by Philip Pilkington via American Affairs Journal

The war in Ukraine has exposed some serious misperceptions about the relative economic size and military power of major nations. Before the war, it was fashionable to say—usually sardonically—thatRussia possessed an economy similar in size to that of Italy or smaller than that of Texas. “The Russian economy will be cut in half,” President Biden tweeted on March 26, 2022, “It was ranked the 11th biggest economy in the world before this invasion—and soon, it will not even rank among the top 20.”1

The statistics that the president used for this comparison were nominal GDP numbers measured in U.S. dollars. It is unclear why the president’s economic advisers did not direct him to the purchasing-power-parity-adjusted (PPP) GDP metrics, which are the standard tool economists use to compare the relative size of national economies. That statistic would have showed that Russia’s economy is the sixth largest in the world—almost as large as Germany’s and well over twice as large as the economy of Texas. Perhaps PPP-adjusted metrics have become embarrassing in D.C. of late as they show that the Chinese economy is roughly 20 percent larger than the American economy.

Yet this is only the first layer of a rather unsettling onion. In a previous essay, I pointed out that even PPP-adjusted GDP measures may be misleading when it comes to determining the relative importance of various economies.2 This argument is simple enough: not all GDP is created equal. A dollar of GDP generated by a casino is fundamentally different from a dollar of GDP generated by extracting oil. Since econo­mies like China and Russia have far larger mining and manufacturing industries, their relative economic importance is far greater than even a PPP-adjusted GDP figure shows. This goes a long way toward explaining why the Western sanctions did not cut the Russian economy in half, as President Biden promised, but instead created a severe energy crisis in Europe—and the continent now faces the horrifying prospect of dein­dustrialization.

Bad metrics can impose severe costs; that much is now clear. But can they also affect military planning? Could underestimating the relative size and power of a rival country’s economy go beyond economics and affect military readiness?

Misleading Macro Metrics

The usual metric that policymakers and pundits reach for when assess­ing military strength is military expenditure as a percentage of GDP. The fact that commentators typically use macroeconomic statistics to assess military strength today says a lot about how we tend to view military affairs in the modern world. It would hardly have occurred to, say, the Romans to assess the strength of their legions in terms of expenditure—much less expenditure relative to the total size of the Roman economy. In the Roman Republic, the Senate seemed mainly concerned with manpower and kept a close eye on recruitment and battlefield casualties.3 It seems likely that this is how military matters were viewed in most places and at most times.

States appear to have become more attuned to the relationship between military and economic power as the costs of large-scale materiel started to rise. Before the emergence of large-scale military equipment, soldiers generally supplied most of their own weaponry. A medieval knight, for example, is widely known to have had to purchase and maintain his own armor, weapons, and steed—all of which became symbolic of his wealth and power.

This seems to have changed in the Age of Sail, when large and expensive galleons became one of the most important weapons of the early modern period. These large ships required significant government expenditures. It was perhaps only natural, then, that a whole theory would develop around accumulating as much treasure as possible through trade so that the treasure could be used to build a large and powerful navy. That theory was known as “mercantilism” and was the de facto governing theory of the economy in Europe between the fifteenth and the eighteenth centuries. Still, one would imagine that when states compared their naval power to that of their rivals during this period, they had their spies count the number of ships on the opponent’s docks rather than the amount of money spent on the ships.

It seems likely that the best explanation for the move to compare military strength using economic measures is simply the creation of the economic measures themselves. The national accounts were developed as a response to new developments in economic theory, most notably the gradual emergence of the field of macroeconomics in the late 1920s and early 1930s. The first modern national accounts were published by the National Bureau of Economic Research (NBER) and were built by the Ukrainian American economist Simon Kuznets.4 This development was just in time for the Second World War, during which modern economic tools were deployed both to restructure the economy and to estimate military strength.

Over time, the use of GDP figures to analyze military strength became commonplace. We now know that, during the Cold War, up to 25 percent of Soviet GDP went to the armed forces, indicating at once the seriousness of the Soviet threat as well the USSR’s deep economic problems. Since the “end of history” was proclaimed after the Cold War, and peer-to-peer military competition seemed to recede, spending as a percentage of GDP is often the only measure of relative military power discussed in nonspecialist media.

Thus, the table below is one many have likely seen before. It shows overall military expenditure by country in nominal U.S. dollars for 2021.5 It also shows this spending represented as a percentage of each country’s GDP and as a percentage of total global military expenditure.

At first glance, this table seems to allow for an easy comparison of relative military strength among major world powers today. Yet a closer look at the actual military power of the countries listed reveals the severe limitations of these metrics. Let us examine Russia and the United Kingdom. If we look at the military expenditure in raw dollar terms, we will conclude that Russia and the United Kingdom are relatively similar in terms of military power. But anyone with even a cursory knowledge of the two militaries knows that this is almost certainly not the case.

