Capacity Utilization: Formula, Meaning, and How to Read the Rate

Capacity utilization is a ratio between observed production and an estimate of sustainable production capacity. In plain terms, it asks how much of the capacity represented by a particular statistical series is being used. The answer depends on the industry, country, measurement method, and release being discussed. A factory survey, a national statistical agency’s index, and the Federal Reserve Board’s industrial measure do not automatically mean the same thing.

The U.S. Federal Reserve publishes monthly capacity and utilization data with its Industrial Production and Capacity Utilization release, known as G.17. The series cover manufacturing, mining, and electric and gas utilities. The accompanying industrial production index measures real output, while related capacity indexes provide an estimated denominator for the utilization calculation. {source:fedG17About}

That scope matters when a headline says that “the economy” is operating at a certain percentage of capacity. The G.17 total-industry aggregate is broader than manufacturing alone, but it still describes covered industrial activity rather than every service, household, or public-sector activity. It is an indicator of one part of the economy, not a measure of all productive resources or economic well-being.

For a Federal Reserve G.17 series, capacity utilization equals the seasonally adjusted output index divided by the related capacity index, multiplied by 100 when expressed as a percentage. In compact form:

Capacity utilization = output index ÷ capacity index × 100

Suppose a made-up industrial series has a seasonally adjusted output index of 76 and a related capacity index of 80. Dividing 76 by 80 gives 0.95, or 95% utilization. These index values are hypothetical and are used only to show the arithmetic; they are not a current Federal Reserve reading. {source:fedG17About} {source:fedG17CapacityNotes}

Both values need to come from the same compatible series and period. Comparing one industry’s output index with another industry’s capacity index would not produce a meaningful utilization rate. Nor can a reader substitute nominal sales, a GDP growth rate, or a company’s reported factory percentage for the specific output and capacity indexes used in the G.17 calculation.

The ratio can move because measured output changes, because the estimated capacity index changes, or because both move. If output stays level while estimated capacity rises, the utilization rate can fall. If output drops while capacity is revised down by more, the ratio could rise. A percentage therefore summarizes a relationship between two series; by itself, it does not explain what caused either series to change.

The word capacity can sound like a fixed physical ceiling: the largest number of units a plant could produce if every machine ran continuously. The Federal Reserve uses a more practical concept. Its capacity indexes aim to represent the greatest output a plant can sustain on a realistic work schedule, allowing for normal downtime and assuming that inputs are available to operate the installed capital. {source:fedG17CapacityNotes} {source:fedG17CapacityMethod}

This sustainable level is not the same as an emergency sprint. A plant might temporarily exceed its normal schedule by adding shifts, postponing maintenance, or running equipment longer. That does not mean the higher pace can be maintained indefinitely. Likewise, a scheduled maintenance stop does not necessarily signal weak demand or a broken supply chain; ordinary operating schedules and downtime are part of the capacity concept.

Capacity is also an estimate, not a meter installed in every facility. The Board combines production data with capacity information from different sources. Depending on the industry, these may include physical-output reports, establishment surveys, and statistical estimates based on capital and industry trends. The resulting index is designed to be consistent with observed production, but it is not a census of all machines running at maximum speed. {source:fedG17CapacityNotes} {source:fedG17CapacityMethod}

The Federal Reserve constructs capacity and utilization measures for U.S. manufacturing, mining, and electric and gas utilities. It publishes rates for detailed industries and broader groupings, including manufacturing, mining, utilities, and total industry. The G.17 documentation describes 89 detailed industry series: 71 in manufacturing, 16 in mining, and 2 in utilities. Most detailed series correspond to three- or four-digit North American Industry Classification System (NAICS) industries. The Board also publishes broader groupings. Choose the series that matches the question: a total-industry rate does not describe one plant, and rates from different groupings are not interchangeable. {source:fedG17CapacityNotes}

Broader G.17 aggregates are not simple averages of their component utilization rates. Through the latest completed year, annual utilization aggregates are capacity-weighted; the Board derives annual aggregate capacity from the corresponding production and utilization aggregates, then interpolates monthly aggregate capacity using a Fisher index of the component monthly capacity series. As a result, averaging several industry percentages yourself may not reproduce the published total. {source:fedG17CapacityNotes}

The G.17 production and capacity indexes are each expressed relative to real output in 2017, but the utilization rate is the output index divided by its matching capacity index. An output index of 76 by itself therefore does not mean that 76% of capacity is in use. Check the denominator before reading an index level as a utilization percentage. {source:fedG17About}

The measure does not provide a direct capacity percentage for the full U.S. economy. Many service activities are outside the industrial capacity series, so the G.17 aggregate cannot stand in for activity across the full economy. Construction is also not included in the stated G.17 definition of the industrial sector, though the Federal Reserve notes that industry together with construction accounts for much of the variation in national output over the business cycle. {source:fedG17About}

