In a comment on Reynolds and Cochran on the Slow Recovery, Marcus Nunes argued, “If there was a lesson in the R&R fracas is that you should take care with numbers, especially if you define “tipping points”. I think 15% is below what would account for the ‘core functions’ of government.” Nunes then referred the reader to ‘Keep it simple’. Nunes’s commentary is an interesting take on recent controversy over the coding error found in one of the Reinhart-Rogoff papers focusing on debt and growth.
‘Keep it simple’ is useful. It provides a summary of other evidence on the impact of the size of government on growth evidence. (John Taylor also focuses on size of government at Coding Errors, Austerity, and Exploding Debt.)
I agree whole heartedly with Nunes’s reflection on the R & R debate [For a summary of the details of this ‘debate’ see Mistakes by Greg Mankiw]. The focus should be on the source the size of government relative to the size of the economy not on the size of the debt or deficit, as Nunes succinctly point out:
I may be missing something vital, but what bothered me about the R&R ‘fall-out’ was that the original study was concerned with public debt/GDP levels. The major finding of the critics was that, contrary to the original study, no ‘tipping-point’ (after which growth is negatively affected) was found.
My take: Debt results from deficits. Deficits follow government spending (given revenues). So why not go to the ‘source’, i.e., government spending, and check if it has a measurable impact on growth.
While there may be a quibble on a ‘tipping point’, Nunes’s conclusion from the data provided is essentially consistent with my argument. His conclusion:
“So yes, “too much government can be bad for the economy´s health”!”
Dick’s comment reflects well my perspective on turning points:
It appears that Professor Cochran is closer with 10% than to consider 15%. Your averages tend to be on the high side because you are using more modern government spending levels. In a functioning economy there is little need for government because market driven production and processes are much more efficient with much higher quality.
Serious analysis would probably show that much less government is necessary to a properly function economy, even less than 10%.