Dallas Fed Services Index

The Dallas Fed Services Index is the vendor label for the headline series in the Texas Service Sector Outlook Survey (TSSOS). The Federal Reserve Bank of Dallas conducts a monthly survey of private services in Texas only, with about 230 firms taking part. The panel spans transport, warehousing, finance, professional services, health, hospitality, retail, and wholesale.

Government bodies are excluded, and respondents are senior executives. The sector is 73% of Texas private activity, with over 12 million jobs. It is not a national gauge.

It measures breadth, how many firms rose versus fell, not dollar amounts. Collection began in January 2007, with public release from June 2011. Revenue from own-firm sales is the key gauge of state conditions.

Questionnaires go out by email near mid-month. Answers are collected over 7 business days in the latter half of the reference month. Release follows the manufacturing survey by one day, so it usually appears on a Tuesday near month end at 9:30 CT (Central Time).

Each release publishes a narrative report, results tables with shares up, unchanged, and down, revenue charts, and anonymized respondent comments. Full history files go back to January 2007. A PDF version is also posted.

Results are never revised month to month. History is revised each January only to recalculate seasonal factors.

Six stand-alone gauges shape the reading. There is no composite headline, because each reply counts equally regardless of firm size.

• Revenue: own-firm sales excluding normal seasonal change, the key gauge of state conditions.
• General business activity: executives' view of broad conditions, the vendor headline series, not own revenue.
• Employment and hours: headcount, part-time staff, and average weekly hours.
• Company outlook and capital spending: change in own-company outlook plus investment activity.
• Prices and labor costs: input costs, selling prices and pass-through, wages and benefits.
• Outlook uncertainty: whether uncertainty about own-company outlook rose, held steady, or eased; added in January 2018.

The March 31, 2020 release showed breadth: how many firms rose versus fell. Revenue fell from 14.0 to -67.0, on 2.4% up minus 69.4% down. Data were collected March 17 to 25, 2020 from 248 service and 56 retail executives. General activity hit -78.8 and company outlook hit -75.3 with 0.0% improved, pinning diffusion near -100 without measuring dollars lost.

Jobs, hours, and prices fell together as employment dropped from 6.1 to -23.8, hours to -43.0, wages to -12.4, and selling prices to -44.5. Uncertainty hit 37.6 while expectations for six months ahead sank to record lows. The April 28, 2020 follow-up held revenue at -65.4 while employment fell further to -35.2.

1. Ask: each executive reports increase, no change, or decrease versus the prior month excluding normal seasonal change. General conditions use improved, same, or worsened. The same set repeats for six months ahead.
2. Count shares: tally the shares reporting increase (I), no change (N), and decrease (D). Each reply counts equally regardless of firm size.
3. Subtract: the unadjusted diffusion index is I minus D, from -100 to 100. For example, 60 of 200 up (30%) minus 40 down (20%) equals +10.
4. Adjust for seasons: apply the X12 procedure where seasonality is significant. It covers 39 of 46 indexes. Adjust I, N, and D separately, then recompute. If adjusted shares do not sum to 100, normalize them to 100. Non-seasonal series go out unadjusted.
5. Publish: publish the index with its I, N, and D shares. There is no monthly revision. History is recalculated each January with an extra year of data.