Tomorrow, the Employment Situation Report comes out and as usual, there will be idiots claiming the government doesn't count people not receiving unemployment benefits, or that the government isn't counting people who should be counted etc. So let's discuss. Question: What exactly do we want to measure and why? Answer: We want to measure the Labor Market and how it is changing, especially how difficult it is to get a job. First difficulty is that not everyone can freely participate in the labor market. There are laws and barriers afffecting groups of people. While children can work, there are significant restrictions as to field, hours, changing jobs etc. At approx age 16 the restrictions are gone, almost all jobs are open for full or part time work. So we ignore anyone under 16. Next group are prisoners. While some "work" it's not really a labor market system and working or not working are not free choices of entry and exit. So we ignore prisoners. Next group are those mentally or physically incapable of working. While many disabled and those with psychological conditions do work, those who are institutionalized or in long term care (such as a hospice) can't. So we ignore those in an institution or long term care. There remains many disabled who can't work but aren't in an institution, but since some can and some can't work, there's no bright line, so we include them all. Next: Retirees. Many countries exclude those over the age of 65. For a country with mandatory retirement, or generous government pensions, that is sensible as almost all those over 65 don't need to work and don't work and including them can distort the picture. But in the US, there is no mandatory retirement (except in military/police) and retirement benefits are not as generous or universal. So no maximum age limit for us. Last, military. For a while, they were counted as employed (1984-1994) but since there's no free exit and it's not a market system, including them just raises the number of employed by a fairly constant level, so it's better to exclude the military. So we have our population: The Adult Civilian Non-Institutional Population. How do we divide that up? The key is to be as objective as possible. Since we have to use a survey (it's just not possible to do a full count every month), any subjectivity will increase our error, so we have to avoid any possible overlaps or judgement calls. Let's look at Employed. Who is employed? Some have suggested a minimum number of hours worked, but is that a good idea? Plenty of people voluntarily work few hours, or in a particular week only work a few hours. Should we say they don't really have a job or aren't really working? That's very subjecive. We don't want to run into cases were someone nomrmally works 40 hours a week, but in the particular week of the survey she only worked 2 hours and counted as unemployed. So we set a bright line of 1 hour for pay, and also include those as employed people temporarily absent due to brief illness, injury, vacation, strike, bad weather etc. We can always look at sub groups later. Unpaid labor. Some people don't get paid (directly) for work, either charity work or working on a family farm or family business. Let's not include charity work, because that's not really part of the labor market. But family business/farm is a different story...the person is either just helping out, (meaning if they don't already have another job they wouldn't take one if they weren't working for family and/or no one would have to be hired in their place) or they are a substantial part of the business preventing them from another job and filling a job slot no one else can take. So let's set a bright line of 15 hours a week and anyone working without pay on family business farm 15 hours or more a week is employed. What about someone hired but not yet working? Many things can happen, so we won't call them employed until they're actually working. Now for the Unemployed. In casual usage, this can mean many different things, but what we want is a precise, objective definition. And what we're looking for is available workers who aren't working. This has an impact on the labor market..supply over demand. Some people don't want to work...retirees, housewives/husbands, full time students, and some people on drugs or alcohol or criminals or lazy. So let's not count anyone who doesn't want a job. Should we count everyone who says they want a job? They could be lying. They could be unable to take a job if offered (meaning they're not available labor). So what bright line can we set? If someone is actually looking for work, sending a resume, going to an interview, on the union list, then we can safely assume that they are really available for work. But if someone isn't DOING anything about a getting a job, then they're not available, anymore than someone who doesn't want to work. Sure they may still consider themselves "unemployed" and they may sincerely want to work, but since they're not available, since they're not part of the actual supply of labor and can't be hired, we exclude them. So we have our Employed, we have our Unemployed....this is our Labor Force, our available supply of people who can work or be hired. Everyone else is Not in the Labor Force. Since the population tends to grow, for an historical comparison, it's better to look at rates/ratios/percentages to get a clearer picture. Labor Force Participation Rate: The Labor Force as a percent of the Population. This can be largely affected by non-economic factors as the increase of women into the labor force in the 60s and 70s showed, but it gives a good look at how much of the population is trying to work. Employment-Population Ratio: this tells us the percent of the population that's actually working. Again, non-economic factors play a role but looking at the changes can tell us useful things. Unemployment rate: Unemployed as a percent of the Labor Force. This tells us how much of available labor...people willing and able and available to work... is not being used. This is all economic factors...the percent of people trying and failing to get a job. Does this give a complete picture of the Labor Market? Of course not...it's not meant to. It doesn't tell us types of job, if people are working as much as they want to, people who don't believe they would be successful. But it tells us what we're looking for: the percent of people who we know for sure cannot get a job. I think that explains the basics. I'll be happy to clarify questions or discuss any real issues.