Conclusion: Household consumption growth is carrying the U.S. economy – for now. We do not think that is sustainable and reiterate our call for the YoY rate of change in domestic economic growth to slow throughout the balance of the year.
With the advent of this morning’s personal income and spending data, the U.S. economy is suddenly looking much better than it had been trending – particularly in YoY rate-of-change terms.
Specifically, real personal consumption expenditures growth accelerated in May on a MoM, YoY and 2Y Average basis as the savings rate ticked down -30bps. These are good numbers and we’d be both remiss and intellectually dishonest to suggest otherwise.
With respect to how our predictive tracking algorithm contextualizes the data, it is now accelerating on a sequential, trending and quarterly average basis:
The inflection in real PCE, which includes spending on goods and services and accounts for 69% of GDP, confirms the positive inflection in retail sales growth that we saw two weeks ago – specifically sales growth within the omnipotent control group that accounts for 24% of GDP.
One of the reasons we contextualize data in trending YoY rate-of-change terms is to have a consistent framework for incorporating new data in our Bayesian inference [modeling] process. It’s either accelerating or decelerating on a given duration of relevance – either monthly or quarterly or some meaningful combination of the two (e.g. a three-month moving average). The charts above and table below contextualize the output of this framework in the most succinct manner we can illustrate.
What you should glean from this table is the trending divergence between the [accelerating] consumption side of the economy and the [decelerating] investment, manufacturing and export side of the economy with the former essentially playing the role of Atlas – for now.
With base effects steepening sharply in 2H15 and national retail gasoline prices up nearly +40% off their YTD lows, we question the suitability of any consumer-driven recovery in the absence of a material acceleration in wage growth.
But as the following chart shows, you DO NOT want to see a material acceleration in wage growth if you’re in the camp that a recession would be bad for the capital market cycle. Wage growth peaks EXTREMELY late cycle, so the next acceleration (if any) may essentially sound the alarm on the next bear market.
Moreover, trends in the services and composite PMI data (also released today) suggest domestic consumption growth took a material turn for the worse here in June – just as the aforementioned steepening of base effects would’ve predicted.
Broadening this discussion back to our process, the primary reason we focus on NSA YoY rate of change rather than the QoQ SAAR rate of change – besides the fact that the latter is extremely volatile and has become increasingly nonsensical – is because the YoY rate of change most closely resembles how most investors analyze companies and helps us translate what we are seeing with respect to top-down trends to what bottom-up analysts are seeing in corporate operating metrics.
Additionally, it affords us true apples-to-apples growth rate comparisons across cycles, whereas the QoQ SAAR figures are subject to ever-changing seasonal adjustment methods that distort the reliability of the data over time.
Narrowing this discussion back to this economic cycle, we expect headline (i.e. QoQ SAAR) growth to accelerate in 2Q15 to ~2.5% before bouncing around ~2% in 2H15. That sounds good when coming off a -0.2% contraction in 1Q15, but that nets out to a full-year growth rate of +2.2% YoY and nominal GDP growth of ~3% - which would be the slowest pace since 2009.
Would you hike interest rates on that? If so, please email us why.
All told, the level of mischaracterization of domestic economic growth among market participants is fast approaching all-time highs, probably second only to mid-2009 when we were ragingly bullish amid a myriad of consensus "Great Depression" storytelling. But when you apply a repeatable, quantitative analytical framework to the discussion, navigating these macro risks becomes a little bit easier.
It's still not easy though!