The "business cycle" is one of the central issues in macroeconomic theory and provides the starting point for understanding the complex relationships between the various measures of macroeconomic performance and the role of government economic policy.
The overall goal of government economic policy is to promote economic stability. By economic stability we mean an unemployment rate at or near the natural rate, price stability with a low inflation rate, and steady growth in economic output. But policy is not perfect and the economy is constantly subjected to unexpected events. What we typically observe is an economy that fluctuates around these goals. Sometimes the economy is overheated with too much demand and price inflation, other times the economy is in the doldrums with low or negative economic growth and high unemployment.
The short-term fluctuations in economic activity we see are called business cycles. Business cycles are recurring patterns of economic expansion (increasing economic growth and price inflation), then contraction (declining economic growth and growing unemployment), then expansion again. These two phases are punctuated by a peak at the end of an expansion when a contraction starts and a trough at the end of a contraction when an expansion begins again.
Figure 7-1. Business Cycles
|Business Cycles - recurring patterns of economic expansion, then contraction, then expansion again. Business cycles are characterized by:
Peak - point at which an expansionary phase ends and a contractionary phase begins.
Contractionary Phase - a period in which real GDP is declining. Also associated with declining inflation rates and increasing unemployment rates.
Trough - point at which a contractionary phase ends and an expansionary phase begins.
Expansionary Phase - a period in which real GDP is growing. Also associated with increasing inflation rates and declining unemployment rates.
The contractionary phase of the business cycle is often described as a recession. There is no firm definition of what constitutes a recession, but it is generally described as a significant decline in economic activity spread across the economy and lasting more than a few months. A recession is often casually defined as at least two consecutive quarters of negative economic growth (i.e., real output of the economy is declining) but this is not necessarily the case. For example, a small decline in real GDP not matched by a corresponding decline in employment would not be considered a recession. A severe recession in both scale and duration, such as in the 1930s, is called a depression.
The Business Cycle Dating Committee of the National Bureau of Economic Research (NBER) at some point in time became the organization that declares when a recession started and ended. Although GDP is probably the best measure of economy-wide output it is measured only quarterly. The Committee also uses monthly indicators such as total non-farm employment, real personal income, aggregate hours of work, real manufacturing and trade sales data, and industrial production indexes.
|Table 7-1. Business Cycle Reference Dates and Durations|
|Reference Dates||Duration in Months|
|December 1854||June 1857||--||30|
|December 1858||October 1860||18||22|
|June 1861||April 1865||8||46|
|December 1867||June 1869||32||18|
|December 1870||October 1873||18||34|
|March 1879||March 1882||65||36|
|May 1885||March 1887||38||22|
|April 1888||July 1890||13||27|
|May 1891||January 1893||10||20|
|June 1894||December 1895||17||18|
|June 1897||June 1899||18||24|
|December 1900||September 1902||18||21|
|August 1904||May 1907||23||33|
|June 1908||January 1910||13||19|
|January 1912||January 1913||24||12|
|December 1914||August 1918||23||44|
|March 1919||January 1920||7||10|
|July 1921||May 1923||18||22|
|July 1924||October 1926||14||27|
|November 1927||August 1929||13||21|
|March 1933||May 1937||43||50|
|June 1938||February 1945||13||70|
|October 1945||November 1948||8||37|
|October 1949||July 1953||11||45|
|May 1954||August 1957||10||39|
|April 1958||April 1960||8||24|
|February 1961||December 1969||10||106|
|November 1970||November 1973||11||36|
|March 1975||January 1980||16||58|
|July 1980||July 1981||6||12|
|November 1982||July 1990||16||92|
|March 1991||March 2001||8||120|
|Average, all cycles:|
|1854-1991 (31 cycles)||17||38|
|1854-1919 (16 cycles)||22||27|
|1919-1945 (6 cycles)||18||35|
|1945-2001 (10 cycles)||10||57|
|Source: National Bureau of Economic Research (http://www.nber.org)|
Business cycles are often called "regular" not because they occur with predictable frequency (each cycle differs in both length and intensity) but because the inter-relationships between macroeconomic variables are quite consistent. For example, when the GDP growth rate increases the unemployment rate declines. When the growth rate of GDP is higher than the long-term trend, the rate of inflation increases. Interest rates, exchange rates, bankruptcies, industrial production, and other measures of macroeconomic performance all seem to follow cycles generally consistent with the overall business cycle.
A primary measure of the health and welfare of an economy is the growth rate of real GDP, or total physical output of the economy. Because of a steadily growing population, the accumulation of real capital (investment), new technology that contributes to increases in productivity, and other factors such as education we expect growth in real GDP over the long run. The production possibilities frontier, more commonly called full-employment output or potential GDP, is steadily moving outwards. For example, real GDP in the United States has grown by an average 3.4 percent per year (0.80 percent per quarter) over the last 50 years. Over the short term real GDP growth rates are observed to cycle around this long-term trend. The economy is sometimes growing faster and sometimes slower.
