Turning Down the Volume on Some Noisy US Data

Several widely watched US economic data reports—including September retail sales and industrial production and a couple of October regional manufacturing surveys—have given experts plenty to discuss in recent days. Yes, consumers have been spending—but is it sustainable? Sure, manufacturing has been resilient—but are the latest readings a setback? As pundits and economists debate their meaning, we suggest investors think like markets instead. Stocks are highly likely looking well past these noisy readings and digesting the far future—the long end of the 3 – 30 month range we think they anticipate.

First, the numbers. September retail sales rose 1.9% m/m, the fifth straight positive month. However, September industrial production (IP) dipped -0.6% m/m—missing expectations for 0.6% growth and breaking a four-month positive streak—with the manufacturing subcomponent down -0.3% m/m. Elsewhere, two Federal Reserve regional manufacturing surveys’ divergent October readings muddled analysis further. The Philadelphia Fed’s Manufacturing Business Outlook Survey surged to 32.3 from September’s 15.0 while the New York Fed’s Empire State Index slowed to 10.5 from September’s 17.0 (readings above zero indicate monthly activity rose). The data spurred the usual interpretations: Ongoing retail sales growth continues to surprise, though many think weakness looms; others see IP’s September contraction signaling a slowing economic recovery—which recent surveys may or may not corroborate.

While it is grand to have a snapshot of the latest economic developments, we recommend not drawing sweeping conclusions. There is just too much noise, too many one-off variables and too many conflicting signals. Consider these seemingly contradictory tidbits from the retail sales and industrial production reports. Automobile and other motor vehicles sales were up 4.0% m/m, its second-straight positive month—yet production of automotive products fell -4.1% m/m, its second-straight negative month.^[i] Were factories pulling back as retailers slashed prices on the last model year to clear old inventory that piled up in the pandemic? Maybe. Does this really mean much on a forward-looking basis? Probably not.

The time of year can color the data, too. Industrial production’s utilities component fell -5.6% m/m, worst of the three main industry groups, due to lower-than-expected air conditioning demand in September.^[ii] That says something about the weather but not much about the economy. COVID-19—which has impacted numerous seasonal adjustment calculations this year—appears to have affected retail sales, too. The Commerce Department applies a seasonal factor to August retail sales due to expected skew from back-to-school spending. But the planned adjustment didn’t—and couldn’t have—anticipated many school districts opting for “virtual learning” to open the year, which pushed a lot of school-related spending later into autumn. The upshot: The latest seasonally adjusted retail sales numbers don’t account for this year’s atypical circumstances. 

We also caution investors against reading too much into narrow, localized datasets—like regional manufacturing surveys. Beyond typical monthly volatility, different industries may have an outsized presence in a given geography or locale. For example, the Empire State survey may say something about the apparel business since New York state accounts for 12.6% of the industry nationwide, second biggest in the country.^[iii] Yet that same survey sheds little light on motor vehicle manufacturing given New York’s 0.8% sliver of the total—surveys covering Michigan would be more telling.^[iv] Also consider scale. Combined, New York plus Pennsylvania, New Jersey and Delaware—the three states covered by the Philadelphia Fed’s survey—account for 9.4% of total US manufacturing.^[v] Treating this as indicative of national manufacturing seems like a bit of a stretch to us.

The beautiful thing about stocks is that they see through all this noise. In our view, these data are old news to forward-looking stocks, which typically focus on the next 3 – 30 months. In early bull markets, stocks tend to look toward the longer-end of that spectrum, fathoming a world one, two and even three years in the future—in this case, when businesses and households have found a way to deal with COVID, whether due to a vaccine or because society adjusted accordingly. Interim data wobbles have little to do with this and are normal even in non-pandemic times. This doesn’t necessarily mean smooth sailing ahead, as risks always exist. For example, governments suddenly returning to draconian lockdowns in response to COVID could panic markets anew. But that is a possibility, not a probability today—and successful investing is based on probabilities. Right now, the largely dour reaction to growthy data indicate pessimism is still rampant—another hallmark of young bull markets. In our view, staying disciplined and tuning out the noise is critical for long-term investors.