Link to What's New ThisWeek Jobs as Counted Commodities

Dear Habermas Logo and Link to Site Index A Justice Site



Labor Market

Mirror Sites:
CSUDH - Habermas - UWP - Archives

California State University, Dominguez Hills
University of Wisconsin, Parkside
Created: April 7, 2004
Latest Update: April 10, 2004

E-Mail Icon jeannecurran@habermas.org
takata@uwp.edu

Wal-Mart and the Job Market
Jobs as Counted Commodities

Site Copyright: Jeanne Curran and Susan R. Takata and Individual Authors, April 2004.
"Fair use" encouraged.

This piece is based on Labor's Lost Jobs, a New York Time's Op-Ed piece on the jobless recovery, reinterpreting it, by Tim Kane of the Heritage Foundation, a conservative think tank. Backup Another conservative viewpoint on jobs is Jay Nordlinger's "Voting 'No' on Low Prices and Good Jobs," in the LA Times, Friday, April 8, 2004. Backup.

Numbers say what you tell them to say substantively. For example, what's a "good job"? On for which you get up every day, show up on time, please customers, and don't complain? Or one at which you make enough money to support yourself and/or your family above poverty level? The left wing position is generally that the community, the society owes it to each of its members to provide adequate means for them to earn enough to eat, to wear, and a place to live. The right wing position is generally that such community needs will take care of themselves if the market is allowed to operate freely.

I'm left radical. That means I don't believe the market has ever operated freely, but has operated always to the advantage of the privileged owners. The right conservative position is that the market does operate freely, and that people could live above the poverty level if they just tried hard enough. The left position is that there needs to be a community or governmentally controlled safety net to protect those who fall through the cracks of opportunity. That doesn't mean they're all hard-working caring infdividuals. But it does mean they are all human, the products of many interdependent personal and interpersonal relations with the infrastructure, and that the minimal safety net to provide food, clothing, and housing must be maintained by all of society. The right position is that indivudals should be responsible for themselves and that Mother Bountiful charity will provide for the rare cases that need help. That means basically privatized corporate charity. But there are no ground rules as to how much charity will be provided and to whom.

Consider, for example, the Bill Gates Foundation. He provides scholarships to the worthy minorities he seeks to help. Who decides who is worthy, under what circumstances, and how much help is needed. Bill Gates, or his foundation. Same thing. This is contrary to the left position which sees no excuse for Bill Gates having such control over another's life.

Walmart spends money helping communities builld softball fields. Gee, that's nice. But who gets to decide how much, which side of town, what the rules are for using the field, and where the adults are who will teach the citizenship needed to maintain equality of athletic opportunity? The right position would say that Walmart's corporate giving is good for the community. The left position would say that if the worker's were paid living wages they wouldn't need Walmart's charity to maintain their athletic fields.

There are very few people who find themselves way to the left anymore, because most of us have accepted scholarships, prizes, discounts, special gifts, charity, etc. from so many corporations that we're scared to let go of it. Walmart spent over a million dollars trying to break the city council of Inglewood and circumvent it. No matter how many scholarships they give us we can never hope to spend such sums to defend our right to govern ourselves. Votes become commodities, to be bought and sold in this process. And ethics are corrupted. Nevermind what Walmart's ultimate intent was. It sought control. Control above and beyond the governance of the citizens of this country. That is scary, indeed.

The first article deals with whether or not there are or are not jobs out there to be had. That means we have to think on how "job" is defined. Tim Kane says that we should use the labor department's measure of household jobs instead of its measure of payroll jobs. Now, in order for you to respond critically and reasonably to all the labor market issues that are going to come up in the next election, you're going to need to know the difference between those measures. Here's what the labor department tells us:

From Preliminary Benchmark Revision - Questions and Answers 2003 Benchmark Revision U.S. Department of Labor. Bureau of Labor Statistics

household jobs - what's the measure?

payroll jobs- what's the measure?

    (1) What is the establishment payroll survey?

    The establishment payroll survey, also known as the Current Employment Statistics (CES) survey, is a monthly sample survey which includes 400,000 individual business and government agency worksites drawn from a sampling frame of Unemployment Insurance tax accounts. The primary statistics derived from the survey are monthly estimates of employment, hours, and earnings for the nation, states, and major metropolitan areas. Preliminary national estimates for a given reference month typically are published on the first Friday of the following month, in conjunction with data derived from a separate survey of households, the Current Population Survey (CPS). The CPS is the source of statistics on the activities of the labor force, including the nation's unemployment rate.

    . . .

    (V) Comparison with the Household Survey

    (1) Both the household survey and the payroll survey measure employment. Why do the measures differ?

