Employment as a Signal of Economic Performance
Total employment rises and falls with real GDP. Pull both series yourself and learn to read employment as a procyclical, lagging indicator.
Reading
0%
Time left
~20 min
Quiz score
0/4
What this means
An economy's performance is not directly observable. There is no dial anywhere that reads "the economy is doing well." What exists instead is a set of measurements, each capturing a piece of the picture, and one of the most informative of them is the total number of people working.
Total employment matters because production requires labor. Firms do not hire workers as a favor; they hire because they expect to sell what those workers produce. When firms expect demand to expand, they add workers. When they expect demand to contract, they stop hiring, then cut hours, then cut jobs. Employment therefore carries information about what firms across the economy believe about the near future.
Because employment and output are linked through production, they tend to move together. An indicator that rises when the economy expands and falls when it contracts is called procyclical. Employment is procyclical. So is real GDP, which is why the two series trace similar shapes over time.
The relationship is close but not simultaneous, and the timing distinction is where this gets interesting. Economists sort indicators by when they move relative to the cycle. A coincident indicator turns at about the same time as the economy. A lagging indicator turns afterward. A leading indicator turns before.
Employment sits awkwardly across these categories, and that is not a flaw in the classification. Payroll employment behaves largely as a coincident indicator, while the unemployment rate behaves as a lagging one. The reason is that firing and hiring are expensive and slow. A firm that senses weakening demand first cuts overtime and lets vacancies go unfilled, because dismissing trained workers destroys value the firm paid to create. In a recovery the same caution runs in reverse: a firm restores hours and pushes existing staff harder before committing to new hires. Output can therefore be rising for months while the labor market still looks bad, a pattern commentators sometimes call a jobless recovery.
Why it matters
If you are entering the labor market, this timing is not academic. Headlines announcing that a recession has ended describe output, not hiring. A graduating class can enter a labor market that is still weak by the measure that matters to them while the economy is technically expanding. Knowing which indicator lags tells you how to read that news.
It also disciplines your reasoning about policy. Judging a policy by employment data collected immediately after it takes effect confuses a lag for a failure. Any serious argument about whether something worked has to account for how long the labor market takes to respond.
Real-world example
Two federal agencies publish the series you need, and both are free. The Bureau of Labor Statistics reports total nonfarm payroll employment monthly from its establishment survey, which asks employers how many people are on their payrolls. The Bureau of Economic Analysis reports real GDP quarterly. The Federal Reserve Bank of St. Louis republishes both in FRED, where you can graph them on the same axes and shade the recession bars automatically. Nobody has to take an economist's word for the relationship between employment and output; the underlying data is public and you can plot it in a few minutes.
Try it
- Choose a ten-year window and commit to it in writing before you look at anything. Pick a span that contains at least one recession, since the relationship you are studying is most visible at turning points.
- Pull real GDP for that window from FRED or directly from the Bureau of Economic Analysis. Record the exact series you used, its units, and whether it is seasonally adjusted. Note that GDP is quarterly.
- Pull total nonfarm payroll employment for the same window from FRED or the Bureau of Labor Statistics. Record the same details. Note that this series is monthly, so you will need to decide how to align it with quarterly GDP, and you must state your method.
- Convert both series to percent change from the previous period rather than levels. Levels both trend upward over a decade and will appear to agree no matter what; the changes are what reveal the cyclical relationship.
- Graph the two change series on a shared time axis. Mark recession periods, which FRED will shade for you.
- Describe the relationship in precise language. Do the series move in the same direction, opposite directions, or unrelated directions? Answer with reference to specific dates in your window, not impressions.
- Examine the turning points specifically. Find the month or quarter when real GDP stopped falling and started rising. Then find when employment did the same. Report the gap in months. Do the same at the peak, where the economy turned from expansion to contraction.
- Based only on your own data, classify employment as leading, coincident, or lagging, and defend the classification. If your evidence is mixed or the gaps differ between the peak and the trough, say so rather than forcing a clean answer.
- Write one paragraph on what your two series do NOT tell you. Consider people who stopped looking for work, people working part time who want full time, and changes in job quality or pay. None of these appear in either series.
Teacher note
Step 4 decides whether this lesson works. Students who graph levels see two upward-sloping lines, conclude the indicators agree, and learn nothing, because almost every macroeconomic level series trends up over a decade. Requiring percent change forces them to confront the cycle itself. Expect resistance, since the levels graph looks tidier and the change graph looks noisy; the noise is the data.
The frequency mismatch in step 3 is a genuine methodological problem, not busywork. Students may average the three months in a quarter, take the final month, or plot on separate axes. Any defensible choice is acceptable as long as they name it and apply it consistently. Watch for students who silently drop months to make the series line up.
Two misconceptions surface reliably. The first is causal collapse: students assert that employment causes GDP or that GDP causes employment, when the honest reading of a co-movement graph is that the two are jointly determined by conditions neither series shows. Ask what a graph of two correlated series can and cannot establish. The second is expecting the lag to be identical at peaks and troughs. It usually is not, and a student who notices the asymmetry and reports it rather than smoothing it over is doing real empirical work.
Do not supply any figures yourself, including the length of the lag. The entire point is that students obtain the relationship from data rather than from you. A student has it when they can state the direction of the relationship, quantify the timing gap from their own series, and explain why hiring responds slowly using the firm's perspective on the cost of hiring and firing.
Check yourself
What does it mean to say total employment is a procyclical indicator?
Real GDP has been rising for two quarters, but employment has barely moved. What is the most reasonable interpretation?
Why should you graph percent change rather than levels when comparing employment and real GDP over ten years?
A student concludes from a graph showing employment and real GDP moving together that rising employment causes rising GDP. What is the strongest objection?
Employment rises and falls with real GDP, but it turns after output does, which is why the labor market can still feel like a recession months after the economy has started growing again.