“At this moment, the decision to begin the normalization process at the September meeting seems less compelling to me than it did several weeks ago.”
When I was a kid you didn’t take Tae Kwon Do.
If you actually got into a full fledge karate stance during an escalating confrontation, you were just going to get beat up worse … and more often.
My son takes jiu jitsu 2 -3 nights a week. At 5 years old, he’s the youngest kid in the class and he doesn’t really get it, but that’s okay.
Roll … get beat up … do it again, figure it out.
It’s all about the reps.
Back to the Global Macro Grind …
Last week, Dudley front ran the Jackson Hole festivities with the comment in the headline quote above.
What’s the takeaway?
It hints at his preferred policy preference but, more than that, I think it offers compelling confirmation of the reactive nature of conventional policy.
The reality of global growth slowing seemingly wasn’t apparent until the China data was overtly terrible, DM markets were treading correction territory, and EM and commodity markets were already in full crash mode. Shifting growth/inflation/market data needs to actually be reported for it to become reality.
Conventional monetary policy and its neoclassical underpinnings are kind of like tae kwon do – pretty and elegantly conceived, but you can’t really use it to manage real-time risk.
Meanwhile, back on the Macro mat, it’s Jobs Friday and there’s no shortage of high-frequency, fundamental reps to be taken on the labor side of the domestic economy, so let’s strap it on.
Macro Melodrama: The manic media and market melodrama that accompanies the monthly employment report is both amusing and annoying. The trend in employment is certainly important but the noise in the month-to-month figures is substantial. Recall, the standard error on the NFP estimate is +/-90K jobs. In other words, if the NFP print was 100K, the BLS is 90% sure we gained between 0 (as in “zero”) and 200K jobs. Given the monthly volatility and the margin of error, it’s all very silly when pundits convictedly opine on the underlying state of the labor market in response to a headline beat/miss of +/- 10K on a given month.
Tracking the Trend: Historically, Initial Jobless Claims have been the most consistent, lead labor market indicator for the economic cycle. While total employment peaks coincident to the end of the cycle (which, of course, is not surprising – how do you think economists retrospectively pick when the cycle peaked/recession started?), peak improvement in initial claims occurs ~7 months ahead of the economic cycle peak and coincident with or slightly ahead of the equity market peak.
The Transition: As we highlighted in reviewing this week’s claims data, the current labor market trend is showing a subtle shift from great to good. Initial claims put in their low watermark ~6 wks ago and are now in a 1.5 steps back, 1 step forward re-convergence higher, moving back toward 300k. While anything sub-300k isn't "bad" per se - remember, it's what happens on the margin that counts. This is especially true when the market is stretched on longer-term valuation metrics and is entering the upper echelons of historical bull market duration.
The same great-to-good trend dynamic has characterized the ADP and NFP data in recent months. From a rate-of-change perspective, February marked the peak in employment growth and while net gains of 211K on average YTD are “good” per se, on the margin, they are less good than the 260K average observed last year.
The (Cycle) Tea Leaves: As Josh Steiner, our Financials sector head surmised yesterday - It's not easy to tease out exact causation here, but a few possibilities come to mind. One possibility is the inevitability of late cycle reality --> great becomes good (early late cycle --> mid late cycle) and then good becomes less good (mid late cycle --> late late cycle) and, eventually, less good becomes bad (late late cycle --> recession).
Another possibility is that the confluence of fears from real energy sector headwinds, August market weakness and rising Global Macro uncertainty are conspiring to facilitate that first shift downward from great to good. It's also possible (likely, even) that the latter is simply this cycle's causal factor for the former.
In other words, markets are non-linear and critical state thresholds (i.e. when a particular stimulus becomes sufficiently strong enough to trigger a large-scale and cascading reaction) are often only clear after the fact.
Goods vs Services: Another relatively low-intensity means of tracking the evolution of the cycle is through the monitoring of spending on and employment in the goods producing and services producing sectors.
Due to consumption smoothing, household spending is typically less volatile than the other GDP expenditure buckets. Consumption of durable goods, however, is more sensitive to macro conditions than is consumption of services and can serve as a decent lead indicator for broader consumption/growth. In short, spending on higher-ticket durables is more cyclical and tends to peak ahead of the peak in demand for more inelastic consumption.
In the Chart of the Day below we show the trend in Goods vs Services employment over the last four cycles. As can be seen, employment growth in the goods producing sector tends to presage the trend in services and aggregate employment.
The bottom-line is that the clock tick is getting louder on the current expansion and it’s time to start preparing for the inevitable cyclical downturn.
Preparing, however, does not necessarily mean going full hazmat suit bearish at every time and price. While we’re late-cycle currently, historical cycle precedents suggest there’s still some modest runway left to the current expansion.
On the employment front specifically, while we’re past peak in employment growth and monthly gains have transitioned from great-to-good, the combination of flattish monthly gains, positive mix (i.e. more higher paying jobs) and a longer work week have, to-date, driven aggregate personal and salary/wage income higher. With credit growth still muted, improving wage income remains the sangre vital of our modern consumption economy.
Tops are processes.
If this morning’s labor data is trend consistent, the normal mania will only be amplified and speculative angst around September lift-off will see its final crescendo.
Slower-for-longer remains the call. More forest, less bark, more risk-managed, late-cycle allocation reps to execute.
Our immediate-term Global Macro Risk Ranges are now:
UST 10yr Yield 1.98-2.22% (bearish)
SPX 1 (bearish)
VIX 21.69-41.95 (bullish)
USD 93.65-97.19 (neutral)
Oil (WTI) 36.47-49.63
Gold 1113-1165 (bullish)
Good luck out there and Happy Labor Day,
Christian B. Drake
U.S. Macro Analyst