Chapter 1: Introduction to Labor Economics#

Why do some people earn more than others? Why does a country’s unemployment rate rise or fall? This chapter introduces labor economics — the study of how workers, employers, and governments shape wages, jobs, and careers. You’ll learn the basic numbers economists use to size up a labor market and the key tool, regression, that helps them spot patterns in data.

The Big Picture#

Labor economics tries to answer a simple but deep question: Who gets which jobs, and why? The answer comes from millions of people making choices — how long to stay in school, whether to hunt for a new job, when to retire — and thousands of firms deciding whom to hire and how much to pay. Governments set rules, taxes, and safety nets that affect those decisions. This chapter gives you a roadmap. We’ll meet the three main players, learn the key numbers that describe a labor market, draw a clear line between what is and what should be, and then get comfortable with regression, the basic statistical tool that helps us move from simple observations to solid evidence.

Who Are the Players?#

Every labor market has three types of decision-makers: workers, firms, and the government. Understanding their goals is the first step.

Workers supply their time and effort. Economists assume workers try to get as much overall happiness — or utility — as they can from the options they have. Utility comes from two main sources: the goods and services they can buy with their income, and the free time (leisure) they enjoy. Taking a second job gives up leisure but brings in more money. Retiring early trades income for more free time. This constant balancing act is what we mean by utility maximization, and it drives every labor supply decision. Workers also invest in themselves by going to school or learning on the job, hoping those investments will pay off later with higher earnings.

Firms demand labor. They don’t hire out of kindness; they hire because workers help produce goods or services to sell. Economists usually assume firms try to maximize profits. To decide how many workers to hire, a firm compares the extra revenue from adding one more worker — the marginal revenue product of labor — with the extra cost, which is the wage. If that extra revenue is greater than the wage, hiring adds to profit. When the two become equal, the firm stops hiring. This profit-maximizing logic gives us a clear way to think about the demand for labor.

Utility: The economist’s word for the satisfaction or well-being a person gets from consumption, leisure, and other aspects of life.

Marginal revenue product of labor: The additional revenue a firm earns by employing one more unit of labor, holding everything else constant.

The government sets the rules of the game. It collects taxes, provides public goods, runs social insurance programs like unemployment benefits and pensions, and passes laws on working conditions, minimum wages, and discrimination. Every one of these actions nudges the decisions of workers and firms, sometimes on purpose and sometimes as an unintended side effect.

📝 Section Recap: Labor markets are shaped by workers balancing income and leisure to get the most satisfaction, firms chasing profit by comparing extra output value to the wage, and governments influencing everything with taxes, transfers, and rules.

Measuring the Labor Market#

Before we can explain what’s happening, we need some clear numbers. The most basic ones come from dividing the adult population into groups.

Start with the working-age population — everyone who is old enough to work and not in the military, prison, or long-term care. From this group we carve out the labor force: people who either have a job or are actively looking for one.

Labor force: The sum of employed and unemployed workers. It does not include full-time students, retirees, stay-at-home parents, or others who are not seeking paid work.

• Employed: Anyone who did any work for pay (or profit) in the reference period, even if it was just a few hours.
• Unemployed: Someone who did not work but actively searched for a job and was available to start.
• Not in the labor force: Everyone else — students, retirees, stay-at-home parents, discouraged workers (people who stopped looking because they couldn’t find a job), and those unable to work.

From these groups we build two key statistics:

The labor force participation rate answers, “What share of the working-age population is either working or trying to work?”

The unemployment rate zooms in on those in the labor force who cannot find a job:

Suppose a country has 200 million working-age adults. Of those, 140 million are employed and 10 million are unemployed but searching. The labor force is 150 million. The participation rate is , and the unemployment rate is .

Notice that the unemployment rate says nothing about people who have left the labor force. A drop in unemployment might be good news (more people found jobs) or bad news (discouraged workers stopped searching and are no longer counted). That’s why economists always look at both rates together.

📝 Section Recap: The participation rate shows the share of adults who are working or trying to work. The unemployment rate shows the share of those active people who can’t find a job. You need both numbers to really understand the labor market.

Positive versus Normative Economics#

When an economist says, “Raising the minimum wage reduces employment among teenagers,” she’s making a claim about what is — a positive statement. When she says, “We should raise the minimum wage to help low-income families,” she’s making a claim about what should be — a normative statement. This distinction is one of the most important habits of mind in economics.

Positive economics deals with what is. It tries to describe and explain the world without taking sides. Positive statements can be tested against data — they are either true or false, even if we don’t yet know which. “Immigration lowers the wages of native-born workers with similar skills” is a positive claim. We can collect evidence and, in principle, settle the matter.

Normative economics deals with what ought to be. It rests on value judgments that data alone can’t settle. “The government should restrict immigration to protect domestic workers” is a normative claim. It may be informed by positive findings, but ultimately depends on whose well-being we care about more.

Positive economics: Analysis that tries to explain how the economy works, free of personal opinions about what is good or bad.