Let us compare these two militaries considering what we have learned from the Russo-Ukrainian war, the first interstate, large-scale land war in Europe since the Second World War. Figure 2 shows manpower, tanks, artillery pieces, and nuclear stockpiles (which allow a belligerent to deter conventional escalation past a certain point) for Russia, the United Kingdom, and the United States.6 Air power is cer­tainly significant as well, but as the war in Ukraine shows, it is not necessarily decisive in a peer-to-peer conflict.

The results are shocking. Even though Russia and the United Kingdom spend approximately the same on their military in dollar terms, Russia has a military that is comparable to—if not substantially larger than—that of the United States. The United Kingdom is not even in the same league as Russia. This raises serious problems for measuring military strength by relative spending. Nor are these problems solved by using a PPP-adjusted metric. If we adjust Russian and British military spending for relative purchasing power, we get $149 billion for Russia versus $77.8 billion for the United Kingdom. True, this implies that Russia spends roughly twice what Britain spends on its military, but this does not explain the extreme differences between the Russian and British armed forces.

Of course, some of this can be explained by quality and some by stockpiles. Much of the Russian armaments, especially their nuclear weapons, are stockpiled from the Soviet Union. But the manpower differential cannot be explained in this way. Moreover, if past military expenditure contributes to current military strength, this only buttresses the case against using current expenditure measures to judge relative military power.

The use of macro metrics has also led to enormous complacency on the part of Western powers with respect to their capacity to reproduce stocks of basic ammunition. As we have seen in the comparison of the United Kingdom and Russia, large expenditures in no way guarantee the procurement of large quantities of the basic materiel needed for a functional military. This fact has been hammered home during the Russo-Ukrainian war. The Center for Strategic and International Studies (CSIS) has released estimates of how long it will take the United States to rebuild stockpiles of core munitions that have been sent to Ukraine.7 Even at surge rates, CSIS estimates that it will take the United States five years to replenish 155 millimeter ammunition stocks; 155 millimeter precision ammunition, 4 years; Javelin, 5.5 years; himars, 2.5 years; and stinger, 6.5 years.8 In general, U.S. defense manufacturing requires any­where from two to six years to replace core munitions after less than a year of sustained conflict—and with allies picking up some of the slack in terms of the required materiel. Macro metrics of military spending appear completely meaningless if the manufacturing capacity is not there to sustain a proxy war with a peer for even a year.

Micro-Level Mischief

The problems do not stop there. Even if we accept that the macro military spending metrics are broken, and that standardized attempts to adjust them fail, we might nevertheless hold out hope for some sort of clever statistical fix. But this hope disintegrates when we zoom in to the micro level and find some rather disturbing discrepancies.

Let us compare two similar weapons systems and see what their relative prices tell us about military spending. The U.S. Virginia-class submarine is a relatively new class of nuclear-powered, fast attack submarine. Its Russian rival is called the Yasen class. Military analysts broadly agree that, while the two ships have their strengths and weaknesses, they are roughly evenly matched.9 But how about the cost? This is where things get interesting. The latest listed price for a Virginia-class submarine is around $3.45 billion per unit.10 Meanwhile, the Yasen-class submarines are estimated to cost around $1.6 billion per unit.11

If we adjust $1.6bn using a PPP metric, we arrive at a purchasing-price-parity cost of around $3.48 billion for the Yasen, almost identical to the price of the Virginia-class. This shows the undoubted power of the PPP framework—and it shows why nominal dollar estimates of military spending are extremely misleading. PPP adjustments may not work perfectly, but they provide us with a better basis of comparison than nominal dollar spending.

The question is whether this holds for all comparable materiel. In figure 3, I have attempted to compare materiel in various categories using publicly available information.12 I have also included PPP-adjusted results and have added a cost-ratio metric.

What we see is that PPP-adjustment sometimes makes the costs of these weapons systems comparable and sometimes does not. What stands out from our cost-ratio metric, however, is that when the PPP-adjustment fails, it always fails in one direction: that is, even when PPP-adjusted, Russian equipment tends to be cheaper than American equip­ment. This means that, when we aggregate up to total spending, even when using a PPP-adjusted measure, we likely underestimate Russian military potential. This fits with the evidence of PPP-adjusted metrics failing when comparing British and Russian spending. The Russians—probably due to being a poorer country with lower wage costs—seem to get more bang for their military buck.

Market Prices for Materiel

Yet even these measures break down when we bring in more qualitative issues. Consider what is arguably the most pressing problem for military planners today: the existence of the high-speed anti-shipping missile.13 In the modern world, navies retain the status that they possessed during the Age of Sail; they are key to global power projection. Navies allow great powers to control shipping lanes and provide a platform to strike at countries far away. The U.S. Navy is absolutely central to America’s ability to project power in the Asia-Pacific region.

