that macro factor flow?
Our immediate-term Global Macro Risk Ranges are now:
UST 10yr Yield 1.50-1.62%
Oil (WTI) 42.13-45.55
Happy Labor Day.
Christian B. Drake
U.S. Macro Analyst