January 29, 2020 Reading Time: 4 minutes

Even somewhat less than eagle-eyed observers will, if only by virtue of the size of the number, take note: the 2020 budget for the Pentagon is roughly three-quarters of one trillion dollars ($738 billion), and yet it has reported $35 trillion in accounting adjustments in 2019, a near $5 trillion increase over the previous year. 

2018 US GDP is estimated at $20.5 trillion. World GDP is estimated at just about $85 trillion. The Pentagon’s accounting adjustments are not only larger than the entire US GDP, but have reached over 40 percent of world GDP estimates.

Does this sound crazy? It should. Something has gone very wrong with the calculations here. 

These accounting adjustments represent “double, triple, and quadruple counting” of funds moved between accounts (departments, projects, etc.), according to one explanation. The General Accountability Office further describes a massive, inefficient, virtually ad hoc accounting system rife with “errors, shorthand, and sloppy record-keeping.” 

But there are more implications to this than simple disorganization. The euphemistically labeled mishandling of money which most taxpayers would rather not hand over in the first place is irksome enough. To know that it’s being used (a) to prop up a regime which increasingly rivals the brutality of the one that cost almost 5,000 American lives to eject, (b) to sustain a twodecade-old war that’s already lost , and (c) to rattle sabers around the world is a tremendous additional affront to taxpayers, especially in light of how readily taxpayers are scrutinized for substantially smaller errors. And that doesn’t take into account a recent purchase of $10 billion worth of accounting software and systems.

When Oskar Morgenstern’s “On the Accuracy of Economic Observations” was first published in 1950, among its early reviewers was Simon Kuznets, 1976 Nobel laureate. While generally praising Morgenstern’s work, he offered several intriguing counterpoints to the recommendation that all economic statistics be published or otherwise disseminated with some form of confidence interval or error estimate. 

For example, a government statistic might take the following form: $11,756,905

To which Morgenstern would argue that, in most cases and for many reasons, that level of exactness is unjustified. A more accurate representation might be: $11,750,000 +/- $15,000

This is likely a more precise representation, both mathematically and intuitively (in most cases). Yet Kuznets offered the following caveat:

Economic statistics are conventions, as in fact most measurements are: they reflect the magnitude of what people agree to recognize as profit, price, sale, etc. rather than what these phenomena may be as defined as … [T]hey are necessarily defective as measures of controlled concepts … [b]ut the data may have value as records of convention. One may be interested in the series on “corporate profits” not only because it is an approximation (full of errors) to what profits truly are as defined by economists, but also because it reflects changes in what the business community sees as corporate profits, and to which it responds. 

Thus the second measure (b) might indeed be a more accurate representation of whatever the range of real (likely unknown, and possibly unknowable) values are, but the first – (a) – may reveal such aspects of the enterprise as the accuracy which the gatherers, reporters, and users of the statistic erroneously believe they are capturing, or more imprudently a purposeful effort to imbue an uncertain quantity with false levels of precision. 

And this applies to the recent reports about Pentagon recordkeeping as well. Beyond the actual accounting adjustments (which, too, are highly uncertain) in one of the largest departments in the government is what they represent as a matter of financial practice, organizational understanding, and political/bureaucratic acceptance. It is difficult to see these accounting adjustments – which number roughly 50 times the actual budget itself – as indicative of anything but some toxic cocktail of incompetence and misprision. 

Another comment by Kuznets alludes to a public policy aspect to accuracy in economic numbers. In the review, he adds that in some cases,

to permit … rounding off … would, in absence of accepted and highly articulated limiting criteria, open the way to uncertainty and flabbiness in the convention, which would thus lose one of its fundamental values – that of rigidity … Professor Morgenstern’s comment on the speciousness of citing the public debt of the United States to the last penny is valid if the series is conceived of as data for analysis; it is invalid if the series is viewed, as it is, as a convention the value of which lies in complete accountability within rigid rules. 

Phrased differently (and more to the point): taxpayers, holders of US dollars, and investors in US government debt securities would, regardless of the underlying circumstances, much rather see the national debt reported to the penny than estimated. Exactness as in this case is almost certainly impossible, but serves an evangelical purpose: the state, Kuznets suggests, must in some ways represent itself as a tirelessly responsible custodian of the funds collected by it or invested in it. 

Or so it was. Clearly that mandate – consider it a principle of good statecraft – has long since been abandoned. Until 2018, in fact, the Pentagon had actually avoided the legal requirement of an annual audit for decades. Yet when, in 2018, it had one, it failed miserably. The response from the deputy secretary of defense? “We failed the audit. We never expected to pass it.”

That kind of response makes spending $7,622 on a coffee pot seem quaint: it was, at least, an accurate number. Maybe. “Improper payments,” defined as funds flowing without approval or documentation, rose to $1.2 billion from $957 million between 2017 and 2018. 

It may be too much to ask, heading into an election in which political candidates are endlessly conjuring plans, proposing projects, recommending “investment,” and otherwise attempting to curry favor in their pursuit of the most powerful office on the planet, that voters consider the historical and philosophical implications of an expansive, fiscally unaccountable government. But one hopes and expects that the example of both the Pentagon’s unaccountable accounting – and, more importantly, its flip, what-are-you-going-to-do-about-it dismissal of the same – would come to figure prominently in those conversations.

Peter C. Earle

Peter C. Earle

Peter C. Earle is an economist and writer who joined AIER in 2018. Prior to that spent over 20 years as a trader and analyst at a number of securities firms and hedge funds in the New York metropolitan area, as well as running a gaming and cryptocurrency consultancy.

His research focuses on financial markets, cryptocurrencies, monetary policy-related issues, the economics of games, and problems in economic measurement. He has been quoted by the Wall Street Journal, Bloomberg, Reuters, CNBC, Grant’s Interest Rate Observer, NPR, and in numerous other media outlets and publications.

Pete holds an MA in Applied Economics from American University, an MBA (Finance), and a BS in Engineering from the United States Military Academy at West Point. Follow him on Twitter.

Books by Peter C. Earle

Get notified of new articles from Peter C. Earle and AIER.