The Corner

The one and only.

More on Government Spending and GDP


Yesterday, I talked about why we should really take with a grain of salt — to say the least– the claim that the disappointing GDP numbers were evidence that sequester cuts shouldn’t be implemented. For instance, as Dan Mitchell of the Cato Institute summarizes, ”According to Politico, ‘the fall was largely due to a drop in government spending.’ Bloomberg specifically cited a ‘plunge in defense spending’ and the Associated Press warned that ‘sharp government spending cuts’ are the economy’s biggest threat in 2013.” 

Among other things, I mentioned the work of Garett Jones about what is measured and what isn’t measured in GDP. Today, he has another piece digging further into this topic. He writes:

The boss hires his worthless unemployed nephew for a summer job.  Does that raise GDP or lower it? To keep things simple think of this as ex nihilo hiring: if not for this hire, the person wouldn’t have been hired otherwise and if the boss doesn’t hire the worthless nephew nobody else is going to get the job.  As we saw last week, the answer to whether the worthless hire boosts GDP turns on whether the boss works for the government or for the private sector.  Let’s take a new look at why.  We usually measure GDP as total spending (by consumers, businesses, governments, foreigners) but it might be clearer if we remember that GDP is also total income:Every dollar spent goes to somebody either as wages or as business profits.  Omitting details (e.g., “wages” includes salaries and bonuses, “profits” are often paid out as interest to the bank), we can sum it up this way: 

GDP = Wages + Profits

But here’s the thing: Government doesn’t make a profit.  

That’s not a punchline: That’s the reason government hiring of the unemployed boosts GDP by definition while private hiring of the unemployed does not.  When the private sector boss hires his worthless unemployed nephew, the nation’s total wage bill rises by, say, $100K, but the nation’s total profits also fall by the same $100K: The firm’s decision to hire the worthless worker is the firm’s decision to take an equal cut in profits.  But when the government hires the same worthless unemployed nephew, the nation’s total wage bill rises by $100K and then…that’s it!  Government doesn’t report profits–at least in the GDP figures–so government doesn’t record any – to balance out the +.  A GAO audit might report wasteful hiring but it won’t show up in GDP. 

On a somewhat related issue, Dan Mitchell writes

But here’s the problem. GDP numbers only measure how we spend or allocate our national income. It’s a very convoluted way of measuring economic health. Sort of like assessing the status of your household finances by adding together how much you spend on everything from mortgage and groceries to your cable bill and your tab at the local pub.

And he asks:

Wouldn’t it make much more sense to directly measure income? Isn’t the amount of money going into our bank accounts the key variable?

The same principle is true – or should be true – for a country.

That’s why the better variable is gross domestic income (GDI). It measures things such as employee compensation, corporate profits, and small business income.

Over at Modeled Behavior, Karl Smith couldn’t agree more. In fact, he argues that for “the sake of the Republic” we should stop using GDP and switch to GDI. 

Here is Scott Summer over at Money Illusion:

The government measures GDP in two different ways; GDP and GDI.  The Gross Domestic Income measure is generally regarded as the most accurate, but unfortunately the complete data comes out one month after the flash estimate of GDP.  Hence the press tends to report the GDP numbers.

Fortunately, the BEA does report most of the GDI data at the same time as GDP.  Today’s report shows the reported part of NGDI rising from $13,430.2 bill in Q3 to $13,569.5 billion in Q4.  About 4.1% at an annual rate.  The same 4% track we’ve been stuck in since mid-2009.  When the missing data is reported (interest and corporate earnings) the number will be revised, but is still likely to be much higher than the 0.5% reported growth in NGDP during Q4.  That number made no sense in light of the steady job growth in Q3 and Q4.

That seems like the number we should focus on.

Update: I really wasn’t clear in this post. My apologies. I was making (poorly) two separate points. First, that there are problems with the way GDP is measured (the Jones argument). A second was that government estimates of GDP and GDI do differ and it seems that GDI is a more accurate predictor of economic growth. Also, it gives a more upbeat picture of Q4. And yes, reported GDI is far from perfect and fails to address Jones’ point. Also, as I have said before, I have no doubt that when sequestration is implemented there will be some short term impact on growth, no matter how we measure it. 


Sign up for free NRO e-mails today:

Subscribe to National Review