Monday, February 4, 2013

Government Purchases Versus Government Spending

It is interesting that PGL has posted a figure showing government purchases and how they have been gradually declining recently.  The recent report of GDP surprisingly declining in fourth quarter 2012 and the role of government spending in that brings this up.  What may be a problem here is that the public reports did not make this distinction, reporting on declines in "government spending," when in fact what is involved here has been a decline in government purchases.  The crucial issue here is that changes in transfer payments do not directly affect GDP, but changes in government purchases do, and it was purchases, mostly through defense spending, that declined.

Some readers may say, "What is the problem here?"  The problem is that a new right wing meme has erupted over whether or not "government spending" has declined.  Indeed, it did not do so during the 4th quarter 2012, at least not federal government spending.  Much as I dislike this source, one can see the numbers taken from US Treasury stats at .

Furthermore, Barron's has gotten all huffy about how the Bureau of Economic Analysis reported the decline in defense spending, which indeed was mostly a decline in purchases leading to a decline in GDP.  They complain that the 22% reported number for the quarterly decline was an annualized number, and that the actual decline was only 5.32%.  This is accurate, but indeed, that decline in purchases by the DOD happened and was unexpected (although following a sharp increase in the 3rd quarter, thought by many to be a forwarding of spending in anticipation of the fiscal cliff/sequester/etc.) and played a major role in the reported GDP quarterly decline.

In any case, the increase in federal spending noted by Breitbart was due to an increase in transfer payments that did not lead to an increase in GDP.  As it is, people should be warned that this meme is now out there.  I became aware of it when I saw the editorial page lead editorial in my very conservative hometown Harrisonburg (VA) Daily News-Record, which was going on about the government lying and Obama lying and so on and so forth, all on the basis of this report that in fact government spending had gone up rather than gone down as the media so widely reported.  It would have been helpful if the original reports on this had been a bit clearer, although I suspect that much of the public is simply unaware of the distinction here or its importance.

Meanwhile, everyone should keep in mind that the much bigger story here has been the ongoing decline in state and local government spending and purchases, with a far larger proportion of their spending being in fact purchases.  Their employment is also much higher than that of the federal government, just over 19 million in comparison with about 4.4 million for the federal government, the latter peaking in recent decades at about 5.3 million in 1987 and reaching a recent minimum in 2000 of about 4.1 million.  Since peaking in August 2008, government employment has fallen more than 700,000, with the overwhelming majority of that being in the state and local sector, with that decline still occurring in the last job report, where the gain of 157,000 jobs was after a 9000 jobs loss in the government sector, again, mostly state and local.  Federal employment has not changed much in this whole period, increasing some in 2009 and slightly decreasing since, although that could change if there continue to be substantial declines of federal government purchases.  

Barkley Rosser


ProGrowthLiberal said...

In an ideal world, one would report the entire fiscal impact of changes in fiscal policy (purchases, transfer payments, and taxes). One problem is that a lot of the changes in transfer payments and taxes are not changes in policy but automatic stabilizers (during recessions spending rises and taxes fall and during booms the converse occurs). Of course, the other issue (which you allude to) is that even policy changes in net taxes often have a less than one for one impact on consumption demand. In fact, Barro Ricardian equivalence types might argue that there is no impact on consumption (albeit this indeed is valid at best for only certain households).

ProGrowthLiberal said...

Nolte uses quarterly observations as to Federal outlays to claim there was some sort of fiscal stimulus but as I look at the actual Treasury figures, I have to wonder if there is some sort of seasonality to transfer payments going on. Not that I've done the hard work to demonstrate this but he hasn't looked into this possibility either.

Anonymous said...

Hmm. But Pelosi, Reid, Obama et al told us those transfer payments DO grow GDP. Remember that propensity-to-consume/fat multiplier at the ZLB thingy? Remember that food stamps have a multiplier of 1.8-ish? Are you saying at present, only DIRECT govt purchases have a multiplier of greater than unity? Tread carefully thinking through this, lest you get your intellectual underpants all wadded-up.

ProGrowthLiberal said...

Anthony - no one (outside of the most ardent Barro-Ricardian types) said changes in transfer payments to borrowing constrained households won't affect consumption - an indirect effect. Barkley's comment was that there was no direct impact. What I'm saying is that capturing fiscal impact from policy changes requires one to segment out the impact of changes in income on actual expenditures (something else not done in the rightwing oped that Barkley critiqued).

Anonymous said...

Ok, fair enough. But the taxpayer finances government **outlays** (now, or in the future). Government outlays have grown; this is no meme here, but rather an accounting identity. This administration cannot have its cake and eat it: either outlays in the form of transfer payment ARE stimulative, or they are not stimulative. If they ARE stimulative, then there is NO austerity. (As if austerity has a non-partisan defn, which it does not.)

ProGrowthLiberal said...

Anthony - try reading E. Cary Brown's 1954 American Economic Review paper on how changes in the deficit are not always the same as fiscal impact. It is a must for anyone who wants to understand macroeconomics. Simply put - taxes can fall and transfer payments can rise without any change in fiscal policy as they also depend on the state of the economy. I tried to make that point earlier but I guess you missed it.

Don Levit said...

"Taxes can fall and transfer payments can rise without any change in fiscal policy," writes Pro Growth Liberal. If revenue falls and expenses rise, then the deficit increases. Does the increased deficit mean the fiscal policy is the same? If so, what policy has changed to indicate the increased deficit?
"The increase in federal spending was due to an increase in transfer payments that did not lead to an increase in GDP," writes Barkley.
Can you explain your rationale, in that I understand government spending is an integral part of GDP.
Don Levit

Anonymous said...

I perused the paper. Key conclusion: "The primary failure of fiscal policy to be expansive in this period is attributable to the sharp increase in tax structures enacted at all levels of government." Whoa. Well then, we're already off to a very bad start for 2013--lot's of new taxes of all flavors. But importantly, transfers are negative taxes. Brown is concluding that negative taxes are expansionary. Pls expound on your take-aways from this work, in a little more detail. Thx.

ProGrowthLiberal said...

"Can you explain your rationale, in that I understand government spending is an integral part of GDP." Don - take a mini course in national income accounting. Consumption + investment + net exports + goverhment purchases. Which does not include transfer payments.