Why Cyclical Co-Movements Matter
Business-cycle analysis often compares real output, employment, unemployment, and inflation. These series do not all behave in the same way. Knowing their typical patterns helps organize economic evidence, but it is equally important to distinguish an observed tendency from a relationship that holds by definition. In particular, employment often moves with output and unemployment often moves against it, while inflation can behave differently across episodes.
The word stylized matters: a stylized fact is a broad empirical regularity, not a law that every country, industry, or business cycle must obey. A candidate should be able to identify the usual pattern, calculate the measures correctly, and recognize when the data show an exception.
Output and Employment
Real gross domestic product (real GDP) measures the inflation-adjusted value of final goods and services produced in an economy over a period. Employment counts people who are working under the definition used by the relevant statistical measure. During many expansions, firms produce more and hire additional workers, so real output and employment tend to move in the same direction. During many contractions, output and employment both weaken. This is the usual basis for describing employment as procyclical.
The tendency is not an identity. Output can change without an equal change in employment because production also depends on hours worked, capital, and output per worker. For example, firms might meet higher demand by increasing hours or using existing capacity before hiring. Conversely, employment can change for reasons that do not track current output, such as changes in workforce availability or hiring decisions. Therefore, “output rises” does not mathematically require “employment rises” in every observation.
Cyclical comparisons also depend on what is measured. The number of employed people is a level, whereas employment growth is a rate of change. A level can remain high while growth slows, or fall only after output has already weakened. Analysts may compare growth rates, index values, or deviations from trend; those are related but not interchangeable descriptions.
Worked Example: Classifying Output and Employment Movements
A hypothetical economy reports the following seasonally adjusted indexes, with the first quarter set to 100. Real output is 100, 98, and 96 in three successive quarters. Employment is 100, 99, and 98. Classify the movements and calculate the quarter-to-quarter percentage changes.
Method. Calculate each percentage change as the change in the index divided by its previous-quarter value, then multiply by 100. This distinguishes the direction of movement from the size of the percentage change.
From quarter 1 to quarter 2, output changes by \(98-100=-2\) index points. Its percentage change is \((-2/100)\times100=-2.00\%\). Employment changes by \(99-100=-1\) point, or \((-1/100)\times100=-1.00\%\).
From quarter 2 to quarter 3, output changes by \(96-98=-2\) points, so its percentage change is \((-2/98)\times100\approx-2.04\%\). Employment changes by \(98-99=-1\) point, giving \((-1/99)\times100\approx-1.01\%\).
Both measures fall in both intervals, so these observations show output and employment moving together, consistent with their usual procyclical pattern. The percentage declines are not identical: an equal index-point change need not imply an equal percentage change when the starting values differ. The common mistake is to treat the shared direction as proof that employment and output have a fixed one-for-one relationship. These three observations do not establish such a rule.
Check. The final indexes are below their initial values for both series, and every calculated quarter-to-quarter change is negative, matching the directions read from the data.
Unemployment: A Definition and a Cyclical Tendency
The unemployment rate is not the share of the whole population without a job. Its denominator is the labor force: employed people plus unemployed people who meet the relevant statistical criteria for being without work and seeking or available for work. People outside the labor force are not counted as unemployed in this measure.
This identity is exact when the counts use consistent definitions and cover the same population and period. The cyclical pattern is different: unemployment typically rises in contractions and falls in expansions, so it is often described as countercyclical. That pattern is empirical, not part of the definition of unemployment. A change in the labor force can also affect the rate, even when employment changes little.
The identity also explains why the unemployment rate is not determined by the employment count alone. To interpret a change in \(u\), one needs both employment and the labor force. A person may enter or leave the labor force; that changes the denominator and may change the number counted as unemployed. The rate can therefore rise even if the change in employment is small.
Worked Example: Employment Falls While the Labor Force Grows
In a hypothetical region, the population used for labor-force statistics is 10.0 million. Initially, 7.60 million people are employed and the labor force is 8.00 million. Later, 7.50 million are employed and the labor force is 8.10 million. Find the unemployment rate at each date, the labor-force participation rate, and the employment-population ratio. Explain how the rates change.
Method. Use \(U=L-E\) to find unemployed people. Calculate the unemployment rate as \(U/L\), the participation rate as \(L/\text{population}\), and the employment-population ratio as \(E/\text{population}\). Keep the denominators distinct.
Initially, unemployed people number \(8.00-7.60=0.40\) million. The unemployment rate is \(0.40/8.00=0.05\), or 5.00%. The participation rate is \(8.00/10.0=0.80\), or 80.0%. The employment-population ratio is \(7.60/10.0=0.76\), or 76.0%.
Later, unemployed people number \(8.10-7.50=0.60\) million. The unemployment rate is \(0.60/8.10\approx0.074074\), or 7.41%. The participation rate is \(8.10/10.0=0.81\), or 81.0%. The employment-population ratio is \(7.50/10.0=0.75\), or 75.0%.
