More on capacity utilization – a look at the historical data

Following yesterday’s post on capacity utilization, I thought it would be interesting to look at the data trend over longer time horizons.  The take-away message is that this information series correlates well with periods of economic contraction.

I looked first at the data going back to the mid 1990s (somewhat arbitrarily starting at Jan 1996):

Capacity utilization dipped by about three percentage points in early 1998, possibly corresponding to the disruption caused by the fall of Long Term Capital Management and the Russian Financial crisis.  Of note, this dip in capacity utilization was not associated with a recession.  I might take the counter-argument that the economic strength seen in the period from 1998 to 2000 was artificial, as financial wealth created by the tech bubble was not echoed by real growth in production.  Perhaps that is why capacity utilization stabilized following the dip in 1998 instead of rebounding.

Capacity utilization declined about eight percentage points in 2000 corresponding to the recession that followed the bursting of the tech bubble.

Yet another, longer time series of the data can be seen here (this time going back to the earliest data available on Bloomberg, starting Jan 1967).  US economic recessions are shown immediately below:

FRED Graph
The timelines from the two graphs above are not aligned. Please look at the dates when comparing data from these two graphs.


I observe the following:

  • Dips in capacity utilization track recessions well (perfectly over this period).
  • The magnitude of capacity utilization declines seen in recessions is about seven to ten percentage points.
  • Over the long term, capacity utilization has gradually trended lower (from high 80’s percent through the mid 1990s to high 70’s percent in recent years).  [This could be the result of increasing efficiency over time, perhaps largely stemming from better IT systems.  This is just a guess and perhaps merits further analysis at some later point.  For now, I identify this merely as an interesting observation.]

It is critically important to remember that most economic activity indicators show correlation in periods of economic contraction and expansion.  I try to look for information that is somehow predictive.  As I wrote in yesterday’s post, capacity utilization data tells me a ‘story’ that has this potential to be predictive.  Specifically (reiterating what I wrote yesterday):

What happens as capacity utilization is increasing?  Demand, relative to supply increases.  As more capacity is utilized, businesses have the ability to begin to raise prices.  Eventually this leads to hiring and expansion in order to build more capacity to meet demand.  This produces a virtuous cycle.  If capacity utilization falls, the inverse is true.  Further, as there is surplus capacity, producers/manufacturers might cut price to ensure that their marginal capacity is utilized.  This is productive if they secure contracts but counterproductive if it results in price wars.  Those manufacturers that don’t match lower prices might see volumes decline.  In almost any case, excess capacity tends to negatively impact profit margins.

Another notable point, especially as it pertains to the recent dip in capacity utilization, is that small declines in (one to a few percentage points) are not unprecedented and don’t necessarily imply an impending recession.  We saw small dips (without recessions) in 1978, 1995, and 1998.  However, such dips are still infrequent.  With this perspective, I stand by my prior conclusion: The data are concerning but not definitively conclusive.  It is appropriate to maintain watchful vigilance and caution but to also be mindful of upside surprises that could occur.

sources: Bloomberg, Federal Reserve Bank of St Louis Economic Database (FRED)

Print Friendly

Leave a Reply

Your email address will not be published. Required fields are marked *