US Industrial Stats Show Worst Monthly Drop in 101 Years
James Anderton posted on May 19, 2020 |
COVID-19 hammers capacity utilization, but the consumer still drives the economy.

The US Federal Reserve has released industrial production and capacity utilization statistics for April, illustrating just how damaging the continued run of the COVID-19 pandemic has been for US manufacturing. Overall, industrial production dropped by 11.2% in April, month over month, it is the biggest monthly drop in the entire hundred-and-one-year history of US government tracking of manufacturing activity. Manufacturing output specifically declined 13.7%, also an all-time record, with all major industries showing decreased output. Motor vehicles and parts declined by more than 70%, while other manufacturing activity dropped by 10.3%. Utilities and mining fared better, and mining went down 6.1% while utilities decreased by only 0.9%. The steady demand for energy and water suggest that industrial activity is in a holding pattern, without a statistically significant number of plant closures on the books, at least not yet. The Fed uses an index which sets 2012 output as the baseline, and total industrial production at 92.6% of the 2012 average represents a 15% decline over April 2019. Capacity utilization in industrials overall is a similar story, down 8.3% to a 64.9% rate in April. Factory utilization is a significant metric, and the April 2020 rate is 15% below the 35-year average and almost 2% below the all-time low set in 2009, in the wake of the financial crisis. The Fed’s March statistics were also revised downward after adjusting for higher-than-expected initial claims for unemployment insurance.

This Fed graph shows the extent of the declines:

It is important to note that capacity is relatively flat, as the immediate effect of the lockdown is to idle existing plants and defer or cancel plant expansions. This is expected in the immediate aftermath of a sudden shock to demand. Manufacturing companies’ idle production lines, with an eye toward a restart in the short term. Highly leveraged companies, however, may be compelled to liquidate assets in the form of plants and capital equipment if the COVID-19 lockdown is prolonged into two or three quarters. If this happens, the resulting decline in manufacturing capacity in the US economy may create a negative feedback loop in the economic recovery post COVID. In that case, capacity constraints will limit supply and finished goods output, resulting in either price inflation or an increase in importation of manufactured goods, which translates to Asia, particularly China. With the Trump Administration’s stated goal of reducing the trade imbalance with China, this would mean either significantly increased exports of goods and services to China (particularly agricultural products), or reduced economic growth in the US economy as manufacturing regains momentum.

This chart shows utilization as percent of capacity. The grey bands represent US recessions since 1970. The correlation is obvious, but one key takeaway is the slope of the curves during recessionary periods. Recessions over the last half-century show periods of rapid decline in industrial capacity utilization and output, and while COVID-19 is an externality and not organic loss of consumer confidence, the rate of decline in manufacturing activity suggests that we are in recession, without the formal declaration.

How do we restart a stalled manufacturing sector?

With the pandemic not yet under control, the manufacturing slump has moved beyond a short-term pause, and into a true force majeure event. But not all industries are equally affected. Take a look at these Fed charts of industrial production and capacity utilization:

All four charts show the same off-the-cliff declines in the month of April, and all trend-lines are down. However, note the difference between non-durable and durable goods, top left. The dramatic plunge in production of consumer durable goods are classic consumer behavior in a recession: fewer appliances, automobiles and home-improvement products as Americans are furloughed, lose their jobs, and save for an uncertain future. The chart top right she was another interesting phenomenon. While overall production capacity has dropped like a rock, the space and defense sectors show minimal impact from COVID-19, similar to their performance in the two previous recessions. Reasons for this are several: defense and space are mostly government-funded initiatives and are normally multiyear projects with demand and timelines that are not dependent on consumer sentiment. Major projects are also difficult to stop, with “kill” fees and termination clauses that incentivize governments to complete projects rather than cancel them.

This new 777X is highly fuel efficient and should be a highly profitable product for Boeing, if COVID-19 doesn’t kill the airlines that buy it. (Image courtesy of Boeing.)
This new 777X is highly fuel efficient and should be a highly profitable product for Boeing, if COVID-19 doesn’t kill the airlines that buy it. (Image courtesy of Boeing.)

