RAND’s study of N. Korea collapse should be required reading at State, USFK

This week, the World Bank recently analyzed a series of governance indicators to conclude that the North Korean regime is stabilizing. Not surprisingly, not everyone agrees. Bruce Bennett of RAND has just published an indispensable, readable, and plausibly terrifying new study of the regime’s stability, and he reaches a very different conclusion. To Bennett, a violent and chaotic collapse looks increasingly likely as North Korea tries to consolidate succession to its third hereditary ruler. (Thanks to a reader for forwarding).

Bennett lays out a series of factors that have accurately predicted the collapse of other totalitarian regimes in third-world ex-colonies. Despite some necessary hedging, a careful read suggests that a violent collapse, humanitarian disaster, and geopolitical anarchy are not a matter of “if” but “when.” Bennett urges the U.S., South Korean, and Chinese governments to discuss–urgently–how to respond to a host of predictable problems to save as many lives possible, to stabilize and reunify Korea as quickly as possible, and to prevent war.

Now, if you’ve read this site for any length of time, you’ve watched my North Korea collapse clock jiggle between 11:52 and 12:03 for nearly a decade. I agree with almost everything Bennett writes. Maybe you don’t, but if you’re interested in North Korea, read his paper. In case you aren’t interested enough to read over 300 pages, however, I’m here to help.

Why Bennett thinks North Korea is collapsing:

Whether you believe that North Korea is stabilizing or increasingly fragile probably depends on what you read and what you believe. Bennett, for example, cites a Joongang Ilbo report of an attempt to assassinate Kim Jong Un in 2012, and he collects a series of similar reports of aborted mutinies and assassinations, starting on page 45. Even if you discount every one of these reports, there is plenty of evidence that the regime’s ideological control is being eroded by cross-border smuggling of radios, cell phones, and DVDs. Critically, this trend also exists within at least some segments of the North Korean military.

The economic tides are also washing the system’s foundations away. Pyongyang looks richer and a few people are growing very wealthy, but the cost is growing corruption and loss of economic control. (Me: one thing I’ve concluded from the North Korean example and others is that poverty doesn’t cause revolutions; jealousy does. Class envy is far more dangerous to Kim Jong Un than famine was to Kim Jong Il.)

Other threats to stability include a decaying infrastructure, a weak currency, corruption in the security forces, and “anxiety” in some quarters of the security forces due to recent purges (the very things that the World Bank sees as signs of stability). To Bennett, those purges also signal the leadership’s dissatisfaction with much of its power structure. Refugee flows out of North Korea are probably our best indication of broad popular discontent.

Preventing anarchy and humanitarian disaster:

Once the central government begins to collapse, North Korea will be like no anarchy we’ve seen in recent times. Different military units could end up fighting each other, armed bands of criminals, or occupation forces from the U.S., South Korea, or China.

Bennett sees information operations as critical to the restoration of order (this may be the point I agree with most emphatically; for more on this, read chapter 4, starting at page 103). We should be talking to the North Korean people about things like reunification, transitional justice, prosecution of crimes against humanity, and property rights. We should be doing more to counteract the state’s xenophobic indoctrination. We (or rather, the ROKs) should also be reaching out to North Korean corps and division commanders about their critical decisions when things start falling part. Read the Hamhung example on page 192 to see what he means.

He has some other very interesting ideas for using information operations for non-military deterrence, starting at page 133. Just read them on your own; they’re well worth it. I particularly liked how Bennett thinks Lee Myung Bak should have responded to the Yeonpyeong Island shelling, and his suggestions for how South Korean propaganda could influence North Koreans effectively, assuming the South Korean left would ever allow that. What I’d like to know is how this could ever be broadcast into North Korean homes from a strictly technical perspective.

The problem of stabilizing North Korea isn’t just that some North Koreans would resist the collapse of the old regime; the greater problem may be anarchy. The infrastructure and public health are broken. Masses of people will leave their homes, either because of hunger, economic disruption, or to flee conflict. Mass migration creates the potential for famine and epidemics. North Korea will need a massive infusion of humanitarian aid to prevent this, and information operations will be key to persuading people to stay in their homes, farm their land, and wait for humanitarian aid to arrive (again, information operations will be essential). Bennett has a lot of very worthwhile thoughts on restoring social stability, starting on page 121.

