Why the War Will Likely Last Years and Why This Makes the Outcome More Uncertain

When the invasion started, I was assuming that Russia would most likely defeat Ukraine’s military relatively easily, but that trouble would start after that if Putin intended to annex most of it. As it turned out, the Russians couldn’t even do what I thought would be the easy part and they had to retreat from Kiev after a month, but this was just the beginning of their troubles and since then they have suffered several more setbacks. This isn’t what I thought would happen, even after Russia’s initial failure in front of Kiev, but it’s what happened and it’s just silly to deny it or to pretend that things are not what they seem. Yet that is exactly what many people in the West, among those who support Russia against Ukraine, have been doing for months. They have been coming up with increasingly contorted theories to the effect that Russia wasn’t actually losing, but that everything was going according to plan, when it’s pretty clear that nothing is and frankly it’s not even clear there is a plan. Nevertheless, I think that people on the other side, who are in favor of supporting Ukraine for as long as it’s willing to fight and no matter the cost (as long as it doesn’t require that NATO enter the war directly), are probably overoptimistic.

In this essay, after outlining what I consider the most plausible scenario in which Ukraine could force Russia to end the war quickly, I argue that it’s probably not going to happen because it would have to successfully conduct another large-scale counteroffensive in the South and I don’t think it will be in a position to do so in the next few months. Indeed, not only does it lack offensive weapons despite Western deliveries, but publicly available data suggest it’s going to run out of ammunition for some critical weapon systems next year. Russia will also have this problem, but it’s going to be even more acute for Ukraine. Moreover, even if Ukraine nevertheless managed to successfully conduct such a counteroffensive and take Melitopol, I argue that it’s unlikely that neither side would be willing to make the kind of concessions that would be necessary to end the war. Thus, on the one hand neither side can force a conclusion militarily, but on the other hand the political conditions for a settlement are not going to be met anytime soon. I therefore conclude that the war is likely to last several years and argue that, if the war turns into a protracted conflict, many things can go wrong for Ukraine.

The scenario of a quick end to the war

At the end of the summer, I predicted that neither Russia nor Ukraine would be able to break the stalemate and make substantial gains anytime soon, which was almost immediately proven wrong as Ukraine managed to successfully conduct a large counteroffensive that expelled the Russians from Kharkiv and more recently recapture Kherson after the Russians abandoned it. My argument was not based on any detailed military knowledge, since as I readily admitted at the time I have none (I still don’t), but rested on the hypothesis that in this conflict the side playing defense had such a massive advantage that neither side would be able to overcome it. Again I have no military knowledge so my reasoning was just that, since Russia had been unable to successfully carry out large offensives despite the huge advantage in offensive weapons and manpower it had initially, Ukraine was unlikely to be able to do so given that even with military aid by the West it would never enjoy anything close to the same advantage in offensive weapons. I concluded that, unless it started committing significantly more resources to the war (which at the time it didn’t seem able or willing to do and in my opinion still isn’t despite partial mobilization), Russia would not be able to overcome this defensive advantage to win a decisive victory. As for Ukraine, it also would not be able to overcome this advantage unless Western countries basically set up a war economy on its behalf, which then as now I regarded as extremely unlikely to happen.

Zelensky with Ukrainian soldiers in Izyum after the liberation of the city in September

I’m still not sure why it was wrong, but I can think of several possibilities. One of them is that I underestimated the impact of morale on military effectiveness. It’s not that I think morale is not important, but I tend to think that material considerations such as manpower and equipment are by far the most important factor, which made me dismiss arguments that gave a lot of weight to the greater morale on the Ukrainian side. Another possibility is that I underestimated how much the quality of the equipment mattered. I knew that Ukraine had more sophisticated weapons than Russia thanks to Western military aid, but I was skeptical that the superiority in quality of the weapon systems delivered by Western countries would be enough to make up for Ukraine’s deficit in quantity relative to what Russia initially had, especially since compared to Ukrainian stockpiles Western deliveries — although they are significant — are pretty limited. Yet another possibility, however, is that my argument was basically sound and that the recent successes by Ukraine were due to special circumstances that are unlikely to be reproduced anytime soon. On that hypothesis, the counteroffensive in Kharkiv was successful because Russia’s lines were overextended, while Ukraine was only able to retake Kherson because the Russian troops on the right side of the Dniepr were in a uniquely bad position due to the difficulty of supplying them.

If this last hypothesis is true, Ukraine will find it difficult to achieve the same kind of success from now on, at least if Russia is minimally competent at training and deploying conscripts to fix its manpower problem, which granted is hardly obvious given what it has demonstrated since the beginning. If on the other hand there was something fundamentally wrong with my argument, whether the fact that I underestimated the impact of morale or the importance of the quality of weapon systems or some other reason, then we should know within a few months if not sooner as Ukraine successfully conducts other large counteroffensives. In that case, the war may be over more quickly than I think. The most plausible scenario I can think of for a relatively quick war is that Ukraine will manage to successfully carry out a large counteroffensive in the South, eventually taking back Melitopol, at which point Crimea would be threatened and Russia might sue for peace to avoid losing it rather than persisting in what would then appear — if that’s not already the case — to be a pointless effort. Indeed, it’s doubtful that Putin’s regime could survive the loss of Crimea, so if it looked as though Ukraine could take it back if the war continued then the Kremlin might actually start looking for a negotiated settlement and be willing to make serious concessions in order to make it happen. There are several reasons why I’m skeptical that such a scenario will come to pass though.

