# Economic Reasoning for Structural change vs Cyclical change

Often when there is an episode of low output growth, a debate takes place: Whether the downturn/recession is due to structural factors or is it cyclical.

I understand that the analysis is largely based on statistical estimation wherein it is checked whether there is decline in trend growth or it's the beginning of trough in business cycles.

How do I understand this in terms of pure economics? Broadly I think answer lies in understanding causes of business cycles. But without statistical estimation, how can one argue that a recession is cyclical or due to structural changes?

## 1 Answer

It is not possible ex-ante, it's all about:

1. economic interpretations, i.e., economic schools of thought
2. expectations on central banks' next moves,
3. political/ideological views,
4. the (hidden and/or unconscious) emerging willingnesses to induce self-fulfilling prophecies

... hence the debates that take place here and there.

Whether a recession is cyclical or due to structural changes can only be assessed ex-post, i.e. with data and (well driven) studies.

• That being said, I of course have personal views on that matter, not shared above. – keepAlive Nov 21 '19 at 13:36
• Okay so let's assume that recession is over and we have all the data. Could you now give an examples where one can argue that a recession was cyclical/structural. – Dayne Nov 22 '19 at 1:33
• @Dayne you could use the canonical econometric approach to Time Series Analysis : SARIMA, see, e.g., Kaiser and Maravall (2001). Also, if there is a structural change, it may be at the source of a new economic cycle periodicity. In such case, you would need more cycles to infer you were dealing with the first one. Be that as it may, sometime a chart is enough, see, e.g. this – keepAlive Nov 22 '19 at 14:14
• these seem to be econometrics approaches. I am looking for only economic reasoning. What I mean is how do I understand in terms of usual equations of macro/eco. For example, a change in production function would be a structural change. A change in market sentiment (which can shift IS curve to the left) might be considered as temporary and thus cyclical. – Dayne Nov 23 '19 at 4:44
• @Dayne what you are searching does not exist. Economics is not hard science, and subjective interpretation is irreducible. This is why econometrics have been "invented", see, e.g. this. Also, all theoretical formalisms of business cycle that currently exist can be found on wikipedia. Put differently, your question is at the edge of economic knowledge. – keepAlive Nov 23 '19 at 10:07