The U.S. Navy has faced a fundamental problem since the time that it became dominant, however. On October 25, 1944, at the Battle of Leyte Gulf, American sailors encountered a new type of weapon: the Japanese kamikaze bomber. The tactic was simple enough: load a fast plane up with massive amounts of explosives and then have the pilot fly it into the enemy’s ships. By the end of the Second World War, thirty-four ships had been sunk by kamikaze pilots and hundreds damaged.

The problem that kamikaze attacks pose is obvious: warships are large and slow while flying vehicles packed with explosives are small and fast. Warships are also very expensive, while flying bombs are relatively inexpensive. Today, of course, air forces no longer need their pilots to fly to their deaths because of the development of high-speed missile technology. Missiles are even more lethal than the old kamikaze planes because they are smaller and much, much faster. A World War II–era Mitsubishi A6M Zero had a top speed of around 350 miles per hour, while a modern Russian 3M22 Zircon anti-shipping missile has a top speed of around Mach 9, or 6,900 miles per hour!

One need not be a military expert to recognize the problem here. A credible case can be made that a ship could defend itself against one of these missiles. A credible case could even be made that it could defend itself against ten of them. But what about a hundred or a thousand? At a certain point, it becomes prima facie illogical to make the case that a large ship can defend itself against hundreds of anti-shipping missiles moving at Mach 9.

Together with the quantitative problems that we highlighted above, all of this suggests that we need a new way of pricing materiel. The underlying problem seems to be that there is no real market price based on combat effectiveness. Of course, defense companies “bid” on defense procurement contracts, but what programs are pursued and what contracts are available are essentially set by diktat, based on the objectives of military planners. This process might be adequate to identify the cost of producing various weapons systems, but it does not necessarily reflect their combat value. Even when weapons are sold to other countries, they are usually sold based on marketing. After all, many have not been tested in real war—by which we mean advanced combined arms warfare.

Yet it may be possible to come up with a market value, or something resembling one, and the case of the anti-shipping missile might give us a hint of what that might be. A piece of weaponry is only as valuable as it is potentially successful on the battlefield. Under conventional military procurement, if an arms manufacturer could convince a government to purchase extremely well-constructed trebuchets with expensive aiming technology for millions of dollars, the prices of those trebuchets would be millions of dollars. But that would not stop the trebuchets from being worth exactly nothing on the battlefield. Their stated price may be millions of dollars, but their value would be zero.

What this suggests is that we should be able, with prudential judg­ment, to determine the value of materiel with reference to its potential battlefield effectiveness. A weapon’s value can only be measured against its potential counter-weapon. Analysis based on this concept, if done properly, should be able to generate a “market price.” We can do so with two very simple equations:

The table below provides the definitions of the variables in an example using the cost of a U.S. Navy Carrier Battle Group and the cost of its Russian counter-weapon, the 3M22 Zircon.14

What these equations do is simply model the market price of a weapon based on its potential for battlefield success when weighed against its counter-weapon. So, taking the price of both, we adjust the counter-weapon using PPP to get a better comparison; we then take the amount of manpower being put at risk in their usage—sadly, a dollar price on human life—and finally, the probabilities that the weapon and counter-weapons will destroy one another. We have no idea how many 3M22 Zircons it would take to sink a Carrier Battle Group, but we have made the conservative assumption of 250—i.e., a probability of a single weapon sinking the carrier of 0.4 percent.

These equations, which are similar to those used in war-gaming, can also be used to “back out” how successful a weapon would need to be against its counter-weapon to justify its price. In the case of our model, a Carrier Battle Group could only justify its cost if the probability of being sunk by a 3M22 Zircon was 0.04 percent—or,what amounts to the same thing, that it would take almost 2,400 3M22 Zircon missiles to sink a Carrier Battle Group. Perhaps more realistically, one could also adjust the value (or price) of the carrier group based on the fact that it can only be deployed in situations where the threat from hypersonic antiship missiles is minimal.

With access to sufficient information about the capabilities of various weapons systems, an economist could, in theory, calculate an assessment of the economic value of a military based on its likely performance. This would be the closest we could come—without access to complete battle and weapon data—to determine the true market value of materiel. Of course, most of us do not have access to the relative performances of various weapons systems; this information is typically top secret. But even with incomplete information about these capabilities, a reason­able approximation is possible.

Many militaries around the world already use similar calculations in making threat assessments. They could therefore provide better statistics on comparative strength, without having to divulge how the statistics are calculated, although this would, of course, raise issues of trust. Nevertheless, given the clear problems policymakers and the general public have encountered in trying to determine military strength from simple economic spending metrics, this seems the most promising way to get something closer to true materiel values and more accurate military comparisons.

This article originally appeared in American Affairs Volume VII, Number 1 (Spring 2023): 90–99.