Why I think this scenario is unlikely to materialize

First, I’m not sure it’s militarily realistic, because again I’m still not sure that my argument from last summer was fundamentally mistaken. To be more specific, I think it’s true that I underestimated the importance of both morale and weapon systems quality, but I’m still not convinced that it’s enough to invalidate my point about the advantage enjoyed by the side that is defending and Ukraine’s lack of offensive weapons. Obviously, I was wrong that Ukraine would not be able to successfully conduct large counteroffensives, but as I noted above it could be that what happened in Kharkiv and Kherson was only possible due to special circumstances that are unlikely to be repeated elsewhere. At the same time, I appreciate that I lack the relevant military knowledge to assess the probability with confidence, so my skepticism could prove completely wrong again and Ukraine might continue to surprise me. Many people who have the military expertise I lack certainly seem to expect it, just as they correctly expected Ukraine to be able to break the stalemate last summer, but I don’t think that makes it irrational for me to remain skeptical.

Indeed, even for people with the relevant expertise and access to information that is not publicly available, predicting the course of the war seems very difficult. For instance, Western intelligence services presumably employ military experts to make predictions about it, yet at the beginning of the war they thought Russia would probably take Kiev in a few days. Of course, that is not a reason to dismiss what experts are saying, but it’s a reason to take their opinion with a grain of salt. This is particularly true when, as I think is often the case among those who hold the view that Ukraine will soon win a decisive victory against Russia, they can’t support it with a coherent argument. A rational person should have some deference toward experts, but this deference should not be unconditional. Trustworthy experts usually don’t simply assert their view and ask that you believe them because they’re experts, but make arguments to defend it. It’s true that people who lack the relevant expertise can’t assess the truth of the assumptions on which those arguments rest, but they can still recognize the logical structure of the argument. However, when I listen to the military experts who predict a decisive Ukrainian victory soon, I don’t see much of that. In particular, I don’t see any quantitative analysis, which I understand is difficult given the lack of data about many relevant aspects of the conflict but still doesn’t inspire confidence.

For instance, I recently listened to The Economist’s podcast with Ben Hodges, a retired US general who thinks Ukraine will have won by next summer, I don’t see much beyond vague generalities and flat assertions. This doesn’t mean that he is wrong, but I don’t see why I should put much credence in what he says. Even Lawrence Freedman, who in general is more careful than Hodges and who I think is a serious scholar (I have read his book on the evolution of nuclear strategy and thought it was very good), contradicts himself in a recent article. On the one hand, he admits that it’s impossible to know for sure whether Russia will manage to turn mobilized soldiers into an effective fighting force “until newly formed units are committed to the war in spring”, but he still concludes that “the war was lost long ago”. In general, many of the arguments seem to boil down to the claim that since Russia has so far been incapable of stopping Ukrainian counteroffensives, it would continue to fail, but applying this heuristic retrospectively to past conflicts shows that it’s not very reliable. Again, it doesn’t mean that people who think that Ukraine is going to continue to make rapid and substantial gains are wrong, but in light of the weakness of their arguments I don’t see any reason to defer blindly to them.

This is particularly true when, as is the case, other military experts disagree with them. For instance, General Milley, the Chairman of the Joint Chiefs of Staff seems to think that the front will stabilize and that unless a diplomatic solution is found soon both sides will suffer heavy losses without making substantial gains in the future. He was wrong in February when he predicted that Ukraine would lose quickly and may also be wrong about that, but presumably he has the relevant expertise, so it must not be obvious that Ukraine is going to keep wiping the floor with the Russian armed forces. Even Valery Zaluzhny, the commander-in-chief of the Ukrainian armed forces, seems to agree with the argument I made in September that Ukraine couldn’t break the stalemate because it didn’t have enough offensive weapons to successfully conduct large counteroffensives. He doesn’t put it like that, but he says that in order to push back the Russians behind the pre-February 24 lines Ukraine would need “300 tanks, 600-700 IFVs, 500 howitzers”. Of course, it’s possible that he is asking for more than he really needs to get as much as possible, but this is not obvious because making unrealistic demands is not necessarily a smart negotiating position. In any case, short of a radical change of stance on the part of the West, it’s very unlikely that it will provide that kind of support.

General Mark Milley, the Chairman of the Joint Chiefs of Staff (the highest-ranked officer in the US military), thinks that the war is going to turn into a bloody stalemate

Indeed, according to the Kiel Institute for the World Economy’s Ukraine Support Tracker, Western countries have delivered 280 tanks, 120 IFVs and 321 howitzers since the beginning of the war. So in order to send what Zaluzhny says he’d need to push back the Russian armed forces to the pre-February 24 lines, they would have to deliver almost twice as many howitzers as they already have, more than 5 times as many IFVs and about the same number of tanks. In the case of tanks, the only countries that have delivered any to Ukraine so far are Poland and Czechia, which gave Soviet T-72 tanks they had in their arsenal and plan to replace them with American tanks. Czechia plans to send 90 more T-72s, after they have been refurbished at the expense of the US and the Netherlands, but other countries have so far refused to send tanks and show no sign of changing their mind for the moment. (Even if they did, it would take a while before the Ukrainians could use them, because unlike T-72s which they know how to use since they already have them in their arsenal they would have to be trained to use Western tanks.) Ukraine has been asking for more offensive weapons for months now, but Western countries seem wary of sending them for various reasons, such as the fear that it might lead to a dangerous escalation. Some people in the West argue we should abandon this restraint, but even if they eventually win this debate (which is hardly obvious), a change of stance is likely to be gradual.

