The Establishment SurveyIn addition to the household survey, the BLS also uses theestablishment survey, (a.k.a. the payroll survey),This survey samples~300,000 establishments, or places ofemployment,about theiremployees.Disadvantages include: Self-employed people not surveyed (not on a company payroll)Newly-opened firmsoftenomittedInformation on employment only, not unemployment. Numbers fluctuate depending on establishments included, oftenrequiringlarge revisionsHowever, a big advantage is that the data are determined by realpayrolls, not self-reporting like the household survey.162015PearsonEducation,Inc
© 2015 Pearson Education, Inc. 16 The Establishment Survey In addition to the household survey, the BLS also uses the establishment survey, (a.k.a. the payroll survey). This survey samples ~300,000 establishments, or places of employment, about their employees. Disadvantages include: • Self-employed people not surveyed (not on a company payroll) • Newly-opened firms often omitted • Information on employment only, not unemployment • Numbers fluctuate depending on establishments included, often requiring large revisions However, a big advantage is that the data are determined by real payrolls, not self-reporting like the household survey
Comparing the Household and Establishment SurveysThe table below gives the data from the July and August 2013household and establishment surveys:Household SurveyEstablishmentSurveyJulyJulyAugustChangeAugustChange-115,000Employed144,285,000144,170,000135,964,000136,133,000169,000Unemployed11,514,00011,316,000-198,000Laborforce-312,000155,798,000155,486,0007.4%7.3%0.1%UnemploymentrateTable9.1Household andestablishment surveydataforJulyandAugust2013Even if all surveys are truthfully and accurately answered, we donot expect the numbers to be identical between the two surveys:Differentgroupsaremeasured·All surveys have measurementerrorsBut weget a morecompletepicture byconsidering both surveys17@2015PearsonEducafion.lnc
© 2015 Pearson Education, Inc. 17 Comparing the Household and Establishment Surveys The table below gives the data from the July and August 2013 household and establishment surveys: Even if all surveys are truthfully and accurately answered, we do not expect the numbers to be identical between the two surveys: • Different groups are measured • All surveys have measurement errors But we get a more complete picture by considering both surveys. Household and establishment survey data for July and August 2013 Table 9.1 Household Survey Establishment Survey July August Change July August Change Employed 144,285,000 144,170,000 –115,000 135,964,000 136,133,000 169,000 Unemployed 11,514,000 11,316,000 –198,000 Labor force 155,798,000 155,486,000 –312,000 Unemployment rate 7.4% 7.3% –0.1%
Revisionsto EmploymentNumbersRevisionstoemploymentdata(thousandsofworkers)150100500-50-100150200-250-300-350JunDecJunDecJunDecDec2007200820082009200920102010Over time, the BLS adjusts its estimates ofFigure 9.5employment and unemployment for previous months. Revisions toRevisions sometimestake place years later.employmentchangesasreportedintheThe large negative revisions were because the BLSestablishmentsurveyunderestimatedthe severityofthe 2007-2009recession.18@2015PearsonEducation.lnc
© 2015 Pearson Education, Inc. 18 Revisions to Employment Numbers Over time, the BLS adjusts its estimates of employment and unemployment for previous months. Revisions sometimes take place years later. The large negative revisions were because the BLS underestimated the severity of the 2007-2009 recession. Revisions to employment changes, as reported in the establishment survey Figure 9.5