4 ms·

It appears to largely be a manipulation / changes in the reporting -- Here's the FAQ attached to the report: https://www.bls.gov/cps/employment-situation-covid

It appears to largely be a manipulation / changes in the reporting -- Here's the FAQ attached to the report: https://www.bls.gov/cps/employment-situation-covid19-faq-may-2020.pdf https://www.bls.gov/cps/employment-situation-covid19-faq-may... > In the establishment survey, workers who are paid by their employer for all or any part of the pay period including the 12th of the month are counted as employed, even if they were not actually at their jobs. > The household survey is conducted by the Census Bureau and normally includes both in-person and telephone interviews, with the majority of interviews collected by telephone. Interviewing for the household survey began on May 17th, 2020. ... > In May about three-quarters of the interviewers at these centers worked remotely to collect data by telephone. > In April, BLS made changes to the establishment survey net birth-death model used in the estimation process. These changes were continued in May. However, after further research, BLS extended the estimation modifications back to March. This change contributed to the revision for March. > During their review of household survey data for May, BLS staff tested for outliers to determine whether any changes were needed to the seasonal adjustment models. BLS staff determined that, as in April, the vast majority of household survey data series had significant outliers in May. Therefore, BLS staff made adjustments to the models used in seasonal adjustment processing to better account for these outliers. > Of the 8.4 million employed people not at work during the survey reference week in May 2020, 5.4 million people were included in the “other reasons” category, much higher than the average of 549,000 for May 2016–2019 (not seasonally adjusted). BLS analysis of the underlying data suggests that this group included workers affected by the pandemic response who should have been classified as unemployed on temporary layoff. Such a misclassification is an example of nonsampling error and can occur when respondents misunderstand questions or interviewers record answers incorrectly. BLS and the Census Bureau are investigating why this misclassification error continues to occur and are making changes for the June collection. (See item 14 below.)

awillen 6y ago
Thanks for this - I figured there had to be at least a partial explanation related to methodology, but this is the first I've seen of the details.
koheripbal 6y ago
It's common for them to add a FAQ to the report - I don't see how any of these points would have "manipulated" the data to the huge degree that we've seen. If anything, the first point would mean that the prior jobless numbers were higher than they should have been.
- vannevar 6y ago
  I agree with you that "manipulated" seems like an exaggeration, these notes all involve decisions that don't seem unreasonable. But however you slice it, it's hard to interpret that first point as a good sign for the future---how long do you think employers will continue to pay employees who are not actually working?
frankchn 6y ago
The correction would be about 3% if we reclassify those as unemployed: https://www.bls.gov/news.release/pdf/empsit.pdf https://www.bls.gov/news.release/pdf/empsit.pdf > If the workers who were recorded as employed but absent from work due to “other reasons” (over and above the number absent for other reasons in a typical May) had been classified as unemployed on temporary layoff, the overall unemployment rate would have been about 3 percentage points higher than reported (on a not seasonally adjusted basis). However, according to usual practice, the data from the household survey are accepted as recorded. To maintain data integrity, no ad hoc actions are taken to reclassify survey responses. If you apply the correction, the rate would be 16.3%.
mud_dauber 6y ago
great summary. Props.

awillen 6y ago

Thanks for this - I figured there had to be at least a partial explanation related to methodology, but this is the first I've seen of the details.

koheripbal 6y ago

It's common for them to add a FAQ to the report - I don't see how any of these points would have "manipulated" the data to the huge degree that we've seen. If anything, the first point would mean that the prior jobless numbers were higher than they should have been.

vannevar 6y ago
I agree with you that "manipulated" seems like an exaggeration, these notes all involve decisions that don't seem unreasonable. But however you slice it, it's hard to interpret that first point as a good sign for the future---how long do you think employers will continue to pay employees who are not actually working?

frankchn 6y ago

The correction would be about 3% if we reclassify those as unemployed: https://www.bls.gov/news.release/pdf/empsit.pdf https://www.bls.gov/news.release/pdf/empsit.pdf > If the workers who were recorded as employed but absent from work due to “other reasons” (over and above the number absent for other reasons in a typical May) had been classified as unemployed on temporary layoff, the overall unemployment rate would have been about 3 percentage points higher than reported (on a not seasonally adjusted basis). However, according to usual practice, the data from the household survey are accepted as recorded. To maintain data integrity, no ad hoc actions are taken to reclassify survey responses. If you apply the correction, the rate would be 16.3%.

mud_dauber 6y ago

great summary. Props.