Even if the West delivered all that Zaluzhny asks, Ukraine would still not come close to enjoying the kind of advantage in offensive weapons that Russia had at the beginning of the invasion, which still wasn’t enough. Let’s estimate how the ratio has changed for each type of equipment, starting with main battle tanks. According to RUSI’s latest report, Russia initially committed ~3,200 tanks to the invasion, while Ukraine had ~900. That’s a 3.5:1 ratio in favor of Russia. In order to estimate what the current ratio is, I have subtracted the documented losses and added the documented captured tanks on each side (according to Oryx), before in the case of Ukraine adding the tanks that were delivered by the West according to the Ukraine Support Tracker. This methodology probably results in a large bias in favor of Ukraine, because Russian losses are better documented and it’s probably less able to refurbish captured tanks as the parts are only made in Russia, but it still finds that Russia has ~1,400 tanks whereas Ukraine only has ~1,300. It also ignores the fact that Russia has a huge number of tanks in storage, more than 10,000 according to the IISS, that it can use to replenish its forces. (I’m sure that most are not combat-ready, but given how many of them it has, Russia was probably able to make up for most if not all of its losses.) Even if Ukraine received 300 tanks, as Zaluzhny asks, it would still be far from the 3.5:1 ratio in favor of Russia at the beginning of the invasion and would almost certainly have far less tanks than Russia once you account for the bias in the methodology.

The same methodology can be used to estimate how the ratio has changed for howitzers. According to RUSI, Russia initially committed ~2,400 howitzers to the invasion, whereas Ukraine had ~1,200. That’s a 2:1 ratio in favor of Russia if RUSI is talking about the number of howitzers initially committed to the invasion, but it’s less clear than in the case of tanks, so it may have been less than that. In any case, after updating the numbers for each state based on the documentation established by Oryx, I reach the conclusion that Ukraine now has ~1,500 howitzers while Russia only has ~1,400. Again, this is biased in favor of Ukraine, due to the fact that Russian losses are better documented and that Ukraine is presumably less able to refurbish captured Russian howitzers. Even more importantly, Russia has vastly more of them in storage (more than 16,000 according to the IISS), so even accounting for the fact that many of them are probably not operational it almost certainly still vastly outguns Ukraine. (As we shall see below, Russia also has a huge advantage in ammunition, which is going to get worse for Ukraine.) This would still be true even if Ukraine received 500 howitzers, as Zaluzhny says it needs to conduct new large counteroffensives.

RUSI doesn’t give any estimate of the number of IFVs Russia initially committed to the invasion, but according to the IISS, it had ~8,000 out of storage plus about the same number in storage whereas Ukraine only had ~1,200. Thus, even though according to Oryx’s documentation Russia has lost far more than Ukraine (1,872 included 580 that were captured whereas only 423 losses included 89 that were captured have been documented for Ukraine), I think it’s safe to assume that even if Ukraine received the 600-700 IFVs that Zaluzhny requested, it would still not come close to the kind of advantage Russia initially had. The methodology and the data I have used are extremely imperfect, so the figures I give should only taken as very rough estimates, but they are sufficiently lopsided that I think we can safely assume that even if the West vastly increased the military aid it provides to Ukraine, which is unlikely for reasons I have already discussed and others I will discuss below, the Ukrainian armed forces will not have the sort of advantage in purely quantitative terms that Russia had at the beginning of the invasion. It may be that morale and the quality of Western equipment will more than make up for it, I lack the expertise to know for sure, but I find that hard to believe.

Another problem that Ukraine is going to face, beyond the lack of offensive weapons, is a shortage of ammunition for some critical weapon systems. Perhaps the worst problem that Ukraine is going to face in 2023 is that it’s probably going to run out of ammunition for air defense systems. Ukraine has been relying mostly on S-300 and Buk systems for air defense. (The Ukrainians have also been making extensive use of MANPADS, shoulder-fired air defense systems, but those can only be used against targets flying at low altitudes.) It’s estimated that Ukraine had 250 S-300 systems and 72 Buk systems at the beginning of the war, although it’s not clear exactly how many were operational and the loss of several has been documented since then. The problem is that the missiles used by S-300 and Buk systems are only manufactured in Russia and it’s difficult to procure them from countries that have stockpiles of them, so Ukraine has been forced to rely on the stockpiles it had at the beginning of the war, which are dwindling and will eventually run out.

Unfortunately, I haven’t been able to find estimates of the number of missiles for S-300 and Buk systems that Ukraine had at the beginning of the war, but the Ukrainian Defense Ministry claims that Russia had 8,000 missiles for S-300 systems and presumably Ukraine’s stockpiles were much smaller. According to the same article, not counting the missiles used by S-300 systems (which are primarily used for air defense), Russia has fired about 2,400 missiles in Ukraine since the beginning of the war. According to the Ukrainians, their air defense systems fire 2 missiles for every Russian missile to increase the chance of interception, so they have probably already used several thousands S-300 and Buk missiles, but it’s difficult to estimate the number accurately. On the one hand, the Ukrainians don’t attempt to intercept every Russian missile, if only because they can’t protect every part of their territory and the Russians try to attack targets that are not covered by Ukrainian air defense systems when they can. On the other hand, the Ukrainians presumably used those systems against Russian planes and drones, not just missiles.

So I don’t think it’s possible based on publicly available information to estimate when Ukraine will run out. It also depends on how many drones and missiles Russia has left to use against Ukraine, as well as on how fast it can produce more, which is very unclear. At the end of November, Ukraine claimed that Russia had about 1,500 offensive missiles left, in addition to almost 7,000 missiles for S-300 systems. (Before the war, Russia’s production of precision-guided missiles with a range of more than 300km was estimated at 225/year, but it’s hard to tell what it is now. On the one hand, Russia now has incentives to increase the production and no doubt allocates more resources to it, but on the other hand sanctions probably make it harder to source components although it has been able to bypass them to some extent.) While S-300 systems are normally used for air defense, it has been reported that Russia has modified some to use them against ground targets. People assume that it’s because it’s running low on offensive missiles, but if Russia has indeed used S-300 systems to perform surface-to-surface strikes (which apparently is still unclear), it may also be because Russia has a lot of ammunitions for those systems and want to use them to deplete Ukraine’s air defense ammunitions. Regardless of whether the reports about the use of S-300 systems to attack ground targets are accurate, Russia is clearly trying to deplete Ukraine’s missiles for air defense systems, as shown by the report that it was using old missiles designed to carry nuclear warheads to do just that.

