Real Business Cycles - Business Cycles

Business Cycles

If we were to take snapshots of an economy at different points in time, no two photos would look alike. This occurs for two reasons:

1. Many advanced economies exhibit sustained growth over time. That is, snapshots taken many years apart will most likely depict higher levels of economic activity in the later period.
2. There exist seemingly random fluctuations around this growth trend. Thus given two snapshots in time, predicting the latter with the earlier is nearly impossible.

A common way to observe such behavior is by looking at a time series of an economy’s output, more specifically gross national product (GNP). This is just the value of the goods and services produced by a country’s businesses and workers.

Figure 1 shows the time series of real GNP for the United States from 1954-2005. While we see continuous growth of output, it is not a steady increase. There are times of faster growth and times of slower growth. Figure 2 transforms these levels into growth rates of real GNP and extracts a smoother growth trend. A common method to obtain this trend is the Hodrick-Prescott filter. The basic idea is to find a balance between the extent to which general growth trend follows the cyclical movement (since long term growth rate is not likely to be perfectly constant) and how smooth it is. The HP filter identifies the longer term fluctuations as part of the growth trend while classifying the more jumpy fluctuations as part of the cyclical component.

Observe the difference between this growth component and the jerkier data. Economists refer to these cyclical movements about the trend as business cycles. Figure 3 explicitly captures such deviations. Note the horizontal axis at 0. A point on this line indicates at that year, there is no deviation from the trend. All other points above and below the line imply deviations. By using log real GNP the distance between any point and the 0 line roughly equals the percentage deviation from the long run growth trend.

We call large positive deviations (those above the 0 axis) peaks. We call relatively large negative deviations (those below the 0 axis) troughs. A series of positive deviations leading to peaks are booms and a series of negative deviations leading to troughs are recessions.

At a glance, the deviations just look like a string of waves bunched together—nothing about it appears consistent. To explain causes of such fluctuations may appear rather difficult given these irregularities. However, if we consider other macroeconomic variables, we will observe patterns in these irregularities. For example, consider Figure 4 which depicts fluctuations in output and consumption spending, i.e. what people buy and use at any given period. Observe how the peaks and troughs align at almost the same places and how the upturns and downturns coincide.

We might predict that other similar data may exhibit similar qualities. For example, (a) labor, hours worked (b) productivity, how effective firms use such capital or labor, (c) investment, amount of capital saved to help future endeavors, and (d) capital stock, value of machines, buildings and other equipment that help firms produce their goods. While Figure 5 shows a similar story for investment, the relationship with capital in Figure 6 departs from the story. We need a way to pin down a better story; one way is to look at some statistics.