Are the Government’s Most Important Items Of Economic Data All Deceptive In Fundamental Ways? – Part I (Accounting For Government Spending In GDP)


Many categories of government economic data are compiled and defined in ways to make bigger government appear more desirable. Can the data be regarded as objective, or instead part of the government’s advocacy for increase in its own size and power? This series will consider three categories of economic data that are the most important in terms of their use in political arguments over economic policy. Those categories are: (1) GDP, (2) the rate of poverty, and (3) the cost of obligations to retirees for pensions and health care.​

GDP is the principal measure of the overall size of the economy, and of whether the economy is growing or shrinking and by how much. ​Some assert that there is a fundamental problem with the compilation of GDP data, which is that government spending on goods, services and salaries is added into GDP dollar for dollar . This fact arguably makes the GDP data useless in a debate over whether government spending should be increased or cut. That is the principal use to which GDP data are generally put.

The portion of GDP that comes from the private economy is based on the rationale that voluntariness and markets mean that transactions of equal dollar value should be given equal weight. That rationale does not apply to government spending. An assumption that a dollar of government spending on goods, services and salaries counts at 100% as an addition to GDP only has some validity with a small government that thinks thrift is a virtue. ​When the government starts spending money to “create jobs” or “stimulate the economy,” do we have a fallacy?

When you count every dollar of government purchases at full value, then pure wasted spending counts to “increase” GDP just as much as the very most necessary expenditure. Suppose you pay someone to dig holes and fill them in (an example actually extensively discussed in Keynes’ General Theory). At the end of a year he has produced exactly nothing — but the GDP numbers add in an amount equal to whatever the government paid him. If the government pays every one of 300 million Americans $50,000 each to dig holes and fill them in all year, they have produced absolutely nothing by the end of the year, and the government records a GDP of $15 trillion. If the government doubles everyone’s salary to $100,000, then it just doubled the GDP to $30 trillion, even though they are all starving (because they were too busy digging holes to produce any food).

These are extreme examples, but we live in a time when the acceptance of the GDP numbers as measuring real changes in the economy is so great that the government can spend hundreds of billions in stimulus and count that as a dollar for dollar increase in GDP. The media offers virtually no push-back to this practice. The same argument holds for “green” energy spending. That kind of spending also is measured as increasing GDP dollar for dollar.​ Hire double the number of airport screeners as needed? Also a dollar for dollar GDP increase. The cost of the drug war? Same. And so forth.

It also works the same way when there is any proposal to cut spending. Under GDP accounting, cuts in government spending, no matter how wasteful that spending, are recorded as reducing GDP dollar for dollar. Thus here we have President Obama in February 2013 opposing the upcoming sequester spending cuts:​

“Our top priority must be to do everything we can to grow the economy and create good, middle-class jobs,” Obama said during remarks at the White House, standing alongside a group of emergency responders. “That’s why it’s so troubling that just 10 days from now, Congress might allow a series of automatic, severe budget cuts to take place that will do the exact opposite.”

Under this logic, you can’t ever cut any government spending, because that will “shrink” the economy.

[A longer version of this post originally appeared at on February 22, 2013.]