Is Some Honesty About To Come To Government Economic Statistics?

A recurring theme at this blog over the years has been the rank dishonesty of many of our government’s economic statistics. Rather than being neutral indicators of the state of the country and its economy, the most important government statistics have been crafted and manipulated to maximize their usefulness to advocates for increases in the size of government and in government spending. Here is a particularly detailed post on this subject from back in December 2016.

The two main areas of focus here have been the statistics on GDP and on poverty. Both of those come from the Commerce Department.

In the case of GDP, the biggest issue is that government spending on goods and services is counted as a 100-cents-on-the-dollar addition to GDP. That means that the most wasteful spending gives an apparent but false boost to the economy; and even more importantly, that any cut to government spending, no matter how wasteful the spending may have been, gets portrayed as a hit to the economy and a harbinger of recession.

In the case of poverty, the issue is that the official measure of “poverty” counts only cash income, while almost all government “anti-poverty” programs provide in-kind goods and services (think Medicaid, food stamps, public housing, etc.) that don’t get counted. Well over a trillion dollars of annual “anti-poverty” spending thus somehow never reduces the number of people in “poverty” by even a little; and advocates point to a continued high “poverty” rate to demand yet more spending.

For today, I’ll consider the GDP statistics. As Trump 2.0 and the Congress move forward with what will hopefully be large cuts to the most wasteful of government spending, some smart people have figured out that any such cuts will in the first instance get recorded as a shrinkage of GDP. Will this mean we have entered a “recession”? You can be sure that the legacy press will loudly proclaim that proposition. Indeed, get ready for that.

To their credit, some of the top administration people have figured out that this is coming. And thus in the past couple of days you have some of them getting out front on this issue. In a wide-ranging interview on Fox Business on Sunday (March 2), new Commerce Secretary Howard Lutnick said:

“You know that governments historically have messed with G.D.P.,” he said. “They count government spending as part of G.D.P. So I’m going to separate those two and make it transparent.”

(The quote appears in the New York Times at this link.) Lutnick’s statement echoed a similar remark from Elon Musk on his X platform on Friday (February 28):

A more accurate measure of GDP would exclude government spending. Otherwise, you can scale GDP artificially high by spending money on things that don’t make people’s lives better. For example, you could shift everyone who is building cars to working at the DMV. That would result in no cars and a much worse standard of living, but GDP would appear to be the same!

Those two remarks inspired a spirited pre-emptive rebuttal from the New York Times today (March 4). The theme: Trump and his people plan to “interfere” with federal statistics. Excerpt:

Comments from a member of President Trump’s cabinet over the weekend have renewed concerns that the new administration could seek to interfere with federal statistics — especially if they start to show that the economy is slipping into a recession. . . . It wasn’t immediately clear what Mr. Lutnick meant. The basic definition of gross domestic product is widely accepted internationally and has been unchanged for decades. . . . “The implication is that it is OK to manipulate economic data for political gain,” said David Wilcox, a fellow at the Peterson Institute for International Economics and director of U.S. economic research at Bloomberg Economics.

Good try. It’s the now familiar theme that there is something illegitimate about the elected President and his people seeking to run the executive departments of the government. How dare they “interfere” with the completely neutral and non-political efforts of the expert permanent bureaucracy!

The problem, of course, is that the efforts of the permanent bureaucracy are the opposite of neutral and non-political. I’m not sure that the initial decision to include government spending as a 100 cents addition to GDP was dishonest. But once that decision was made, it became catnip to functionaries advocating for bigger budgets for themselves and/or resisting any and all cuts.

As prime examples of wasteful government spending going at 100 cents on the dollar into GDP, consider USAID grants to NGOs funding Hamas and Black Lives Matter protests; or Inflation Reduction Act subsidies to wind turbines. The latter would amount to hundreds of billions of dollars if they go forward, clearly a net negative to the people’s well-being, yet counted at full value as an increase to GDP.

At his Substack called The Honest Broker yesterday, Roger Pielke, Jr., takes issue with Lutnick and argues for continued inclusion of government goods and services spending in GDP:

Remember that GDP is simply a summary of economic activity, not the worth of its components. There are longstanding debates over how to interpret and value government spending, but its inclusion in the top line GDP measure is standard practice — Users of GDP numbers are of course free to redefine the metric however they like.

Pielke correctly points out that the Commerce Department provides detailed GDP reports (here is the latest one for the full year 2024), that break out line items for all the components, including the government spending. Thus, if a user wants to go to the details to see how the economy would have changed without the changes to government spending (on goods and services) it is possible to do that.

The problem is that almost nobody goes to that trouble, most particularly the media that breathlessly report on the latest quarterly GDP statistics. The extent to which GDP changes are an artifact of changes in government spending generally gets completely lost.

What 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. It’s the one-page press release that counts. Nobody reads the twenty pages of detailed charts of numbers that follow.

I should note that even if Lutnick accomplishes what I suggest, that would only address a part of the problem. The portion of GDP recorded as federal government spending is only about $1.9 trillion (see page 9 at the link), out of a GDP of $29.2 trillion (page 8 at the link). The federal government spent some $6.75 trillion in fiscal 2024 (which ended on September 30, 2024); and since spending only goes up, it undoubtedly spent even a little more than that in calendar 2024 that ended on December 31. So what happened to the other almost $5 trillion of government spending? Is it somewhere in GDP?

I don’t know the answer completely, but the big pieces are (1) interest on the national debt at about $1.1 trillion (not counted in GDP), and (2) transfer payments (social security, Medicare, Medicaid, food stamps, etc.). These latter items add up to at least $3+ trillion. To avoid double counting, they are not counted in GDP under the federal spending category, but rather are counted as personal expenditures by the people who receive them and spend them. Thus, if some substantial reductions can be achieved in the transfer programs (as for example by identifying and eliminating fraudulent payments to illegal aliens or others), that will turn up as a reduction to GDP under the personal spending categories. Maybe Lutnick has some way to avoid this outcome, but if there is a way, I haven’t learned of it yet.

Meanwhile, next they should get to work on fixing the poverty statistics.