Figure 7-2. Quarterly Changes in Real GDP and Recessions
The components of GDP (consumption, investment, government spending, exports, and imports) can behave very differently during recessions. In Table 7-2 the change in real GDP during recessions of the last 50 years is broken out by its components. We can make some general observations:
|Table 7-2. Change in Real GDP and its Components During Recessions|
(annualized seasonally adjusted billions of chained 1996 dollars)
|1948Q4 - 1949Q4||- 24.9||+ 33.1||- 52.5||+ 20.6||- 3.1|
|1953Q3 - 1954Q2||- 51.2||+ 8.8||- 25.5||- 49.8||+ 7.9|
|1957Q3 - 1958Q1||- 81.5||- 17.6||- 38.9||+ 4.1||- 11.8|
|1960Q1 - 1960Q4||- 38.1||+ 14.3||- 65.8||+ 24.3||+ 13.3|
|1969Q3 - 1970Q4||- 21.8||+ 55.9||- 56.2||- 24.2||+ 9.2|
|1973Q4 - 1975Q1||- 141.1||- 26.8||- 164.5||+ 32.4||+ 40.6|
|1980Q1 - 1980Q2||- 101.1||- 73.2||- 64.3||+ 3.1||+ 31.8|
|1981Q1 - 1982Q1||- 118.2||+ 11.2||- 91.5||+ 6.1||- 13.0|
|1990Q2 - 1991Q1||- 100.3||- 41.3||- 117.9||+ 20.0||+ 43.5|
|2000Q4 - 2001Q3||- 57.4||+ 82.1||- 192.5||+ 39.9||- 0.5|
|Note: Component changes may not add up to total GDP change due to the "residual" in GDP accounts.|
Source: Bureau of Economic Analysis (http://www.bea.doc.gov)
During business cycle contractions the unemployment rate rises and during expansions the unemployment rate falls. The low point in the unemployment rate usually occurs just before the peak. The high point usually occurs just after the trough. It appears that the increase in the unemployment rate is usually faster than the decline. In other words, the unemployment rate may surge upwards to a peak and then slowly fall back. This may be because hiring is more costly and time-consuming than firing, or that firms are reluctant to let go of staff until and then do so in a rush.
One interesting characteristic of the unemployment cycle is the change in the duration of unemployment. The Bureau of Labor Statistics categorizes how long people have been unemployed for: less than 5 weeks, 5 to 14 weeks, or longer than 14 weeks. Figure 7-3 shows that during recessions the long-term unemployment (15 weeks or more) share increases dramatically while the share of the total unemployed who have been out of work less than 5 weeks declines. During recessions there are more unemployed and it takes much longer to find job. The message is that if you are graduating from college or want to change companies or professions, don't try to do it when the economy is in a recession.
|Figure 7-3. Duration of unemployment.
Data Source: U.S. Dept. of Labor, Bureau of Labor Statistics (http://www.bls.gov/)
During economic contractions, when output is falling, the inflation rate also declines. During recoveries, when the economy nears the peak of the business cycle, the rate of inflation increases. The inflation cycle does not perfectly match the business cycle. While inflation generally declines during contractions, the decline does not stop when the trough is reached and recovery begins. Inflation continues to fall during the early stages of the recovery.
There can be cycles in the inflation rate independent of the business cycle. For example, a sudden and brief increase in oil prices can cause a temporary surge in the inflation rate that is not associated with the business cycle. Increases in the inflation rate in some months have been attributed to such cost or supply-side factors as energy, housing, and food price increases. Generally, cycles in inflation that are related to the business cycle are demand-driven while other movements in the inflation rate unrelated to the business cycle are often initiated by supply-side shocks. Prices and inflation also respond to changes in money supply and interest rates that do not translate to changes in the trend of economic output. A combination of events can also lead to rising inflation during recessions as occurred in 1974 with the Arab oil embargo.
We mentioned above that Business cycles are often called "regular" because the inter-relationships between macroeconomic variables are quite consistent. Some of the relationships were described in general terms but we can be much more specific. The inverse relationship between output and unemployment is measured by Okun's Law. The inverse relationship between inflation and unemployment is illustrated using the Phillips Curve.
The unemployment rate is usually inversely related to the growth rate of real GDP. When the economy is at the peak of the business cycle the economy is growing faster than normal and the unemployment rate declines. When the economy is near the trough economic growth is slow and the unemployment rate rises.
A second characteristic of the relationship between the growth rate of real GDP and unemployment is also important. Output fluctuates more than unemployment during the business cycle.
