STIR: December Rate Cut Prospects Hit By Hawkish Powell

How would data collection be affected?  As the Department of Labor warns, "All active data collection activities for BLS surveys will cease....A reduction in quality of data collected might impact the quality of future estimates produced." Using 2013's October employment report as an example of how this might play out: 

• The report was scheduled for release on Friday Nov 1 but was pushed back a week to Nov 8 as the BLS needed more time to get the report together.
• For that report the household survey week was Oct 6-12, and the survey scheduled to start Oct 13 was postponed by a week because of the shutdown. The survey reference week was unchanged and the response rate was "within its normal range". For the establishment survey, the reference week is the pay period that includes the 12th of the month. As such this survey was similarly impacted as the household one was, but there's also the revisions to consider, per the BLS post-mortem of that episode: "Collection continues for about 2 months, and the estimates are revised twice after their initial release to reflect the additional data collected. From October 1 through October 16, 2013, BLS normally would have collected data to be used for the August final, September second preliminary, and October preliminary establishment survey estimates. October data collection began shortly after BLS reopened." That having been said, "The response rate for the establishment survey was above average."

How would the data itself be impacted? Again using the October 2013 payrolls report as an example: for the establishment report, the BLS notes there were "no discernible impacts of the partial federal government shutdown on the estimates of employment, hours, and earnings. Federal employees on furlough during the partial government shutdown were still considered employed in the payroll survey because they worked or received pay for the pay period that included the 12th of the month" - this is because workers who are paid for any part of a pay period are defined as employed.

• However, "some of the estimates from the household survey clearly reflect an impact of the partial government shutdown. This is due to differences in the concepts and definitions used in the two surveys. In the household survey, workers who indicate that they were not working during the entire reference week due to a shutdown-related furlough and expect to be recalled to their jobs should be classified as unemployed on temporary layoff, whether or not they are paid for the time they were off work. In October, many furloughed federal employees were so classified, contributing to a rise in the overall number of persons unemployed on temporary layoff."
• This meant 223k federal workers were on temporary layoff in the month on a non-seasonally-adjusted basis (around 200k larger than normal), with 307k classified as employed with a job but not at work (again, around 200k larger than usual). The BLS doesn't adjust the household survey findings, but estimated that "If the federal workers who were recorded as employed but absent from work had been classified as unemployed on temporary layoff, the overall unemployment rate would have been slightly, but not substantively, higher than reported" - an estimated 0.1pp higher.
• We'll look more closely at impacts on CPI if the shutdown persists into next week, but here is what the BLS said in the October CPI report (which was pushed back a week): "Data collection, data review and index computation commenced shortly after the end of the shutdown. In order to minimize the impact of the shutdown on the quality and timeliness of the index, resources normally devoted to maintenance and improvement work were redirected into data collection and index production. The sample of prices used to calculate the October index was about 75 percent of the amount usually used in the CPI." Given methodological issues in the current year with data collection, with impacts from funding cutbacks already pinching, this could make for a very idiosyncratic October report if it comes to that.
• In terms of GDP, there could be a hit based on the BEA's methodology of reducing its growth estimate based on the number of federal government worker hours lost (feeding through to the real government consumption expenditure category). We've seen estimates of an impact around 0.1 (Morgan Stanley)-0.2pp (Deutsche ) subtracted from quarterly annualized GDP for each week that the shutdown goes on, assuming 800k employees are furloughed. Coming into this week, the Atlanta Fed GDPNow estimate for the quarter is for close to 4%, so this wouldn't be a major drag on activity. However the longer it goes on for, the more impact it will have, especially to the degree that it spills into private sector activity.

Amid signs that the federal government will "shut down" starting Wednesday October 1, we answer some of the most frequently asked questions we've received from clients.

• What are the chances of a shutdown? How long would it last? Prediction markets (Predictwise, Kalshi) are signalling around 70-80% chance of a shutdown starting Wednesday. There's a higher than 50% market-implied chance that it lasts more than 5 days, and 40% it lasts more than 10 days. This would be the first full (and not "partial") shutdown since 2013 - that one lasted from Oct 1 to Oct 16 (regular operations resumed Oct 17).
• How will this impact data releases? With only essential government services being performed during a shutdown, this would effectively mean that any data from federal agencies including the Bureau of Labor Statistics, Bureau of Economic Analysis, and other agencies (commodities data via Dept of Agriculture and Dept of Energy) would be postponed until after the end of the shutdown. There would also be potential delays to subsequent releases as data collection is not able to be conducted.
• What are the data releases that are set to be postponed? The two key data releases that would be impacted by a shutdown through mid-month are Friday's Employment report for September, and the CPI report on Oct 15 (and PPI the next day). All the other not-quite 1st tier data, from construction spending to housing starts to trade balance, would also not be published (thankfully, the week of Oct 6 - 10 is unusually light). The Department of Labor's shutdown plans suggest there will be no weekly jobless claims reports published. These were published in 2013's shutdown, as while they were compiled and released by the BLS, they use state-by-state data. We go through the 2013 episode in more detail in the next note by way of illustrating what it could look like this time.
• What data will we get? Private-sector compiled data will still be released after September 30, and this will serve as an alternative of sorts to the official government data to gauge economic developments. Looking at the more immediate releases: for the labor market, this includes the ADP employment report and Challenger job cuts out this Wednesday and Thursday. For broader activity, we get ISM Manufacturing / Services (Weds and Fri). For retail sales we get indicators including the weekly Redbook release and Wards Automotive sales. Also Federal Reserve releases including the regional Fed surveys would be unaffected: in the 1st half of the month that includes NY Fed consumer survey and manufacturing surveys, as well as the Federal Reserve Beige Book.
• When would data releases resume? In 2013, releases were pushed back to slightly after the end of the shutdown. That meant that there were some odd days of the week: the September 2013 payrolls data for example came out on Tuesday Oct 22 (instead of Fri Oct 4). The Department of Labor's notice summarizes the impact of a federal shutdown: "BLS will suspend all operations. Economic data that are scheduled to be released during the lapse will not be released.... The BLS website will not be updated with new content or restored in the event of a technical failure during a lapse. The releases of economic data will likely be delayed if a lapse is prolonged." See the table the BLS provided in 2013 after the shutdown ended to get a sense of the impact of delays.
• Will Treasury auctions be affected? No, these are considered "essential" and will be conducted as normal. There is a possibility that bill auctions could be adjusted for cash management purposes the longer the shutdown goes on, but the near-term impact is minimal.

• Date $MM Issuer (Priced *, Launch #)
• 09/29 $1.4B #Codelco WNG $700M 2035 Tap +130, $700M 2055Tap +155
• 09/29 $1.25B #Dominion Energy $625M each 2056 Taps 6 & 6.2
• 09/29 $1B *Scotiabank 60NC10 6.875%
• 09/29 $500M #Fortitude Global Funding WNG 3Y +100
• 09/29 $500M *Vakifbank WNG PerpNC5.25 8.5%
• 09/29 $500M #Corebridge Funding 5Y +73
• 09/29 $500M Starwood Property Trust 5.25NC 5.75%
• 09/29 $1.25B #Societe Generale 11NC10 +130
• 09/29 $1.4B Waterbridge Starts $400M each 4Y, 8Y investor calls
• 09/29 $Benchmark Kuwait 3Y, 5Y, 10Y investor calls
• 09/29 $Benchmark Egypt 3Y, 7Y Sukuk investor calls
• Expected Tuesday:
• 09/30 $Benchmark Rentenbank 5Y SOFR+41a