    The household survey provides a count of employed persons age 16 and over (multiple job holders are counted only once), including the self-employed, agricultural workers, private household workers, and unpaid workers in family businesses. The payroll survey provides a count of wage and salary jobs, excluding agriculture, private households, and self employed.

    (2) Why does BLS recommend the payroll survey as the preferred measure of nonfarm wage and salary employment change?

    Both surveys provide valuable information and have strengths and limitations. However, the payroll survey has a much larger sample (400,000 establishments) than the household survey (60,000 households).

    Also, the payroll survey benchmarks once a year to the UI data. The household survey does not have a benchmark to a universe count of employment.

    In addition, the CES sample size yields less volatile series and a lower threshold for a statistically significant change than the household survey. A monthly change is significant for the payroll survey at 105,000, and the household change is significant at 290,000. However, the payroll survey does not include self-employed persons, those employed in agriculture, and in private households. The household survey does provide data for those components not obtainable from the payroll survey.

    Last Modified Date: October 3, 2003

    From Preliminary Benchmark Revision - Questions and Answers 2003 Benchmark Revision U.S. Department of Labor. Bureau of Labor Statistics

Discussion Questions

  1. What's the sample for the Payroll Survey?

    "The establishment payroll survey . . . is a monthly sample survey which includes 400,000 individual business and government agency worksites drawn from a sampling frame of Unemployment Insurance tax accounts."

  2. What's a sampling frame?

    A sampling frame is a list of every possible member of a population. This comes up with issues like the census. One thinks of the census as just having to send a census from to each house or apartment. Not so. Several families may live in the same house together, not in separate apartments. People may live in illegal rental places, like garages, or guest houses. People live in their cars when they have no other space, and some live under bridges. So an important part of the census taking is finding all the possible places that people could be, and making sure that each of them is included in the sampling frame. This is a very expensive undertaking. One reason that surveys are so expensive. Easy to find every house that has a telephone. But not every house does.

  3. Do you still have sampling error if you have a sampling frame?

    YES. A sample is a sample is a sample, and has sampling error.

  4. Does the sample for the Payroll Survey include government jobs and payrolls?

    Yes. "400,000 individual business and government agency worksites".

  5. What's the population for the Payroll Survey?

    The population is all the individual business and government agency worksites included in the sampling frame of Unemployment Insurance tax accounts.

  6. If a small, upstart company couldn't afford and didn't pay its unemployment insurance tax, would it be included in the sampling frame?

    Only if the people who made up the frame found it.

  7. How often is the Payroll Survey conducted?

    The Payroll Survey "is a monthly sample survey."

  8. How are the Payroll Survey Statistics interpreted?

    "in conjunction with data derived from a separate survey of households, the Current Population Survey (CPS)". So the figures of payroll jobs are combined with the figures of unemployed adults, taken for the household survey. Suggests that neither tells the whole picture. How are these data combined? Good question, but not unless you're really going to follow through in this area. I don't know. Not my area. jeanne

  9. Why include the household survey? What does it cover that the payroll survey doesn't?

    "The household survey provides a count of employed persons age 16 and over (multiple job holders are counted only once), including the self-employed, agricultural workers, private household workers, and unpaid workers in family businesses." So where the payroll survey counts the number of people receiving salaries from business and agencies that pay unemployment insurance taxes, the household survey counts people who are almost certainly not covered by unemployment insurance taxes. If you added the two counts together, you'd be assuming that their weights were the same. For example, that one 16-year-old who works for free in the family restaurant counts the same in the work force as an executive who heads the personnel department of a large agency. Can you see why statisticians might be able to interpret this data differently by playing with it, like by counting two 16-year-old non-paid workers to equal one exec? I made that up. I don't know how they add them up. I just want you to see the complexity of counting this way, and then insisting that the way you've interpreted it is the "right" way to interpret it, and the "only" way.

  10. Why does the Bureau of Labor Statistics recommend the use of the Payroll Survey for general labor statistics?

    Because the data are more sensitive to change. "A monthly change is significant for the payroll survey at 105,000, and the household change is significant at 290,000." That means that you can see the effects on the job market more readily, and the household survey only increases sensitivity by adding the self-employed. Especially since people self-employ when regular employment is not available, I'd be suspicious of this self-employment statistic.

  11. Would any of these statistics tell us the difference in the effects of working for Wal-Mart at a less-than-adequate raise to support your children and working for L.A. county?

    No. In each case you would have one salary, and no way to tell how that salary affected the person's life style or poverty level.

  12. Does the addition of household survey data add more low income or high income people to the mix of the employed?

    Consider the average wages of those who work in agriculture, self-employed, and who work in private households.