We mentioned in the previous chapter on the labor market that the unemployment rate is an imperfect measure of the under utilization of the labor resource. There is the potential for underemployment. An individual may be counted as working, but may not be working as many hours as he or she would like to. During recessions businesses do not necessarily reduce the number of employees. One reason for not reducing the quantity of labor when production declines is because of the hiring and retraining costs when the economy recovers. Firms try to retain their investment in human capital. As the economy weakens, firms are as likely to reduce the number of hours worked as they are the number of workers. Thus employment generally falls less rapidly than output. As the economy moves into a recession, the productivity of labor consequently declines. Similarly, during the early stages of recovery from a recession, businesses can also increase their output before they need to hire more workers.
As a result, output fluctuates more than unemployment during the business cycle. The historical trend in the U.S. has been that when real GDP is growing at 3.4 percent per year, the unemployment rate is stable at the natural rate of unemployment. If the growth rate in real GDP declines by 2 percent (from 3.4 percent in one year to 1.4 percent the next year), the unemployment rate is expected to increase by 1 percent. For every 2.0% change in the growth rate of real GDP, the unemployment rate moves about 1.0% in the opposite direction. This tendency for output to fluctuate more strongly than unemployment is known as Okun's Law. (The late Arthur Okun served as chairman of President Lyndon Johnson's Council of Economic Advisors.)
|Okun's Law - The unemployment rate is negatively related to changes in the growth rate of real GDP. Moreover, output fluctuates more than unemployment over the business cycle.|
Figure 7-4. Okun's Law
Okun's law can be represented by the percentage change in output as a function of the percent change in the unemployment rate:
|where,||Yi = output (e.g., real GDP) in year i
Ui = unemployment rate in year i (percent)
The left-hand side of the equation, [(Y2 - Y1) / Y1] * 100, represents the percentage change in real GDP between year 1 and year 2. The is the standard way of calculating the relative percent change between two years. For example, real GDP increased from 8,856 in 1999 to 9,224 in 2000, a 4.2% increase.
The right-hand side of the equation (U2 - U1) represents the absolute percent change in the unemployment rate. For example, if the unemployment rate increases from 4% in year 1 to 5% in year 2, there was a 1% increase in the unemployment rate.
There is also a constant term on the right hand side of the equation. The 3.4 represents the long-term trend in the GDP growth rate. For example, if the unemployment rate in year 1 is 4% but does not change (i.e., U1 = U2), then the associated increase in real GDP is 3.4 % (because U2 - U1 = 0). This long-term trend in the real GDP growth rate can be attributed to many factors such as population increases and steady improvements in technology and productivity. Consequently, this value may change over time. The 3.4 percent long-term growth rate may have been true for the baby boom years of the late 1960s through early 1980s and the technology boom of the 1990s, but may not hold for 21st century. It may be smaller because of the slowing growth rate of the labor force or larger if advances in technology and worker productivity continue on an accelerating pace.
Now let's say the unemployment rate increases from 4% to 5% in year 2. We have a 1% increase in the unemployment rate. The associated change in the growth rate of real GDP from year 1 to year 2 is now lower at 1.4% (where 3.4 - 2 * (5 - 4) = 3.4 - 2 = 1.4). In other words, when the unemployment rate increases from 4% to 5%, the growth rate in real GDP declines by 2% (from 3.4% to 1.4%).
Again, the implication is that changes in the growth rate of real GDP are larger than changes in the unemployment rate. As the economy moves into a recession, rather than lay off workers firms have an incentive to keep their staff even as production declines. Keeping underemployed workers can be cheaper than having to pay the costs of hiring and training new workers when the economy recovers. So, as the GDP growth rate declines by 2%, the unemployment rate increases by only 1%. Conversely, as the GDP growth rate increases by 2%, the unemployment rate declines by only 1%.
During economic contractions, when output is falling and unemployment rising, the inflation rate declines. During recoveries, when the economy nears the peak of the business cycle and unemployment is low, the rate of inflation generally increases. The Phillips curve provides a graphical picture of the inverse, or negative, relationship between the inflation rate and the unemployment rate.
The Phillips curve originated in a 1958 article by A.W. Phillips ("The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861-1957," Economica, November 1958) in which he observed the inverse relationship between unemployment rates and changes in changes in wage rates in the United Kingdom over a 97 year stretch. A similar stable relationship appeared to hold in the United States. Because the prices of goods and services are directly related to wage rates, economists also used Phillips curve concept to relate general price inflation (as opposed to wage inflation) to unemployment rates (Figure 7-5).
Figure 7-5. Phillips Curve, 1950s and 1960s
Politicians adopted the Phillips curve as their foundation for debate on economic policy. The Phillips curve presented a menu of policy trade-offs. Should an economy operate at low unemployment at the cost of high inflation or is low inflation at the cost of high unemployment preferred? Democrats held true to their stereotype of a party of labor by supporting lower unemployment rates at the cost of slightly higher inflation while Republicans reinforced their image as the pro-business part by advocating lower inflation rates.
