Labor Market Recovery Still Has Legs, But S&P Hindered By Declining Momentum

|
Includes: SPY
by: Williams Analytics LLC

Summary

The labor market recovery continuing w/ declining momentum.

Future S&P returns may be hindered.

Negative 10-year correlations expected to ease.

Today we're looking at the monthly Labor Market Report wherein US labor data is used to strategically forecast* S&P 500 E-mini and 10-Year US Treasury Note front month futures prices. We begin with Total US Employment. Here, we see that Employment has been steadily increasing after the financial crisis and is forecasted to continue to increase (in a general sense) over the next year

Of course, Employment's path ahead is not without seasonal whipsaws. Specifically, we are anticipating drop in (Non Seasonally-Adjusted) Employment data until early-2016, a resurgence beginning around February 2016, and a leveling off around mid-2016. After mid-2016, Employment is expected to meander lower and then experience a slight rise into the New Year (2017). Our November 2017 forecast for non-seasonally-adjusted Employment currently stands at 146,198 (thousand).

While, as noted above, Employment is expected to increase (in aggregate) across the first-half of this year, the momentum (i.e. year-over-year percentage change) of this increase is anticipated to decline to a post-recovery, year-over-year rate of about 1.45% by year's end. This number is almost an entire percentage point drop in momentum relative to the peak of the labor market recovery (circa early-2015; 2.4%). Our forecasting models have and are continuing to provide indications of a slowing in Employment growth over the coming year; the data had not disappointed.

Looking at another metric of US labor market health is the commonly-reported U3 Unemployment measure. U3 has clearly declined since its peak (circa early-2010) over the past few years indicating that the US Labor Market has substantially improved, relative to where it was shortly after the financial crisis (i.e. around 11% unemployment). Granted, U3 is still above zero but this is to be expected. That is, a non-zero U3 measure does not indicate dysfunction but rather reflects the US economy's need for structural unemployment. Based on a long-term trend, and keeping structural unemployment in mind, we are forecasted to be near the US Economy's long-run employment rate. This, in turn, is a good indication that labor market improvements have mostly run their course.

While somewhat subtle, next year's forecasted U3 values (red) are trending at a gentler slope than the prior (actual; teal) declines in U3. Our forecasting models are predicting non-seasonally-adjusted U3 to reach 4.55% by November 2017.

However, just like with Employment, the pace U3 improvements, as measured by the year-over-year percentage change measure, is expected to slow significantly. Whereas U3 was declining on an year-over-year basis by about 1.5% in most of 2014, U3 momentum is anticipated to decline to a year-over-year value of about - 0.2% by the end of this year. Thus, like Employment, U3 improvement is slowing (again) indicating clear expectations that the labor market's recovery continues to slow.

In terms of the financial markets, we now want to examine how labor market conditions have impacted and are anticipated to impact the S&P E-mini and 10-Year futures contracts. We'll do so using our "Correlation w/ Macro" Measure that looks at rolling correlations between measures of labor market health (e.g. Employment, Average Hours Worked, and so on) and a given price series.

This measure ranges from -1 to +1 where positive values indicate that increases (decreases) in labor market health are associated with positive (negative) increases in the price series at hand. Thus, positive correlation values imply that good labor market expectations "lead" to higher prices and that bad labor market expectations lead to lower prices. Alternatively, negative correlation values mean that an increase (decrease) in labor market health "leads" to lower (higher) prices. Thus, negative correlation values mean that the market interprets good labor market expectations and conditions as being bad for price.

Starting with the S&P 500 E-mini, as the labor market has consistently improved over the last few years (to no one's shock) the S&P 500 has been blasting higher and higher with recent breaks in the trend representing its only sustained pause.

For that reason, the S&P's Economic Correlation measure has been strongly and consistently positive since the beginning of 2014. While the S&P's relationship with other macro indicators is certainly different (especially the credit markets), positive labor market news has generally been interpreted at positive for the S&P, leading to positive price moves. The reverse also has also held true: what few truly negative (or disappointing) labor market announcements that have been made have generally been associated with negative S&P price moves.

While this relationship is, again, different for other types of macroeconomic news, the S&P has responded in a manner consistent with economic fundamentals meaning that, despite the likely distortions caused by various central banks' policies, the S&P is not completely detached from (some of) the underlying fundamentals of the US economy. Further, based on our forecasting models, we are currently anticipating that the S&P will continue to exhibit a strongly-positive relationship with measures of labor market health for the year ahead.

As a cautionary note, however, we must reflect that a positive relationship implies that increases in labor market health are matched with increases in the S&P (and vice versa). Also, as noted above, many measures of labor market health, Employment and U3 in particular, are exhibiting declining momentum as the US labor market recovery wanes. Thus, assuming the S&P is still strongly linked to the labor market, the labor market's slowing may have an attenuating effect on S&P returns in the year ahead.

Continuing with the 10-Year US treasury Note, we see that it has had a much different recent trajectory than the S&P 500. Specifically, while the S&P has continued ever higher, the 10-Year began its second sustained decline (in an overall multi-decade bull trend) around early-2012.

Also, quite different from the S&P, the 10-Year's Economic Correlation Measure has been consistently negative since the beginning of 2014 indicating that improvements in the US labor market have actually been negative for 10-Year prices. Looking forward to the year ahead, we are anticipating that the 10-Year's Economic Correlation measure will be negative with a modest reversion back to less-correlated territory. Traders watching and participating in the 10-Year should be aware that the 10-Year's future market conditions and reaction to news may be different (same direction but less strong) than in prior years.

Thank you for your time and interest; we hope you have a profitable trading day!

Technical Note: All forecasts reported in this post and in Williams Analytics' Indicator Reports are generated from Williams Analytics' proprietary statistical forecasting models. Macroeconomic variable forecasts are based on the most-recent historical data possible on a quarterly or monthly basis, depending on the specific series at hand. Economic forecasts use own- trend, seasonal, and mean-reverting components in dynamically-adjusting time-series econometric models to find "optimal" t+k step-ahead forecasts. Note that each macroeconomic variable's forecast is independently estimated based on its own, "best fitting" model. Strategic price forecasts for various assets, e.g. the S&P 500 E-mini and 10-Year Treasury Note front month futures contracts, use the most-recent market observations available to provide monthly, t+k step-ahead statistical forecasts. In addition to using own- trend, seasonal, and mean-reverting components, Williams Analytics also includes higher-moment (e.g. volatility, etc.), inter-asset index, broad-asset index, and future macroeconomic expectation information (see above) in calculating each asset's independently-estimated, dynamically-adjusting time-series econometric model and associated forecast(s). All other reported macroeconomic and asset price information (e.g. volatility, year-over-year percentage change, Marginal Economic Influence, etc.) is based on both historical data and Williams Analytics' statistical forecasts.

Supporting Documents

  1. TR03-StrategicMacro.pdf

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.