Thus, although it’s impossible to know exactly when this will happen, there is little doubt that Ukraine will soon run out of ammunitions for S-300 and Buk systems. When it does, it will have very little air defense left, leaving both critical infrastructure and troops on the front extremely vulnerable to attacks from the air. Indeed, while Ukraine has hundreds of S-300 and Buk systems, Western countries have only delivered 20 air defense systems according to the Ukraine Support Tracker. A handful more have been pledged, but it will take months or in some cases years before they are delivered and they won’t make up for the loss of Ukraine’s S-300 and Buk systems. For instance, while the announcement that the US would provide a Patriot air defense system to Ukraine was made with much fanfare, we’re just talking about one battery whereas Ukraine is about to lose hundreds of air defense systems. Norway has promised 8 NASAMS, but only 2 of them have been delivered and the others still have to be manufactured, which could take up to 2 years. Germany has promised 4 IRIS-T systems but only has delivered one and the others are still on production lines.

In a report on Russia’s air war published in November, RUSI warned of dire consequences for Ukraine if this happened:

If Ukrainian SAMs are allowed to run out of ammunition, then not only will Ukrainian infrastructure and other key target sets become dramatically more vulnerable to Russian missile strikes, but the Russian VKS fixed-wing fleet would suddenly again be able to start penetrating deep into Ukrainian-controlled airspace at medium and high altitudes.

Unlike in the case of tanks and IFVs, which the US could provide in large number to Ukraine if it wanted to, Western countries don’t seem to have many air defense systems they could give to Ukraine even if they wanted to. According to that same report, that’s because after the end of the Cold War they largely stopped investing in the production of that kind of systems, as they’ve had overwhelming air superiority in every conflict since then and expected that to continue.

A S-300 system of the sort Ukraine relies on for air defense

In order to prevent such a disaster from happening, the authors recommended that Ukraine partner with Western countries to produce missiles for Soviet air defense systems, but as far as I know this hasn’t happened. Even if Western countries had shown a willingness to make that kind of investment, which so far hasn’t been the case, it would no doubt take months if not years before they learn how to produce those missiles and start making them in large enough numbers. Even MANPADS are going to become scarcer, as Raytheon had stopped producing Stinger missiles before the war and had to resume production this year, but will not be able to ramp up production significantly for a while because it takes time to restart a munitions production line and some of the components are not even commercially available anymore, so the company will have to redesign some of the electronics in the missile. Thus, I think that sometime next year Ukraine will probably lose most of the air defense capabilities it has had since the beginning of the war, which in turn will complicate things considerably for Kiev. It will have to choose between protecting cities and infrastructure in the rear and troops in the front. I think it will probably do the former, so any counteroffensive will presumably be more vulnerable to Russian attacks from the air than in the past.

Ukraine will not just suffer from a lack of ammunitions for air defense systems next year, it will also face a shortage of ammunitions for artillery, which is critically important if it’s to recover more territory and defend against Russian offensives. Ukraine has many Soviet artillery systems, both tube artillery that fires shells and rocket artillery, but it started to run out of ammunitions for them several months ago and now relies almost exclusively on Western systems that use different ammunitions. (The Ukrainians apparently captured ammunitions for their Soviet artillery systems from the Russians during the Kharkiv offensive and recently announced that it had started to produce 152mm and 122mm shells for Soviet howitzers for the first time since 1991, but with the Russian attacks on the electric grid it’s probably safe to assume that volumes will be limited. Similarly, the US has somehow been able to find ammunitions for Soviet artillery that are included in the latest military aid package, but it just amounts to a few days worth of consumption at the rate Ukraine is currently firing them. Thus, while it will not completely stop using Soviet artillery systems, it’s safe to assume that Ukraine will continue to rely mostly on Western systems.) According to the Ukraine Support Tracker, Western countries have delivered 321 howitzers and 29 multiple launch rocket launchers, including 16 HIMARS. The US is going to deliver 4 more HIMARS pulled from its stockpiles and pledged to send 18 more later, but those will have to be procured from industry because US stockpiles are running low and the military is getting nervous, so they won’t arrive for months.

Although it’s vastly outgunned by Russia, Ukraine still consumes artillery ammunitions at a very high rate. According to a Ukrainian official quoted by The Guardian in June, Ukraine was using 5,000 to 6,000 shells/day back then. According to a US official quoted by NBC News last month, Ukraine was firing between 4,000 and 7,000 rounds/day. (Both sources estimated that Russia was firing 20,000 shells/day. Another Ukrainian officials claimed in June that Russia was firing as many as 50,000 rounds/day, but I suspect he was exaggerating to impress on people in the West how dire things were for Ukraine.) As far as I know, the methodology used by Western officials to estimate Ukraine’s consumption was not explained anywhere, but I think they can estimate it pretty accurately. Indeed, the barrel of howitzers has to be changed because of wear after they have fired a certain number of rounds (about 2,500 for the most common type of artillery, the M777 howitzer, delivered to Ukraine by the West), so presumably Western officials can estimate Ukrainian consumption based on how often they’re asked to replace them.

