Depression: The Forgotten Part of the Business Cycle

The term “business cycle” refers to the cyclic nature of economic activity. The following definition comes from Investopedia:
What Does Business Cycle Mean?
The recurring and fluctuating levels of economic activity that an economy experiences over a long period of time. The five stages of the business are growth (expansion), peak, recession (contraction), trough and recovery. At one time, business cycles were thought to be extremely regular, with predictable durations, but today they are widely believed to be irregular, varying in frequency, magnitude and duration.   
Wikipedia provides a more extended discussion of the business cycle, which it summarizes as comprised of periods of expansion separated by periods of contraction.
John Smith, at Associated, gives the most common definition of the stages of the business cycle: expansion, contraction and recovery.
These definitions are very basic, but we need to think about these as we take an analytical look at the business cycle. One thing will become clear in the following discussion: the business cycle is composed of events and processes. Some descriptions of the business cycle confuse these two things (events and processes).
The Traditional Business Cycle
The concept of recessions is based on the idea of the business cycle. The classical diagram to summarize the business cycle often looks like the following graph. 

This is far from the shape of real business cycle because the expansions and contractions are often of much different durations in time and the trend through several business cycles almost always has a non-zero slope, often upwards. Most of the time, but not all, the second peak is at a higher level of economic activity that the preceding one.
An example of a business cycle is shown in the following graph for the recession of 1973-75, following the value of real GDP. The shaded gray area indicates the NBER (National Bureau of Economic Research) defined recession.
 As we proceed forward, it is important to distinguish between processes (which occur over a period of time) and events (which occur at a point in time). Processes are expansion and recession. Events are peaks and troughs. Throughout some discussions I find that events and processes are used in definitions without distinction regarding the time dependent factor. It is important in the discussion that follows to focus on the differences between processes and events.

The New Business Cycle
The business cycle could be much better defined if the recovery and expansion phases were identified as two individual steps. The recovery ends and expansion begins when the economic variable (in this discussion GDP) recovers to the level of the previous peak. In the introduction, only the definition given by John Smith clearly identifies recovery and expansion as two separate processes. They are linked by an event, the surpassing of the prior peak. The expansion does not begin until new highs are achieved; gains from the trough simply recover what had been produced in the previous expansion until the prior peak is equaled. The following table shows these relationships:
 In the above graph, the recovery would be complete in 3Q/75. The new expansion takes place from 4Q/75 to 1Q/80. The following graph shows the type of illustration that would give a better representation of the business cycle.

Using the new business cycle definitions enables a precise definition of recession and depression:
  • Recession is the economic contraction from peak to trough.
  • Depression is the cycle from an economic peak through a trough and back to the level of the preceding peak (recession plus recovery).
  • Thus every recession is part of a depression.
This is not an idea that exists only in my mind. In fact, it may underlie a lot of economic thinking. Very recently, for example, I saw statements in an article by Edward Harrison that referred to the current projected slow recovery as a depression. I do believe the concept can stand further clarification and, with that, add to better characterization of various business cycles.
The following table gives the peak, trough, recovery and expansion (to the next peak) dates associated with recessions since the late 1940s, using real GDP as the measure of economic activity.
GDP data is quarterly. The month designated in the above table is the middle month of the applicable quarter. Thus the only months listed for the business cycle dates are February, May, August and November.
The duration of official recessions (NBER designations), as well as the durations of business cycle recessions, recoveries, depressions and expansions as measured by real GDP are shown in the following table.
 Some observations:
  •  All NBER official recessions have been equal to or longer than the business cycle recessions (based on real GDP) except for the recession of 1990-91.
  • There have been four recoveries that took longer than the associated business cycle recessions.
  • There have been four recoveries that were quicker than the associated business cycle recessions.
  • The last two completed business cycles had recessions and recoveries of equal length.
  • All business cycle depressions have been of longer duration than official recessions, except for the 2001 recession.
  • The average duration of official recessions (NBER) has been 11.6 months (1948-1990), with a standard deviation of 3.5 months. The 2001 recession lasted 8 months. This data point belongs to same distribution as the previous recessions.
  • The average duration of business cycle depressions based on GDP has been 18.1 months (1948-90), with a standard deviation of 4.4 months. The 2001 depression lasted 6 months. This is a clearly an outlier from the previous depressions.

