So Krugman has posted his latest proof that he’s smarter than anyone who disagrees with him.
His assertion is that austerity is finally proven harmful and shows us this graph as proof:
What is the definition of GDP? Y = C + I + G + (X − M)
So government expenditures are by definition a part of GDP. So if G decreases (increases), if nothing else changes, then Y must decrease (increase). Obviously this is causal, not a spurious correlation that Krugman has challenged us all to argue in favor of.
That’s not what those who disagree with Krugman are saying. At least not me. Better questions than whether definitions hold true in the broadest sense might include:
- What are the unintended consequences of increased government spending? Most importantly, if the money to finance this spending is borrowed, will future growth be slowed? (a lot of research points to definitely YES, at least as debt levels increase). And by how much? Is it worthwhile to push the pain into the future?
- How much increase in GDP does the spending get us? Krugman is careful to point out that since he has graphed percentages, “you can’t read the slope of a trend line as a multiplier.” If government spends more, but GDP increases by less than the amount spent, you might argue that the government spending is somehow harming the economy. Or, you might argue that it just shows how much the government spending is needed. Anyway, people talk about the multiplier enough that he mentioned it, so it’s not irrelevant. Of course, the reason he mentioned it was because it appears to be less than one, if you tried to use the numbers on the axes that way.
- Why does a slowdown in GDP growth, or outright GDP contraction, demand government action? If there is an imbalance in the economy that causes negative changes in GDP, why does there have to be a government reaction? Is it impossible for the economy to self correct? It seems to me that the government should assist those who are harmed by economic changes, but why is some specific GDP print even a goal?
ASIDE: Krugman states in his post that this chart is based on 33 European countries, 2010 – 2013, annual data, from Eurostat. I am not able to replicate this result.
First, I only get data for 28 countries from Eurostat for the change in real GDP. Second, his graph does not show the Greece 2011 GDP change, which was -8.9%. So I’m not sure I have the correct data for GDP.
Third, the “government purchases” that he refers to as his X axis is not referenced by that name on Eurostat. There are two possibilities that I see data for:
Total general government expenditure% of GDPTotal general government expenditure is defined in ESA2010 8.100 and chapter 20 by reference to a list of categories: intermediate consumption, gross capital formation, compensation of employees, other taxes on production, subsidies, payable property income, current taxes on income, wealth, etc., social benefits other than social transfers in kind, social transfers in kind – purchased market production, other current transfers, adjustments for the change in pension entitlements, capital transfers and acquisitions less disposals of non-financial non-produced assets.
Government final consumption expenditure, current pricesMillion EUR – seasonally adjustedGovernment final consumption expenditure (ESA 2010, 3.98) includes two categories of expenditures: the value of goods and services produced by general government itself other than own-account capital formation, and purchases by general government of goods and services produced by market producers that are supplied to households – without any transformation – as social transfers in kind. Values are seasonally adjusted (SA). The ESA 2010 (European System of Accounts) regulation may be referred to for more specific explanations on methodology.