 In American industry, Boeing carries unusual weight, and the company’s fortunes were already cloudy due to the 737 MAX issue. With two thirds of the world’s airliners grounded and the future of global air travel uncertain, only airfreight is robust. Unfortunately for major air-framers, this downturn is hastening the retirement of relatively fuel-efficient large twins, forming a low-cost supply base for the freighter conversion industry. This puts downward pressure on new air freighter pricing at a time when it’s most needed to keep lines operating. The collapse of global oil prices and the resulting significant reduction in fuel costs also work against new build air freighters as fuel efficiency has been a major driver of new aircraft sales.

The auto industry operates by a different set of constraints. This Fed graph tells the story:

While both the aviation and automotive sectors are ultimately consumer demand driven, autos are a consumer durable, and this chart shows the impact of COVID-19. Drop in output is unprecedented in half a century, although the rate of decline, the negative slope portions of the chart, is consistent with the 2008 recession. In all the recessions charted, upslope is similarly steep. As auto production increases in automation and supply chains become more sophisticated, it becomes easier for automakers to ramp up production when consumer confidence returns in the waning months of recession. While the trends suggest that declines are always followed by rapid improvement, the pandemic is different for several reasons. One is that the downturn has not been caused by natural economic forces such as consumer exhaustion at the end of a long period of growth, or higher interest rates, or increased energy prices. In a sense this is a “synthetic” recession, which may simply defer consumer demand, fueling a major production ramp up as lockdown restrictions ease. Tailwinds for the auto industry include an aging vehicle fleet, and an overall reduction in US consumer debt which is particularly important considering the amount of subprime auto loan debt, as well as the considerable number of consumers who are “upside down” on existing auto loan debt, meaning consumers who owe more for their vehicles than their market value. Auto manufacturers and dealers have been fighting this problem for several years and have taken advantage of very low interest rates to roll over existing consumer debt into new vehicle purchases. If a long-term effect of the pandemic is an increase in remote work and a decrease in overall miles travelled however, both the need for new vehicles and the rate at which the existing fleet wears out will be reduced. On the other hand, if social distancing rules go on for a considerable time after the lockdown is eased, which appears likely at this point, use of public transit may be discouraged. This may increase demand for private vehicle use as well as use of ridesharing services.

Is there a way out?

In a word, “yes”. Even without policy action by federal state and local governments, the epidemiology of all virus outbreaks tell us that it will end. Sweden is conducting a nationwide experiment now with minimal government intervention and no declared lockdown. It’s too early to determine if this course of action, which appears to depend on the rapid development of herd immunity, is the best way forward. US policy is not unified, with individual states taking different approaches to quarantine. At this point, a growing public backlash against an open-ended quarantine is developing at the state level, and there is considerable evidence that compliance is breaking down in many jurisdictions. The effects on the spread of the virus is unknown. From an industrial production perspective however, there are several factors that make pandemic fundamentally different from conventional economic recessions in the past:

  • The virus disproportionately affects the older demographics. This demographic tends to consume less, although they hold a significant portion of national wealth.
  • The sudden crash in consumer demand has been created by the closure of businesses, rather than an organic loss of consumer confidence.
  • Interest rates are still at multi-generational lows, which will help with demand for consumer durable goods.
  • The switch to online retailing, even for large consumer durable goods, may enhance consumer demand medium to long term.
  • Social distancing may accelerate the adoption of process automation in the manufacturing industries, improving productivity at a time when productivity growth has lagged in the US.

US manufacturing has been heavily dependent on the consumer since the Industrial Revolution and there is little evidence to suggest that even a generational event like a pandemic has changed consumer behavior or attitudes. The unanswered question is, how long can manufacturing remain closed without fundamentally disrupting supply chains and diluting the skill set of manufacturers’ employee pool? Statistically, it is likely that May and perhaps June numbers will worsen, but there has been remarkably little social unrest generated by this 100-year event. If the hoped-for V-shaped recovery emerges, US manufacturing should recover as quickly as in 2009.

Recommended For You