According to a New York Times article from the end of November, a third of the howitzers delivered to Ukraine by the West is out of action at any given time, so I think the estimates of Ukrainian consumption are plausible. They could even be independently verified if we knew how much time it takes on average for a howitzer whose barrel needs to be replaced to be taken away from the battlefield, sent to Poland where the Pentagon set up a repair shop (other Western countries are repairing the howitzers they sent to Ukraine in other neighboring countries) and sent back to Ukraine to be deployed on the battlefield after the barrel has been replaced. (If we assume that Ukraine fires 4,000 rounds/day, making a few simplifying assumptions, we can infer from the fact that the repair process is steady state and that a third of the howitzers sent by the West are out of action at any given time that it takes about 2 months on average or equivalently that on average 1.6 howitzers leave the battlefield and the same number come back from repair every day.) The problem is that it’s not going to be possible even for the West to deliver ammunitions fast enough to sustain Ukraine’s consumption at the current level.

Indeed, the US is currently producing only 14,000 155mm shells a month, which is barely 1/10th of Ukraine’s consumption if we assume that it fires 4,000 rounds/day. Washington recently announced that it would “dramatically” increase production to be able to keep supplying Ukraine while replenishing US stockpiles, but according to that plan, production would just reach 20,000 rounds/month by the spring and 40,000 rounds/month by 2025. Thus, even by the time the US is done ramping up production, it will at most be able to sustain a third of Ukraine’s consumption at current levels. Western countries have so far pulled ammunitions from their own stockpiles to supply Ukraine, but they can no longer do that because they are emptying rapidly and defense officials are getting nervous. So even taking into account the fact that Ukrainian started to produce shells for Soviet howitzers, I don’t see how it’s not going to face a shortage of ammunitions for howitzers in the next few months, which means that it will have to reduce consumption. (Moreover, Soviet howitzers also suffer from wear, but I don’t know if Ukraine can produce the barrels it would need to keep fielding them.) It’s harder to find data on Ukraine’s consumption of ammunitions for MLRS, but it’s likely high and Western production capacities for them are also very low, so presumably it will have the same problem with rocket artillery.

The Russian armed forces will also have a hard time maintaining their current level of consumption, but as we have seen they start from a much higher level, so they can reduce their consumption and still vastly outgun the Ukrainians even at their current level. Experts disagree wildly about whether Russia will face shortage of shells. Some of them argued that it would happen by the end of this year, but others think it can sustain the current level of consumption for more than a year. There was recently a report that Moscow bought millions of shells to North Korea, which some experts interpreted as a sign they were already running out, but others argue that Russia may just have done that to shore up short-term supplies while it’s expanding domestic production and preparing for a long war. A Russian opposition website estimated that before the war Russia was producing about one million shells. (They give a slightly higher estimate, but their calculation is a bit off, so I corrected it.) Another paper estimated Russia’s pre-war production of 152mm shells at more than 700,000/year. This would only suffice to sustain a small share of Russia’s current level of consumption, but it’s still vastly more than what the US is planning to produce even after it has implemented the recently announced plan to expand production capacity, although it’s hard to know how Russia’s production capacity has changed since the war started. On the one hand, Moscow obviously has a strong incentives to allocate more resources to ammunition production now, but on the other hand sanctions probably make it harder other things being equal.

Sanctions will probably have less impact on shell production than on precision-guided munitions though, because they don’t require electronic components, so I think Russia will most likely be able to maintain pre-war production and perhaps even expand it. On the other hand, it will probably have a harder time producing MLRS ammunition, so I think it will likely have to reduce its use of rocket artillery. Nevertheless, I think it will almost certainly be able to fire vastly more rounds of artillery than Ukraine in the future, especially if it can buy shells from North Korea and other nations that have stockpiles of ammunitions for Soviet-made artillery systems. To be sure, Russia and Ukraine will both have to reduce their consumption of artillery shells, but Ukraine will probably have to do so more than Russia. This will presumably be a serious problem for Ukraine and, at the very least, should complicate its efforts to recapture the territory it has lost to Russia earlier this year.

US officials have apparently made similar calculations and reached the same conclusion, but they think Ukraine can make up for the lack of ammunition with training and better tactics:

“I think if we can train larger formations — companies, battalions — on how to employ fires, create conditions for maneuver, and then be able to maneuver like you’ve seen [the U.S. military] maneuver on the battlefield, then I think we’re in a different place. Then you don’t need a million rounds” of artillery, a senior U.S. defense official said.

However, due to the fact that Ukraine lacks the air support and the logistics of the US military (as we have seen it will likely run out of air defense on the front soon), some military experts are skeptical. I’m inclined to think they are right, but I don’t have any military expertise and Ukraine has already surprised me in the past though, so maybe it will do so again. However, as I argue in the next section, I don’t think the war would end even if the Ukrainian armed forces managed to successfully conduct another large counteroffensive in the South and take Melitopol anyway.

Peace would remain unlikely even if Ukraine managed to maintain its momentum

Another reason why I doubt that the war could end relatively quickly is that, even if my skepticism about Ukraine’s military prospects is unjustified and the Ukrainian armed forces manage to make a breakthrough in the South in the next few months that puts them in a position to threaten Crimea, it’s hardly obvious that it would lead to a negotiated settlement that would quickly end the war. Indeed, for that to happen, both Ukraine and Russia would have to accept difficult compromises and it’s doubtful that it would be the case even in this scenario. In the case of Ukraine, the government has consistently and repeatedly claimed that it sought the return to the country’s internationally recognized borders, which includes not only the part of Donbas that has been controlled by separatists with the support of Russia since 2014 but also Crimea. While such a claim may initially have been part of a negotiating posture, it makes sense to make unrealistic claims at first which once dropped can be presented as concessions during the negotiations, I doubt that is the case today. Repeating that claim so often doesn’t just make it easier to present dropping that demand as a concession, it also makes it harder to actually drop it once negotiations start, because it creates a strong expectation on the part of the Ukrainian population that Kiev is actually going to go through with it and therefore makes abandoning that goal very costly politically.