There has been a lot missed in understanding the cyclic behavior of the economy by not making precise definitions. A remedy has been attempted here. While this discussion has been trivial for academic macroeconomists (but I hope not for all of my readers), these trivialities must be clearly understood to appreciate what I will try to discuss in upcoming articles.

Related Article

Do Not Confuse Unemployment Rate with Employment  by Steven Hansen

8 replies on “Depression: The Forgotten Part of the Business Cycle”

  1. Here we are tracing the sometimes problematic metric of GDP (see the Sarkozy Report). The real question to my mind, at least, involves investment. Does investment not drive the business cycle?

  2. demand side – – –

    I agree with your thoughts about investment. I have started some work to look at the cyclical behavior of several important economic metrics. The cyclical behavior of of employment will be posted in a few days. I am revising some earlier work that I published at and Seeking Alpha on employment cycles. I expect to do articles on several other business cycle indicators and drivers in the coming weeks.

    Steve Hansen has written extensively over the past few years about why he doesn’t think that GDP is the best measure of economic activity. I expect he will have more to say about that going forward.

  3. “Does investment not drive the business cycle?”

    Opens up an interesting point. What, exactly, gets invested? Once you start down this rabbit hole, it’s clear that “initiative” is what gets invested, ala 1776, not to mention 1941-1945. Specifically, our greatest wealth is created by coordinated public initiative, which we post hoc denominate with created currency bookkeeping.

    Here’s a classic statement on that:

    “ECCLES: We created it.
    PATMAN: Out of what?
    ECCLES: Out of the right to issue credit money.
    PATMAN: And there is nothing behind it, is there, except our government’s credit?
    ECCLES: That is what our money system is.”
    – Federal Reserve Board Governor Marriner Eccles in testimony before the House Committee on Banking and Currency in 1941, during questioning by Congressman Wright Patman about how the Fed got the money to purchase two billion dollars worth of government bonds in 1933.
    [Makes you wonder why we bother with the charade of T-bonds, but that’s another story.]

    There’s a whole rabbit hole to explore here.
    business cycle
    currency investment drives cycle
    public projects create currency when/as needed
    public initiative creates & initiates coordinated projects
    public awareness triggers public initiative
    (note: whole host of necessary but not sufficient catalyst processes go into creating an aware & responsive population & nation state)
    so coordinated personal eventually initiative drives public awareness/initiative/investment/business cycles
    (yada, yada – so the $100 question is: how do we keep next decade’s aggregate – of population size “N+M” – within growth tolerance limits? If 1941-1945 & 1776 were any precedents, it seems we first need metaphorical 2x4s to even achieve simultaneous attention from that many people)

    Finally, we’re left with the circular logic that 2x4s, or more generally, context, drive business cycles? Translation: business cycles = damped oscillation of our cultural processes as we try to track unpredictable context paths. All described by an infinite polynomial with endless degrees of freedom. Consequence: any economist that tells you they know how the economy works is totally full of BS. Business cycles are not so different from the motorneuron & muscle activity cycles that surfers exhibit as they attempt to ride successive waves. We should be discussing this from the perspective of regional/national adaptive rates, not the extremely arbitrary & limiting concept of business cycles.

  4. Roger – – –

    Good essay. Econometric types (I am one of those nerds) try to discuss the economy in terms of parameters that have numerical values. That limits our effectiveness in describing the complex system you describe which would be better characterized by n-variant statistical interactions and feedbacks, where n approaches infinity.

    However, I still get satisfaction from finding small cells of correlative sense in the primordial statistical soup.

Comments are closed.