More On Counting Federal Spending As A Full-Value Addition To GDP
/My last post on Tuesday has inspired a spirited debate in the comments about how federal spending should properly be accounted for in GDP. What is the right answer? After reading the comments, it occurs to me that there are several more points to make.
For those criticizing or disagreeing with my post — led by prolific commenter Richard Greene — the main theme has been that many large categories of federal spending make an obvious positive contribution to the economy. Examples given include the Defense Department, teachers/education, and national parks. Surely excluding those kinds of things entirely from GDP accounting would provide at least as deceptive an indicator of the true size of the economy as including them at full cost value. And if those kinds of things, and many others, are not included at full cost value, what is the alternative? Some flat percentage discount could be applied, but there is no obvious constant level of discount that would be appropriate for all categories of spending; and reasonable people could disagree on varying levels of discount for different categories. Maybe defense should even be included at a premium!
My answer to this critique was at least suggested in the prior post, although perhaps not stated with sufficient clarity. From the prior post:
What [new Secretary of Commerce] Lutnick could do that would be a real service would be to change the Commerce Department’s communications with the public to emphasize the changes to GDP excluding government spending, and to de-emphasize the changes in government spending and/or the figures that include that spending.
To expand on that, I’m not proposing that the government should stop keeping track of the amount of federal spending on goods and services, or that it should completely stop publishing a figure for what GDP would be if government spending on those items were included at full cost value. But I do propose that the principal GDP figure reported on a regular basis should be the figure that excludes federal spending on goods and services. The total with federal spending excluded is the number that should be the headline and the emphasis of the quarterly press releases. This proposal applies particularly to statements about measuring changes in the overall size of the economy. The reason for the proposal is that the size of the private economy is far the better indicator of overall wealth generation, and of whether the economy is improving or deteriorating.
The most important flaw in the current system is that by far the biggest and most frequent use of the GDP figures by the public and the press, and by the government itself, is as a measure of the extent to which the economy is changing, growing or shrinking, and by how much, from period to period. Are we in a boom or a bust, a period of strong growth, or weak growth, or maybe a recession, or even a depression?
For these purposes, the inclusion of the federal spending in GDP at full cost value becomes much more misleading than it is useful. The reason is that changes to government spending can be easily manipulated; and under the current system, in any given month or quarter, or even for a full year, changes in government spending can be the dominant factor in reported changes in GDP. In an effort to show GDP growth, a government wanting to make its economic numbers look better can accomplish that by taking on hundreds of billions of dollars of debt and then wasting the money on its cronies. This is exactly what the Biden administration did with “Build Back Better” and the “Inflation Reduction Act.” Not so many years ago, it was the same story with the Obama “stimulus.” In the other direction, cutting government spending on even the most obvious and total waste — think of Lee Zeldin’s effort to stop Biden’s $20 billion giveaway to cronies in the IRA’s “greenhouse gas reduction fund” — can be portrayed by critics as a reduction in GDP, an indicator of impending recession, and a failure of Trump’s economic policies (or of the economic policies of any other President who makes an attempt to be fiscally responsible). Such statistics then get misused to put the government on a one-way ratchet of always increasing in size, while never making the people better off.
Even in today’s enormous economy approaching $30 trillion per year, when government spending is included in the measure of GDP, then any big changes in government spending will appear to be very significant in measured changes to GDP. Biden’s “Build Back Better” law of 2021 was said to involve some $2.2 trillion of spending. It is impossible to pin down exactly how much of that would be attributable to any given year, and probably the full amount would be over ten years. But even at $220 billion per year, that would be about 0.7% of GDP annually, which would be a very noticeable part of annual growth that has been averaging about 2-3% per year. The Inflation Reduction Act was separately reported to represent another $550 billion of increased spending, and much of that in the form of uncapped benefits for “green” energy projects that, if fully claimed, could easily exceed $1 trillion over the ten year measuring period, or $100+ billion per year. Put the two acts together, and you could have 1% or more per year of fake economic “growth.” (The same criticism could validly be made of inclusion in GDP accounting of Trump’s blow-out “Covid” spending in 2020.)
In the other direction, if President Trump’s expense reduction efforts succeed in cutting $500 billion of waste per year from federal spending (highly optimistic), that could be recorded as an annualized 1.7% of economic shrinkage, even if all of that spending were complete waste.
Meanwhile, don’t lose track of the fact that for every federal expenditure that unquestionably contributes in some positive way to real GDP, there is another that not only does not contribute positively, but is actually destructive and only serves to reduce real wealth and impoverish the people. At the top of my list of such things are salaries of the regulation-writers in EPA who concoct things like the regulation to force closure of all fossil-fuel-based power plants, to be replaced with wind turbines and solar panels; or their colleagues at the same agency who work to ban gasoline-powered vehicles; or the people in the Energy Department who have written the rules to ban washing machines and dishwashers that work; or the people in the Biden White House or Interior Department who have blocked pipelines or banned leases for drilling on federal lands; or the people in the Education Department who have worked to require schools to require that boys be allowed to participate in girls’ sports. I could come up with dozens more such examples. In GDP accounting, all of these activities have been accounted for by adding the salaries of the regulators to the GDP total. The much larger diminutions to the real economy caused by the regulators’ destructive actions just never enter into the books.
Finally, let me apply the above thoughts in a response to Mr. Greene’s comments on the effect of government spending reductions on the economy of Argentina. In his most recent comment on the previous post, Mr. Greene states:
Your daydream ignores the relevant experience in Argentina where the 10% government employee reduction caused a serious recession:
Argentina Real GDP
2024 -3.5% to -3.8%
Argentina Unemployment Rate
November 2023 5.7%
April 2024 7.6% (up 35% in 5 months)
Undoubtedly, what Mr. Greene reports is what the official statistics show, with the salaries of these government employees recorded as a full addition to GDP. But these statistics illustrate exactly what is my point. The people President Milei has fired are the very people who have been tying up Argentina’s economy with subsidies and red tape. The reduction in GDP of -3.5% to -3.8% represents the removal of their salaries and other government spending from the official accounts. But to what extent does it represent a real reduction in the size of the productive economy or the well-being of the people?
I would be very interested in finding a figure for how Argentina’s economy fared in 2024 if the change in government spending is excluded from the figures and only the private economy is considered. I am not able to find that. Maybe others can.
What I can find are many, many optimistic predictions of economic growth for 2025 for Argentina, now that the dead weight of excess bureaucracy and regulation has been lifted. Even the IMF (generally brain-dead Keynesians who always advocate for more government spending) is predicting 5% real growth. Others are predicting even more — up to 7%. I don’t know that that will happen. But from what I can find, there appears to be a consensus of 5% growth or more.
So I suggest that the -3.5% economic shrinkage figure cited by Mr. Greene for 2024, while it may be an accurate rendering of the official statistics, provides a highly misleading picture of the underlying economic reality. And the reason the picture is misleading is precisely the convention of including the government spending at full value in the measure of GDP.
I’m not saying that there has not been some economic pain experienced by the government workers who got laid off. But now their efforts can be put to productive use in the economy. The country will quickly be much better off, and before long, the laid off workers in their new jobs will also be better off.
So I wish Mr. Lutnick the best of luck in his effort to reform the reporting of economic statistics by the Commerce Department to better reflect underlying economic reality. No system will be perfect, but the changes I recommend will be greatly superior to the easily-manipulated measures that we have now.