A poll conducted in September, after the Kharkiv counteroffensive had started but before it was over, showed that a large majority of Ukrainians were in favor of continuing the war until they have recovered all the territory lost since 2014. This position is likely even more popular today after the liberation of Kherson and it would no doubt be even more so in the scenario I’m considering here, in which the Ukrainian armed forces have managed to conduct another successful counteroffensive in the South, leading to the liberation of Melitopol. In this situation, the capture of Crimea would seem within reach for Ukrainians (this impression may be misleading because a serious attempt to retake the peninsula would probably lead Russia to a significant escalation but there is no doubt that it’s how people in Ukraine would perceive it), so it’s dubious that even Zelensky (who has become extremely popular since the beginning of the war) would have the authority to abandon this goal and would take the political risk to do so in exchange for peace. Frankly, the Ukrainian leadership would probably find it hard to do so even independently of the political cost, because people get drunk on their success and history demonstrates unambiguously that political leaders are not immune to this phenomenon. While at the beginning of the invasion, Ukraine was apparently willing to consider a change of Crimea’s status, the situation has changed since then and it’s dubious that the Ukrainians would be willing to make that kind of offer as long as things continue to go well for them militarily.

However, unless Ukraine expressed a willingness to abandon the goal of retaking Crimea in exchange for peace, it’s dubious that Russia would agree to stop the war even after another defeat in the South. No matter how bad things are going for Russia militarily, it’s hard to see how the Kremlin could ever accept losing Crimea. The annexation of Crimea is seen by Putin as his greatest accomplishment and a key part of the legacy he intends to leave behind. It’s also unclear that his regime could survive the loss of legitimacy that would result from the recovery of Crimea by Ukraine, which provides a strong incentive against making that concession to end the war. Putin also made it considerably more difficult for him to reach a negotiated settlement by annexing Donetsk, Kherson, Luhansk and Zaporizhzhia in September. By doing so, he effectively burned his bridges. It’s hard to see how even another defeat in the South could force him to negotiate seriously. Even if Ukraine pulled that off and Putin were willing to abandon the recently annexed oblasts, both of which — especially the second — I regard as highly unlikely, he probably couldn’t abandon Crimea even if he wanted because not only the ruling elite but even Russian public opinion would never accept it. Indeed, while a recent poll shows that a majority of Russians support negotiations to end the war, 78% oppose returning Crimea and 66% think the same thing even about Donbas. Even accounting for social desirability bias, Putin would likely be prevented from making that sort of concession even if he were inclined to do so, which he clearly isn’t at the moment and probably still wouldn’t be even if Ukraine inflicted another defeat to his armed forces in the South.

Putin at a rally to celebrate the eighth anniversary of Russia’s annexation of Crimea in March

In theory, the West and particularly the US could increase the probability of a negotiated settlement by inducing Russia and Ukraine to make concessions with the right combination of sticks and carrots, but it’s very unlikely to happen for political reasons. For Western countries to have a chance to induce both sides to make significant concessions, they would have to credibly make threats and promises to both Ukraine and Russia that I don’t think they can. The Ukrainians aren’t going to stop fighting, especially if they manage to carry out another successful counteroffensive in the South, unless they believe that Western countries will cut their support if they don’t make significant concessions and even in that case it’s not obvious the Ukrainian government would be able to do so. But the Ukrainians know that, as long as they keep winning, cutting support outright would be politically very difficult in the West, especially in the US where the blob would go ballistic, so I think we’d have a hard time preventing them of having a go at Crimea if they retook Melitopol. We could make it more likely by making credible promises of support for reconstruction and relatively quick EU membership after the war, but this would be so expensive that Kiev would have every reason to doubt the promises we make unless there is a mechanism to hold us to them, which is hard to imagine. So although I think most Western countries don’t want the Ukrainians to try to retake Crimea and will try to talk them out of it should they ever be in a position to make the attempt, which to be clear I think is the right position because it would result in total war with a high degree of probability (more on that later), I doubt they would be successful.

Even if they were, it would only solve half of the problem, because Russia would also have to be willing to negotiate and, even in a scenario where Ukraine inflicted another defeat to the Russian armed forces in the South, I don’t think it would happen unless the West credibly made promises and threats that I don’t think it would be in a position to make. First, we’d have to credibly threaten that unless Russia makes serious concessions we’d continue to provide Ukraine with the same level of support that we have so far even if the war continues for years, but Putin clearly doesn’t believe we are going to do so and for reasons I will explain below I think he is probably right. But we’d also have to credibly promise the end of most of the sanctions that we took in response to the invasion and I don’t think we could. Indeed, while promising to lift some of the sanctions to facilitate a peace settlement could probably be done, most of them would have to remain in place because no Western political leader could afford to lift them even in exchange for peace. People would say that we’re rewarding aggression and that would be the end of it. But if people in the Kremlin think that Russia would remain a pariah state even after making what for them would be huge concessions, they are unlikely to make such concessions, even if the war continues to go badly for them. The sunk cost fallacy is one hell of a drug and the West is not doing anything to make it easier for Russia to resist the temptation to fall into it.

