Some background on cycles in macroeconomics

One of the most interesting aspects of the history of economic thought is the "fads" and "fashions" that go with the ebb and flow of thinking about how the economy operates. If you look back to the early part of the last century it is clear that much of the effort of researchers working in the first part of the century on macroeconomics centred on understanding the cyclical nature of the macroeconomy. Perhaps the most famous contribution came from a distillation of research conducted by Arthur Burns and Wesley Mitchell in 1946 (see the NBER reference).

But although it's quite obvious that the business cycle ( - the cycle of boom and bust/positive growth and recession) is the most important cycle in the economy, many others have been hypothesized. Perhaps the most famous long cycle is that of
Kondratieff (who was executed during Stalin's purge) and his "long wave" of roughly 50 years in periodicity. But there were also others - among them names like Clement Juglar, Joseph Schumpeter, Joseph Kitchin and Simon Kuznets. There are also theories of credit cycles which largely originate from the Austrian school and also theories of financial instability (due to Hyman Minsky). The point though is that the business cycle is not the only cycle that exists in macroeconomics - it's just the cycle that hits the headlines most often because it has such drastic consequences.

For some reason (likely the Great Depression), the emphasis in economics shifted to economic policy and the role of government in stimulating the economy. In the post-WWII era, the emphasis shifted as events dictated - oil price shocks in the 1970s, inflation in the late 1970s and 1980s, etc. and never really returned to the idea of cycles in macroeconomics.

What's interesting here is that the fact that cycles in macroeconomics are not strictly "periodic" - in other words they don't have regular periodicity like a sine or cosine function, and don't have a regular amplitude (height of fluctuation) either, which makes them harder to detect. In addition changes in policy will likely change the nature of the cycles, distorting what we would otherwise observe. Also what's also interesting is that (as one might expect) there isn't necessarily just one cycle in economic growth - there is a combination of cycles all operating at different frequencies. So separating these out is no easy task.

Next I'll discuss why we should expect there to be cycles in macroeconomic data.