Employment falls by 0.10 million, while the labor force grows by 0.10 million. The unemployment rate rises from 5.00% to 7.41%; participation rises from 80.0% to 81.0%; and the employment-population ratio falls from 76.0% to 75.0%. The common mistake is to divide unemployed people by the population: the correct denominator for the unemployment rate is the labor force.
Check. At the later date, \(7.50+0.60=8.10\) million, as required by \(L=E+U\). The computed rates also use the stated denominators.
Inflation and the Price Level
Inflation is a rate of increase in a general price index over time. A consumer price index (CPI), for example, tracks the cost of a specified basket of consumer goods and services. Inflation over a period is commonly calculated as the percentage change in the index over that period. The index is a measure of the price level; the inflation rate describes how quickly that level is changing.
Inflation’s cyclical behavior is less uniform than the typical output-employment pattern. In some episodes, inflation rises as economic activity strengthens and eases as activity weakens. In other episodes, inflation may rise while output falls, or may continue to decline even as output begins to recover. The direction depends on the circumstances and the period examined. Thus, inflation should not be called invariably procyclical or countercyclical.
A particularly important distinction is between a lower inflation rate and a falling price level. If inflation moves from 4% to 2%, the price index is still rising when both rates are positive. A falling price level requires a negative inflation rate over the relevant period. Also, a CPI inflation rate and a GDP deflator inflation rate can differ because the indexes cover different prices.
Worked Example: Output Falls While Inflation Remains Positive
A hypothetical economy’s real-output index is 100, 98, and 97 in successive years. Its CPI is 250, 257.5, and 262.65. Calculate the annual output percentage changes and CPI inflation rates. State whether the price level falls, and describe the observed co-movement without treating it as a universal rule.
Method. Apply the percentage-change formula separately to the real-output index and the CPI. The output index measures real production; the CPI measures the price level, so their percentage changes answer different questions.
Output from year 1 to year 2 changes by \(98-100=-2\), giving \((-2/100)\times100=-2.00\%\). From year 2 to year 3 it changes by \(97-98=-1\), giving \((-1/98)\times100\approx-1.02\%\).
The CPI rises from 250 to 257.5, a change of 7.5. Inflation is \((7.5/250)\times100=3.00\%\). It then rises from 257.5 to 262.65, a change of 5.15. Inflation is \((5.15/257.5)\times100=2.00\%\).
Real output falls in both years, while inflation remains positive but slows from 3.00% to 2.00%. The CPI rises in both years, so the price level does not fall. These data show declining output alongside positive, easing inflation; they do not establish a general relationship between output and inflation. The common mistake is to infer that a lower inflation rate means prices have declined.
Check. The CPI values increase from 250 to 257.5 to 262.65, consistent with positive inflation in each interval. The second increase is a smaller percentage of its starting index, consistent with the lower inflation rate.
How to Read Stylized Facts Carefully
When comparing cyclical series, first identify what each series measures and whether it is a level, a growth rate, a rate such as unemployment, or a price-index change. Then establish the period and frequency: quarterly output growth should not be compared casually with an annual inflation rate or a monthly employment count. Seasonal adjustment can also matter because recurring calendar patterns may otherwise obscure changes associated with the business cycle.
Finally, use the appropriate strength of language. “Typically” and “tends to” describe stylized facts. “Must” and “by definition” should be reserved for identities or stated assumptions. In particular, \(L=E+U\) is a definitional relationship for consistently measured labor-force counts; the tendency for unemployment to rise in contractions is not. Likewise, the percentage change in a price index defines its inflation rate, but no identity requires inflation to fall whenever real output falls.
These distinctions are useful when summarizing economic data and when evaluating claims about a business cycle. A pattern may be common without being present in every episode; a definitional relationship should hold when its terms are measured consistently. This is aligned with the SOA VEE Economics (Validation by Educational Experience) syllabus.
Check Your Understanding
Work through each problem, paying attention to whether the statement concerns a definition or a typical pattern.
- A business-cycle summary says a series usually falls during expansions and rises during contractions. What term describes this pattern? (Answer: Countercyclical.)
- A hypothetical CPI rises from 180 to 189 in one year. Calculate inflation and state whether the price level rose or fell. (Answer: Inflation is 5.00%; the price level rose.)
- Real output falls from an index of 500 to 490, while employment falls from an index of 20.0 to 19.8. Calculate both percentage changes and state whether the observed movements are consistent with procyclical employment. (Answer: Output changes by \(-2.00\%\), employment by \(-1.00\%\); both fall, consistent with the usual procyclical pattern.)
- A labor force initially has 5.00 million people, of whom 4.80 million are employed. Later, 4.75 million are employed and the labor force is 5.10 million. Calculate the unemployment rate at both dates. (Answer: Initially \(0.20/5.00=4.00\%\); later \(0.35/5.10\approx6.86\%\).)
- A CPI rises from 300 to 306, then to 309.06, while real output falls by 1.5% and then by 0.5%. Calculate inflation in each interval and interpret the price-level movement. (Answer: Inflation is 2.00% in the first interval and 1.00% in the second; the price level rises in both intervals even though inflation slows.)