A lot of things can go wrong for Ukraine if the war turns into a protracted conflict

Thus, even if Ukraine manages to maintain its momentum and inflict another defeat to Russia in the South (which as I have argued we have good reasons to doubt), I think it’s unlikely to bring the war to a rapid conclusion. However, while so far things have been going well for Ukraine and people therefore assume that it will continue, if the war turns into a protracted conflict there are good reasons to think this might change. In general, the longer it lasts, the more difficult it is to predict how it’s going to end because as it drags on it becomes more likely that a low-probability event, which by definition is harder to predict, will occur and change the war’s dynamic. Of course, not all of those low-probability events would favor Russia and some on the contrary would accelerate the current trend in Ukraine’s favor (e. g. Putin could be overthrown and the Russian government be subsequently paralyzed by the ensuing struggle for power), but everybody is talking about the second kind of low-probability events while relatively few are talking about the first kind so I want to focus on them. Moreover, while some of the events that could change the dynamic of the war in Russia’s favor are low-probability and hard to predict, some are actually very likely and easy to predict. In some cases they will most likely occur not in a few years but sometime next year. Here are some of the things that could, and in some cases probably will, go wrong for Ukraine if the war doesn’t end soon:

• One of the most obvious possibilities, if the war goes on for years, is that Ukraine will eventually run out of manpower before Russia does. Indeed, Russia’s male population of military age is more than 3 times as large as that of Ukraine (I looked at men between 15 and 64 because the data were more readily accessible but presumably the ratio is roughly the same for men between 18 and 45), so if the war drags on this is a very serious possibility. The Ukrainians have been impressively resilient so far but everyone has a limit and beyond a certain rate of casualty even the Ukrainians will no longer be willing to continue to fight.
  The Ukrainian government says that between 10,000 and 13,000 of their soldiers have died. This is probably an underestimate, but even if we accept it and assume a 4:1 wounded-to-killed ratio, that’s already between 50,000 and 65,000 casualties. Milley recently estimated that Ukraine had suffered 100,000 casualties, but in his case this may be an overestimate, since he made this claim in a context where he was arguing that Ukraine should negotiate a settlement this winter and therefore had an incentive to exaggerate the cost of the war for Ukraine. In any case, between 50,000 and 100,000 casualties after 9 months is already a lot and, given the advantage the side that is defending seems to enjoy in this war, it will likely get worse if Ukraine attempts more counteroffensives, especially now that Russia has been able to fill its ranks with conscripts.
  It’s possible that Russia will reach the breaking point before Ukraine, because although it has a much larger manpower potential the Russian armed forces also have lower morale for the moment and there is less popular support for the war, which presumably makes it harder to mobilize resources for the war and lowers the tolerance for the pain it causes. But it’s hardly obvious and it’s also possible that the Kremlin will be able to continue to conscripts men of military age, though more gradually and discreetly than before, to take advantage of its much larger manpower potential even as it suffers a high casualty rate.

• Another risk if the war continues for more than a year is that Ukraine’s economy will eventually collapse and make it impossible for the country to keep fighting. The Ukrainian economy is already in shambles as Russia occupies 1/5 of the country and Black Sea ports except for grain shipments. A large share of the workforce is now under arms and can no longer contribute to the economy, while close to 8 million people have fled the country, with several millions more who have been internally displaced. The World Bank expects Ukraine’s GDP to fall by 35% in 2022 and forecasts a growth of 3.3% in 2023, but this was before Russia started to systematically destroy the country’s energy infrastructure.
  In September, before those attacks started, the World Bank estimated that $349 billion worth of physical assets had already been destroyed. The public deficit has increased massively as revenue collapsed while military expenditures increased rapidly. In October, Zelensky stated that Ukraine would need $55 billion in external support next year ($38 billion to finance the deficit and $17 billion to rebuild critical infrastructure), but some officials now think that because of Russian attacks on the country’s energy infrastructure it may need another $2 billion/month in external financing in 2023. The US and the EU have provided support in the form of grants or loans this year, but disbursement has not been fast enough (EU officials and Germany have even blocked some loans over concerns that Ukraine will default on its debt), forcing the Ukrainians to dig into their foreign reserves to buy crucial imports and to print money to finance the deficit. As a result, inflation reached 26.5% this year.
  For 2023, the EU and the US have only committed to provide $30 billion, barely more than half what Ukraine estimated it would need even before Russia started to systematically destroy critical infrastructure, so this situation will continue next year and likely become even worse. According to the same article, Ukrainian officials have tried to get more but apparently they “found a cool reception” from Western officials, who “were already wary of appearing to support too much aid for Ukraine”. This cannot continue indefinitely and, unless international support increases significantly (which at the moment seems unlikely), Ukraine will eventually no longer be able to keep fighting because the government won’t be able to pay its bills anymore and the population will lack basic necessities.