Edmund Phelps and Milton Friedman surprised economists when they published articles in 1967 and 1968, respectively, in which they argued that a stable Phillips curve was a fantasy. Phelps ("Phillips Curves, Expectations of Inflation and Optimal Unemployment over Time," Economica, 1967) and Friedman ("The Role of Monetary Policy," American Economic Review, March 1968) argued that the Phillips curve was a simple empirical relationship that had no foundation in economic theory. In other words, the Phillips curve may have provided a reasonable representation of the economy in the past but that had no bearing on the future. The key theoretical advances that Phelps and Friedman made were that changes in money supply and peoples' expectations can lead to any possible combination of unemployment and inflation.
The 1970s quickly proved Phelps and Friedman to be right. High unemployment coexisted with high inflation. President Jimmy Carter called this the "misery index" (the sum of the unemployment rate and the inflation rate) in his successful 1976 campaign against President Gerald Ford.
Figure 7-6. Phillips Curve in the 1970s
The Phelps/Friedman theory of expectations and the subsequent explosion of the Phillips curve led to a revolution in the development of macroeconomic theory as the traditional foundations appeared to disintegrate. The following chapters in this course will discuss the foundations for macroeconomic theory and investigate how the expectations revolution has changed the face of macroeconomics.
The profession of many economists is forecasting business cycles. Forecasting involves identifying how different measures of economic performance change in relationship to each other. Some measures move in the same direction at the same time. Others move in opposite directions. A change in one measure may precede a change in another measure. These relationships can be used to project (with uncertainty) what will happen one month or one year from now.
The behavior of economic variables over the business cycle are evaluated on the basis of two characteristics:
There are other important behavioral characteristics that should also be considered such as volatility (how big are the expansions and contractions) but we will not discuss these.
Output and inflation are often termed pro-cyclical because both of these measures of economic performance are increasing as the economy is in the expansionary phase and decline during the contractionary phase. The unemployment rate is usually called counter-cyclical since it declines during the expansionary phase and increases during the contractionary phase. Many other measures of aggregate economic performance such as interest rates, housing starts, real wages, inventories, and others are usually described as pro- or counter-cyclical because they also follow cycles that relate directly to the business cycle.
|Pro-cyclical - measures of economic activity that increase when the economy is expanding a fall when the economy is contracting. Examples of pro-cyclical economic variables are real GDP, interest rates, and inflation.
Counter-cyclical - measures of economic activity that decline when the economy is expanding a increase when the economy is contracting. Examples of counter-cyclical economic variables are the unemployment rate and unemployment insurance claims.
Some measures of economic performance are better than others at indicating changes in the business cycle. For example, the Bureau of Economic Analysis regularly surveys firms for the value of new orders for capital goods they have received. The total value of new orders is a good indicator of what production of new capital goods will be in the near future. When there is an increase in the total value of new orders we might expect the growth rate of real GDP in future months to increase. This is what is called a leading indicator. Good news now means predictable good news in the future. Bad news now means bad news later.
|Economic Indicators - key economic statistics that provide information about business cycles and trends in overall economic performance.
Leading Indicator - an economic indicator that changes before the economy has changed. Examples include new orders for capital goods, building permits, unemployment insurance claims, and stock prices.
Lagging Indicator - an economic indicator that changes after the overall economy has changed. Examples include investment spending, the unemployment rate, and interest rates.
Some examples of leading indicators are average weekly hours worked, average initial claims for unemployment insurance, new orders for consumer and capital goods, building permits, stock prices, money supply, and others. Surveys of attitudes are also used as leading indicators. For example, surveys of consumer confidence are regularly reported and business executives are asked what their outlook for the economy is every quarter. Selected leading indicators can also be combined to provide a useful summary statistic called the composite or combined index of leading indicators. One popular composite leading indicator is published monthly by The Conference Board (http://www.tcb-indicators.org/).
The cycles of some measures of economic performance fall slightly behind the business cycle. These measures are called lagging indicators. For example, the high point in the unemployment rate may actually come several months after the bottom point in the business cycle. This happens because after the economy bottoms out, firms are able to increase production without hiring new labor. Moreover, firms may wait a short period in order to develop confidence there is an economic recovery before reversing their staffing plans.
Lagging indicators are useful in forecasting because they add confidence. When several leading indicators start providing positive news we may cautiously predict economic recovery. But that may not be good enough to commit millions of dollars to new capital investment. When the lagging indicators also turn around then we might more confidently make that business decision.
File last modified: May 1, 2003
© Tancred Lidderdale (Tancred@Lidderdale.com)