Normative economics: Analysis that prescribes what should be done, based on explicit or implicit value judgments.

Why does this matter? Because policy debates mix the two constantly. A clear thinker learns to spot which part of an argument is a testable hypothesis and which part is a moral stance. Good labor economics provides the positive foundation — the best evidence we have — so that normative choices are made with eyes open.

📝 Section Recap: Positive economics asks “What is?” and can be checked with data. Normative economics asks “What should be?” and involves values. Keeping them apart makes analysis honest.

Regression: A First Look#

Labor economics is based on evidence. We want to know how much an extra year of schooling boosts earnings, whether a higher minimum wage destroys jobs, or if job training programs actually help. The most common tool for answering such questions is regression analysis.

Imagine we have data on a group of workers: their hourly wage and their years of schooling. We suspect that more schooling is linked to higher wages. A simple regression writes this idea as a straight line:

Here, labels each worker. The dependent variable (the thing we want to explain) is the wage. The independent variable (the possible cause) is years of schooling. and are numbers we estimate from the data. is the error term, which captures everything else that affects worker ’s wage but isn’t in the model — luck, ability, family background, and so on.

Dependent variable: The outcome we are trying to explain or predict (often called ).

Independent variable: The factor we think might influence the outcome (often called ).

The slope coefficient is what we focus on. It tells us how much we expect the dependent variable to change, on average, when the independent variable goes up by one unit. If the estimate , then each extra year of schooling is associated with about $0.10 more per hour. That doesn’t mean schooling causes the higher wage — we’ll get to that — but it describes the pattern in the data.

When we run a regression on a sample, the estimates aren’t perfect. The standard error of measures how much the estimate would bounce around if we took a new sample from the same population. A small standard error means the estimate is precise; a large one means it’s noisy.

Standard error: A measure of the statistical uncertainty surrounding an estimated coefficient. It tells us how much sampling variability we should expect.

To judge whether a slope coefficient is reliably different from zero, we use the t-statistic:

A rough rule of thumb: if the absolute value of is greater than about 2, we can be pretty confident that the true slope isn’t zero. In that case, we say the estimate is statistically significant at usual levels. For example, if with a standard error of , the -statistic is , which is strong evidence that schooling and wages are positively related in the data.

t-statistic: The ratio of an estimated coefficient to its standard error. Large absolute values (usually above 2) suggest the estimate is statistically significant.

📝 Section Recap: Regression fits a line through a cloud of data points. The slope tells us the average link between two variables, the standard error tells us how precise that estimate is, and the t-statistic helps us decide whether the link is likely real or just sampling luck.

Multiple Regression and Causal Inference#

A simple regression with one independent variable is often not enough. The world is messy. Workers with more schooling also tend to have higher ability, come from wealthier families, and live in cities with more job opportunities. If we leave those factors out, the slope on schooling will pick up their effects, giving us a misleading answer. That problem is called omitted variable bias.

Multiple regression lets us bring several independent variables into the same equation:

Now estimates the relationship between schooling and wages while keeping experience and ability constant. By including control variables, we compare workers who are similar in every measured way except their schooling. This gets us closer to isolating the partial relationship we care about.

Multiple regression: A regression that includes more than one independent variable, allowing the researcher to hold other factors constant and reduce omitted variable bias.

Even with many controls, regression alone does not prove causation. Causal inference is the challenge of moving from “X is correlated with Y” to “X causes Y.” In labor economics, that leap is everything. Does education actually make someone more productive, or do more productive people simply get more education? Does a higher minimum wage cause job loss, or do minimum wages rise in places that already have strong job markets?

The gold standard for causal inference is a controlled experiment: randomly assign some people to a treatment (say, a job training program) and others to a control group, then compare outcomes. In labor markets, true experiments are rare. Instead, economists rely on natural experiments — real-world situations where something changes for reasons unrelated to the outcome. A new law that raises the school-leaving age in one state but not the next creates a kind of quasi-experiment. Economists can then use that policy change to estimate the causal effect of schooling on wages, often with techniques like difference-in-differences or instrumental variables. Don’t worry about those names now; the simple idea is: find a source of variation in the independent variable that mimics a random experiment, and trace its effect.

Causal inference: The process of determining whether a change in one variable actually causes a change in another, rather than merely being correlated with it.

Multiple regression is the starting point. It helps us describe the data and control for obvious confounders. But whenever you see a regression table, ask yourself: What makes the independent variable vary? Could something unmeasured be driving both the treatment and the outcome? The best labor economics blends careful statistics with a research design that makes a causal story believable.

📝 Section Recap: Multiple regression adds control variables to reduce omitted variable bias, but it doesn’t automatically prove cause and effect. To establish causation, you need a research design — like a natural experiment — that makes the independent variable change for reasons unrelated to the outcome.

Summary#

We’ve introduced the key players in the labor market and the numbers used to track it. We saw how important it is to separate facts from values. And we learned regression, a powerful tool that helps us measure relationships in data, while remembering that correlation isn’t always causation.