• If despite the shortage of ammunitions for air defense and artillery, Ukraine manages to continue to recover territory from Russia, it will eventually have to perform attacks on Crimea and internationally recognized Russian territory on a regular basis. Indeed, if Ukraine gets closer to Russia’s borders, the Russian armed forces will presumably move ammunitions depots and other storage across the border, so the Ukrainian armed forces will have no choice to disrupt Russian logistics other than to attack them over there. This will predictably result in civilian casualties on Russian territory even if Ukraine tries to avoid it. If the Ukrainians try to take back Crimea, they will also have to cut supplies to the peninsula, which will lead to widespread civilian suffering.
  After Russia’s initial effort failed at the beginning of the invasion, when it became clear that Russia couldn’t win a decisive victory without committing more resources to the war, many people have been surprised that it did not immediately do so. After the Kremlin declared partial mobilization in September, Russia has admittedly increased defense expenditures, but probably not as much as it should have given the difficulties it’s facing in Ukraine and the likely cost of fielding another 300,000 men. (If Russia actually manages to deploy those conscripts to Ukraine, the number of men it’s fielding will more than double, yet it’s only going to increase defense spending by 43% in 2023. Of course, the defense budget includes more than the cost of the troops that are being deployed in the ar, but I still doubt that it will be enough to equip and train adequately so many conscripts.) As I argued before, this is likely due to the constraints inherent to the nature of the regime, which among other things make it harder to extract resources from society to support the war effort. This is probably why so far Putin has tried to shield the Russian population from the effects of the war as much as possible and in particular didn’t reduce consumption to fund the war effort.
  Now, if Ukrainians start killing Russian civilians, popular support for the war may increase and make it easier to do so. I don’t think the average Russian cares much about Kherson, but if civilians in Crimea are starving and getting shelled by Ukraine, he may start to care a lot more and be more willing to make sacrifices to punish the Ukrainians. Of course, it’s hard to know for sure until it happens (maybe the Russian population will turn against Putin instead), but if the war turns into a prolonged conflict this is a possibility that can’t be ruled out. Even after the surge in military spending it recently announced, Russia is still going to spend only ~7.5% of its GDP on military expenditure in 2023, which is ridiculously low for a country that is fighting a large-scale conventional war. This means that it has a lot of room to commit more resources to the war and, while the elasticity of military power with respect to defense spending is probably relatively low for Russia due to sanctions and a weak industrial base, it can’t be zero.

• Finally, if the war lasts several years, there may eventually be a split between the US and its European allies over the support for Ukraine. While the US is relatively insulated from the economic consequences of the war, this is not true of Europe, which used to import a lot of its energy from Russia. Except for Hungary, Central and Eastern European countries have been enthusiastically supporting Ukraine because they still perceive Russia as a security threat, but with the exception of the UK the same thing can’t be said about Western European countries, which have been dragging their feet and arguably had to be bullied into supporting Ukraine as much as they currently do.
  I don’t believe for a second they will completely stop to support Ukraine or publicly insist that it make territorial concessions to end the war, because anyone who does would be destroyed by the media, but I doubt they will continue to provide the same level of support for several years and I think it’s likely they will pressure Ukraine to do it by quietly lowering the amount of aid they provide or delaying the disbursement of funds as they already have been doing. Putin is clearly making the calculation that the West will blink before him and I’m frankly not sure he is wrong.
  In theory, the US could pick up the slack and increase the aid it sends, but it can’t totally ignore its European allies if they want the war to end badly. If the US has to choose between the European alliance and Ukraine, it’s doubtful that it will choose Ukraine. The fact that Sullivan recently urged Zelensky to signal that he was open to a negotiated settlement, which reportedly was to keep the Europeans on board, suggests that it’s not a far-fetched possibility. Europeans are also complaining about US policies, such as the Inflation Reduction Act, which compounds the effects of the war for them, while the US gets to sell liquefied natural gas at record high prices.
  Moreover, even in the absence of frictions within NATO, the US itself may not be willing to support Ukraine indefinitely. Not only could the next administration be less enthusiastic about supporting Ukraine against Russia, but even the current administration has already shown that it was not willing to support it at any cost. As we have seen, it has so far refused to send many weapon systems that Ukraine has been asking for months, while people in the administration like Milley have signaled that they thought Ukraine should negotiate this winter and even Sullivan reportedly asked Zelensky to reconsider the goal of retaking Crimea. A recession could also make it politically more difficult to spend billions of dollars to help Ukraine. Thus, even if Biden is still president after 2024, the US may eventually reduce the aid it provides to Ukraine.

Conclusion

It’s clear that, to say the least, Russia’s invasion of Ukraine has not gone according to plan. It has suffered huge casualties and lost a very large amount of equipment, is facing economic sanctions that will significantly hinder the war effort and have a devastating effect on growth in the long-run by denying it access to Western markets and depriving it of Western capital and technology and blocking access to Western markets, which in turn will probably have very bad demographic consequences. I also doubt that Russia can resume large offensives anytime soon. To be sure, it has recently mobilized a lot of men, but even in the best case scenario for Russia it will take months before all of them have been properly trained and equipped, so I doubt they can make up for the more qualified personnel they lost earlier. Many of the soldiers that were initially committed to the invasion had received years of training and yet they still failed, so I don’t see how people who have just received a few months of training can do better, especially when they have less equipment and ammunition. I don’t think Russia can win a decisive victory without committing significantly more resources to the war, something that is not easy to do for the regime, but I also think it’s a mistake to assume that Ukraine has already won and that it’s only a matter of time before the Russian armed forces are forced to go back to Russia, as several analysts claim.

I have argued that, although there is a plausible scenario in which Ukraine is able to force Russia to end the war by conducting another successful counteroffensive in the South, this scenario is unlikely to occur and that even if it did the war probably wouldn’t end anytime soon. However, if I’m right about that and the war turns into a protracted conflict, many things could go wrong for Ukraine and it’s hardly obvious that it would continue to be as successful as it has been recently. The only way to prevent that from happening would be for the West to set up a sort of war economy on behalf of Ukraine, but that’s overwhelmingly unlikely to happen for political reasons. While I also don’t think that Russia can force Ukraine to make the sort of concessions Putin would need to stop military operations without losing face in the next few months, it means that such a scenario can’t be ruled out in the future. It’s also possible, if neither side manages to force the decision on the battlefield and Putin doesn’t commit enough resources to the war (maybe because he can’t), that it will eventually turn into a frozen conflict without any clear resolution. I make no mystery that I’m not a military expert, so it’s entirely possible that some of the things I’ve said are wrong. However, if something I said is wrong, I still need to know why. It may be that the people who claim that Russia has already lost and that Ukraine will be able to end the war on its terms next year are right, but I think the points I raised in this essay are reasonable and I would like to see them addressed before I believe that.