Other statistical systems can use different approaches. OECD business-tendency guidance, for example, recommends asking firms for their current level of capacity utilization as a percentage in a quarterly survey. That kind of reported survey response is not automatically identical to the Federal Reserve’s output-index-to-capacity-index ratio. When comparing countries or sources, check who is surveyed, which industries are covered, how capacity is defined, and how the aggregate is calculated. {source:oecdBusinessTendencyCapacityUtilization}

There is no direct monthly observation of total industrial capacity. The Board estimates annual capacity indexes from available source data, then combines them with monthly industrial production information to construct the published series. Some industries have physical output or capacity data from government and trade sources; for many manufacturing industries, capacity estimates use responses to the Census Bureau’s Quarterly Survey of Plant Capacity. The precise inputs vary across industries. {source:fedG17CapacityNotes} {source:fedG17CapacityMethod}

Capacity estimates and production indexes are updated as new source data and benchmark information become available. The Federal Reserve revises the monthly industrial production series during the regular reporting window and incorporates broader benchmark information in annual revisions. Its 2026 performance evaluation reports historical revision measures for both production and capacity utilization. Those averages describe past revisions across particular samples; they do not guarantee that a specific future release will be revised by the same amount. {source:fedG17OmbReview2026}

This is why a chart or article should identify the release vintage when it discusses a particular value. A later annual revision can change the historical series, and current estimates may rely on less complete information than subsequent estimates. A revision does not make the initial release useless; it means the number is an estimate that can be improved as information arrives.

Start by naming the exact series. “Manufacturing capacity utilization,” “total industry utilization,” and a survey’s company-level utilization response can refer to different coverage and construction. Compare the same series over time, and note its frequency, seasonal adjustment, reference period, and release vintage. Do not assume that 90% in one industry means the same degree of pressure as 90% in another.

The Federal Reserve notes that industrial plants commonly operate below 100% of estimated sustainable capacity, and that the highs and lows occur at different times across series. A rate below 100% is therefore not automatically evidence that output could rise without cost or delay. Capacity indexes already account for normal downtime and realistic schedules, while actual expansion can still require labor, materials, investment, maintenance, and time. {source:fedG17CapacityNotes}

There is no universal percentage that means “overheating” or guarantees that a bottleneck exists. A rising rate can reflect stronger output, a change in estimated capacity, or both. An industry-level rate also does not show whether idle capacity is located in the same firms or regions where demand is strongest. Use the measure as one piece of industrial context, then look at production, orders, deliveries, prices, investment, and other relevant evidence before drawing a causal conclusion.

Capacity utilization focuses on output relative to estimated sustainable capacity in the industries covered by a particular series. The macroeconomic output gap instead compares broad real GDP with an estimate of potential GDP. The Congressional Budget Office describes potential output as the economy’s fundamental ability to supply goods and services and documents its U.S. estimation framework. Both concepts involve an estimated capacity benchmark, but their coverage, units, models, and interpretation differ. {source:cboPotentialOutputMethodology}

A change in industrial utilization can provide information about manufacturing, mining, and utilities without showing that the whole economy has moved by the same amount. Service output, household activity, and other sectors can follow different paths. Conversely, an estimated economy-wide output gap does not tell a reader what share of the G.17 industrial capacity index is currently being used. See What Is the Output Gap? Potential GDP Explained for the broader GDP concept. {source:fedG17About}

Neither measure is a recession rule. A recession chronology combines its own definition, data coverage, and dating procedure. For a U.S. example, see What Is a Recession? The Two-Quarter Rule and NBER Definition; that article distinguishes the NBER’s retrospective monthly dates from the two-quarter GDP shorthand. A utilization rate can describe industrial conditions without establishing whether the economy meets an institutional recession definition.

A high utilization rate may be consistent with limited spare capacity in some covered industries, but it does not prove that a particular price increase was caused by demand. The G.17 rate measures output relative to estimated capacity, not prices or the causes of price changes. A claim about how capacity conditions relate to prices therefore requires separate evidence over time.

The denominator is estimated, the industrial aggregate excludes many activities, and the rate can change when either output or capacity changes. These features make it unsuitable as a stand-alone inflation forecast or as a universal policy trigger. A rate near a historical high for one series can be informative context, but the comparison does not show that every firm is constrained or that a broader price index must accelerate.

A careful explanation states the series, country, period, and data vintage; separates the observed rate from an interpretation; and checks complementary evidence. If a headline uses a single percentage to diagnose an entire national economy, ask whether the measure really covers that economy and whether the article has explained how its capacity estimate was built. For a related guide to inflation mechanisms, see Cost-Push vs. Demand-Pull Inflation: Causes and Key Differences. The G.17 rate can inform a discussion of industrial pressure, but it cannot settle that broader causal question by itself.