Richmond Fed Services Revenues Index

The headline is the Revenues row of the Fifth District Survey of Service Sector Activity, now called the Fifth District Non-Manufacturing Survey. The Richmond Fed, one of 12 regional Federal Reserve Banks, surveys senior leaders about their own establishment.

Coverage spans DC, Maryland, North Carolina, South Carolina, Virginia and most of West Virginia, starting in November 1993. Monthly reports are available back to January 1994. It measures regional breadth (how many firms rise versus fall), not national output or dollar receipts, and it excludes manufacturing and government.

Responses are usually collected over about 3 weeks, from 2 business days after the prior release until the Wednesday before publication. Panel members receive a link before 10 a.m. Eastern on day one, with 3 reminder emails. Results usually appear on the 4th Tuesday at 10:00 to 10:10 a.m. Eastern. Publication moves to Wednesday when Monday is a holiday. Each release posts 1 PDF with narrative, charts and a table for current and 6-month-ahead expectations. Charts cover revenues with a 3-month average plus jobs and prices. A history spreadsheet adds seasonally adjusted and unadjusted series. Corrections appear in the history file, not in the PDF reports. Seasonal factors are recalculated each July with history revised, while readings above 0 signal expansion and below 0 signal contraction.

Revenues move with demand, local conditions, spending plans, jobs and prices across surveyed service firms.

• Demand: whether customer demand for its products or services rose, held steady or fell.
• Local Business Conditions: how the respondent rates local business conditions.
• Capital and services spending: whether investment, equipment and software outlays and services spending rose, held steady or fell.
• Jobs, wages and hiring ease: whether headcount rose, held steady or fell, with pay reflecting raise size more than breadth, plus hiring ease.
• Prices Paid and Prices Received: average % changes in input and selling prices over 12 months, trimmed 5%, not diffusion indexes.
• Retail breakouts: retail sales, big-ticket goods lasting 3 years or more, shopper traffic and inventories for retail firms only.

The April 28, 2020 release showed revenues falling from 1 in March to -87 in April. It was the lowest reading and largest one-month drop on record. Breadth drove the fall toward the -100 floor, while demand and local conditions hit record lows and capital spending fell sharply. Employment, wages and workweek hit survey lows, while skills availability rose above 0 for the first time since 2017 as layoffs made hiring easier. Input price growth rose while selling price growth fell, widening the gap between prices paid and prices received, which are unadjusted.

1. Classify responses: Each firm reports whether activity rose, showed no change or fell versus the prior period, and expectations cover 6 months ahead.
2. Diffusion index: Subtract decrease share from increase share, so Index equals 100 times (up minus down) divided by respondents, ranging from -100 to 100. For example, 70 up, 50 down and 40 unchanged out of 160 equals 43.75 minus 31.25, or 12.5, reported as about 13. With 30% up and 45% down, the index is -15.
3. Seasonal adjustment: Adjust each diffusion index for normal seasonal swings using an Arima-13 process and publish the adjusted value as the headline. Re-estimate factors each July and revise the full history, while unadjusted values remain in the spreadsheet.
4. Price trends: Report prices paid (input prices) and prices received (selling prices) as average % changes, trimmed 5% to limit outliers. Current gauges cover the past 12 months and expectations cover the next 12 months, without seasonal adjustment. They are not diffusion indexes.