Hurricanes Harvey and Irma and September 2017 payroll data

Background:

Hurricane Harvey first made landfall on Friday, August 25, near Rockport, Texas, 30 miles east-northeast of Corpus Christi, Texas. Rated as a Category 4 hurricane, Harvey stalled in southeast Texas, causing catastrophic damage from strong winds, storm surge and heavy rainfall, before returning to the Gulf of Mexico. Downgraded to a tropical storm, Harvey made landfall a second time on Wednesday, August 30, near Cameron, Louisiana. The Federal Emergency Management Agency (FEMA) declared 39 counties in Texas eligible for federal disaster assistance to individuals and households.

Hurricane Irma hit the lower Florida Keys on Sunday, September 10. After moving north, Irma made landfall again later that evening, on Florida’s southern coast. Initially rated as a Category 5 hurricane, Irma dropped to a Category 4 passing through the Florida Keys, and was then downgraded to a Category 2 hurricane when hitting mainland Florida. FEMA declared 48 counties within the state of Florida eligible for federal disaster assistance to individuals and households.

Summary:

BLS analysis suggests that the net effect of Hurricanes Harvey and Irma was to reduce the estimate of total nonfarm payroll employment for September 2017, although the impact cannot be precisely quantified because the CES survey is not designed to isolate effects from catastrophic events on CES employment, hours, and earnings.

In September, total nonfarm employment changed little (-33,000), after increasing by an average of 172,000 per month over the prior 12 months. A steep decline in food services and drinking places and below-trend growth in some industries likely reflected the impact of Hurricanes Harvey and Irma in September.

BLS closely monitored, and continues to monitor, collection rates in the affected areas. For the first preliminary September 2017 national estimates, the collection rates were within normal range for the nation and for the affected areas. (CES national collections rates are available at www.bls.gov/web/empsit/cesregrec.htm.) As such, BLS did not modify its methodology. BLS will continue to monitor collection rates in the affected areas for upcoming releases of employment data, including for the State Employment and Unemployment release on October 20, 2017.

The remainder of this webpage provides answers to questions related to Hurricanes Harvey and Irma and September 2017 payroll data.

How many jobs are in the disaster areas?

According to BLS’s Quarterly Census of Employment and Wages, about 11.2 million workers were employed in March 2017 in the FEMA-designated disaster counties and represented about 7.7 percent of national employment. Not all workers are within the scope of the CES survey. Farm laborers, the self-employed, and workers in households, for example, are excluded from the CES.

Establishments and employment in disaster counties
State Establishments* Employment* Percent of National CES Employment

Florida

594,588 7,694,762 5.3

Texas

187,930 3,461,172 2.4

Total

782,518 11,155,934 7.7
*Establishments and employment for March 2017 represents actual counts from the Quarterly Census of Employment and Wages

How did BLS collect data from businesses in the areas affected by the hurricanes and how will BLS continue data collection in those areas?

BLS collected data from businesses using normal collection procedures. However, BLS will continue to monitor data collection within the affected areas for the second preliminary and final estimates. The CES sample has just over 23,000 active reporting units in the disaster areas, representing about 6 percent of the entire CES active sample.

CES sample in disaster counties
State Worksites Percent of sample

Florida

16,300 5

Texas

7,000 2

Total

23,300 6

Over 50 percent of the sample in the affected areas is collected by means of direct electronic transmission (EDI). EDI is a method of collection which involves multi-establishment firms. Under EDI, the firm provides an electronic file to CES each month in a prescribed file format, which includes data for all of the firm's establishments. This one file transmitted to CES each month is sent from a single source. Since these sources are located mostly outside of the impacted areas, BLS expected and received normal EDI response in September.

Data are also collected by CES Data Collection Centers primarily through computer-assisted telephone interviews (CATI), representing about 20 percent of the active sample in the affected areas. Self-reporting via the web accounts for another 12 percent of the sample in these areas. Nonresponse prompts are typically emailed to respondents approximately 10 days prior to the initial collection deadline; a week before the initial collection deadline, those who have not responded will also receive a phone call reminder. A final round of nonresponse email prompts are sent the morning of the initial collection deadline. BLS expected, and received, normal CATI and web-based response in September.

Sample distribution by collection method in disaster counties in Florida and Texas
All disaster counties Worksites Percent distribution

      CATI

4,300 19

      EDI

12,500 54

      WEB

2,800 12

      OTHER

3,400 15

      Total

23,300 100
How is employment defined by the payroll survey?

The hurricanes caused large-scale evacuations and severe damage to many homes and businesses. In the CES survey, employees who are not paid for any part of the pay period that includes the 12th of the month are not counted as employed. Consider, for example, that an establishment is completely destroyed and there is no place to work. If the company pays workers during the designated pay period, CES still considers those workers to be employed. If, on the other hand, an individual does not work or receive pay for any part of the reference pay period, that individual is not counted as employed.

Was there a surge in the employment caused by paid emergency workers that went to the disaster area?

Emergency workers are counted by their employer’s established location, even if they are working in the disaster area. Volunteer workers are not counted as employed if they are not paid. Individuals called to active duty for the National Guard are considered to be on active military duty and are outside of the scope of the CES survey. If they were employed when called to duty, their pay status determines whether or not they are counted. If the employer pays the individuals for any part of the reference pay period, they are counted as employed. The paid individual is counted as employed regardless of the type of pay, such as pay for actual work time, paid leave, or a pay adjustment to make up the difference between an employee’s regular pay and their military pay.

Did CES estimation procedures change or will they change in upcoming releases of September 2017 data?

The CES program reviewed, and continues to review, all aspects of its estimation procedures to determine if any changes are necessary. For the first preliminary release of September 2017 national estimates, the CES program determined that no modifications to current methodology would be made because the collection rates in the affected areas were within normal ranges. BLS will continue to monitor the situation and any modifications to the methods for national CES estimates will be explained with the upcoming releases, as appropriate. Any potential modifications to methods for states and areas will be explained with the State Employment and Unemployment news release on October 20, 2017.

Do monthly U.S. employment estimates include Puerto Rico and the U.S. Virgin Islands, and how will estimates for these islands be affected by the recent hurricanes?

National CES total nonfarm employment estimates do not include Puerto Rico or the U.S. Virgin Islands.

The estimates of employment, hours, and earnings for Puerto Rico and the U.S Virgin Islands are published by BLS in the State Employment and Unemployment news release. Puerto Rico also produces geographic area estimates with guidance from BLS. These estimates are published in the Metropolitan Area Employment and Unemployment news release.

Because of the devastation caused by Hurricanes Irma and Maria, Puerto Rico and the U.S. Virgin Islands were not able to conduct normal data collection for their establishment surveys. Given the lack of adequate sample data, reliable September 2017 estimates for Puerto Rico and the U.S Virgin Islands cannot be produced at this time. BLS plans to publish revised August 2017 estimates on October 20, 2017, for Puerto Rico and the U.S. Virgin Islands. BLS will continue to monitor the situation and will produce estimates for these territories when adequate sample data become available.

Last Modified Date: October 6, 2017