Manufacturing Production MoM

Manufacturing Production MoM is the % change in US factory output versus the prior month. It is the manufacturing leg of G.17, the Federal Reserve Board monthly release on Industrial Production and Capacity Utilization. It compiles about 296 industry series, not a single survey, and covers factories plus logging and publishing for history, and excludes mining and utilities. It measures real output with value-added weights that avoid double counting and has no fixed threshold.

The release covers the prior month and arrives mid-month, around the 15th, at 9:15 AM Eastern on pre-announced dates. Each release shows manufacturing plus total output, mining, utilities, capacity and utilization, market and industry detail, vehicle assemblies and history tables. The first estimate is preliminary and revised over the next 5 months, giving 6 estimates as coverage rises from 70% to 98% of value added.

A yearly benchmark revision updates benchmarks, late data, seasonal factors and methods. A rise means expansion and a fall means contraction.

Manufacturing moves with its 3 published parts, weighted by value-added shares that change each yearly revision.

• Durable manufacturing, 39.04%: long-life goods like vehicles, machinery, electronics and metals, the most cycle-sensitive leg.
• Nondurable manufacturing, 35.34%: everyday goods like food, chemicals, fuel, plastics, textiles and paper, the steadier leg.
• Other manufacturing, 1.51%: logging and publishing kept in manufacturing for history.
• Motor vehicles and parts, about 4%: volatile durables part where a 10% jump adds about 0.4 pp to total output.

In the May 15, 2020 release covering April 2020, total output fell 11.2%, its largest drop in 101 years. Manufacturing fell 13.7%, its largest fall on record. Motor vehicles and parts fell more than 70% as plants shut, while durables fell 19.3% and nondurables fell 8.2%, so vehicles drove the headline. Mining fell 6.1% but utilities fell only 0.9%, showing sibling legs can diverge because weights blend the total.

1. Annual benchmarks: annual levels set the year-to-year trend from Census, energy and mining sources, deflated by chain-type price indexes.
2. Monthly indicators: monthly moves use physical product counts where available and BLS factory hours times a productivity factor for 130 series where not. High-tech goods use detailed price and quantity indexes, with edits for strikes, storms and shutdowns.
3. Index relatives: each series is expressed against the base period, with 2017 average = 100. Physical data use daily averages and hours use monthly averages.
4. Seasonal adjustment: every series is adjusted with Census X-13ARIMA-SEATS, then aggregates sum the adjusted parts since 1972.
5. Fisher aggregation: monthly change is a Fisher index, meaning the geometric mean of output valued at month m and month m-1 value added. For example, links of 1.0060 and 1.00618 give 1.00609, or about +0.61%.
6. MoM %: manufacturing MoM % equals (index in month m / index in month m-1 − 1) × 100 from unrounded indexes. Any aggregate move also equals the weighted average of part moves using prior-month value-added shares.