The best news is that there will be plenty of work for lawyers in post-apocalyptic North Korea. They’ll have to sort out complex questions about post-unification property rights, the detention of North Korea military personnel, what categories of criminals should be prosecuted or amnestied, and the logistical and evidentiary burdens of detaining and trying criminals fairly.

Lest that lighten your mood too much, Bennett devotes several pages to the problems of seizing North Korea’s prison camps to prevent a massacre of potential witnesses:

It will take some time (perhaps at least weeks) for sufficient ROK and U.S. forces to reach as far north as the major political prisons, although the smaller prisons might be handled by air assault, if sufficient helicopters could be committed and the air defense environment mitigated. Specific ROK and U.S. forces should be allocated to each prison and a plan established to deliver them to the prisons as rapidly as possible.

This potential for delay in reaching the political prisons raises questions about alternative approaches. For example, the prisoners significantly outnumber the guards but yet generally do not act against the guards because of the known consequences. To give the prisoners a chance to survive, should ROK and U.S. airpower be used to attack guard weapon facilities, guard towers, and other places guards use in controlling the prisoners? Should airpower be used to open holes in the security fences keeping prisoners in the facilities and to kill the power at camps to disable electrical fences and lights used at night for monitoring the prisoners? Should ROK and U.S. special forces be sent to the camps either to assist the prisoners in escaping or to assist the prisoners in seizing control from the guards? Under such circumstances, are the prisoners more likely to survive by rebelling against the guards? Or will such a rebellion just give the guards the excuse they want to exterminate many of the prisoners? All these actions carry considerable risk but potentially no more so than waiting for weeks before dealing with each camp, thereby allowing the camp officials time to exterminate many prisoners. Should the camps be closely monitored and these interventions begun only when extermination efforts start? Or would delaying that long give the guards too much of an advantage in killing the prisoners? All these factors need to be considered and trade-offs analyzed to determine preferred courses of action. [pp. 237-38]

Either that, or we can just let the Chinese go in, bulldoze all the evidence, and put the guards in charge of a factory where child prisoners work in small unventilated rooms full of toxic fumes 18 hours a day, making Elmo plush dolls. I’ll let you read the rest on your own. He even flatters this humble blog with a footnote cite. There’s another long passage about dealing with North Korea’s human rights disasters, starting on page 223.

Risks of Chinese intervention and conflict

Bennett worries, with good reason, that ROK, U.S., and Chinese forces will end up streaming into North Korea from opposite directions, and that the result will be either a re-division of Korea or war. The Chinese refuse to talk to us about post-collapse planning, which feeds distrust on both sides and amplifies that risk. In a fast-moving situation, the North Korean bitter-enders you think you’re shooting at might just turn out to be ChiCom special forces.

Refugee flows into China could also cause Chinese forces to enter North Korea to establish a “buffer zone,” and this could evolve into a permanent division and a long-term source of conflict. There are other motives, too:

China might also be interested in reaching [WMD] facilities because evidence of Chinese assistance to the North Korean WMD program might be found there, evidence that could taint China’s global image.

China might decide to go further and race the ROK and the United States to Pyongyang. Control of Pyongyang would give China a strong bargaining position for postcollapse negotiations, should it decide to leave North Korea. Alternatively, if China decides to sustain a North Korean government and abort Korean unification, control of Pyongyang would give it a means of legitimizing the North Korean puppet government it could decide to install. Chinese control of North Korean territory north of Pyongyang would give it free access to ports on the East Sea/Sea of Japan and the ability to exploit most of the North Korean mineral resources, resources that are now expected to be substantial. According to one observer, “unencumbered Chinese access to North Korea’s minerals, labor and ports would fuel China’s ever-growing economy. Through the calculated distortion of history, China is therefore being proactive against scenarios on the Korean peninsula it dislikes.”

Most of the North Korean military faces south along the DMZ, not north along the Yalu. Little would stand in the way of the Chinese as they raced south. In the event of such a race, China would end up occupying most of North Korea’s territory and population, which would cause an outbreak of nationalist rage in South Korea, and more potential for conflict.

Another risk Bennett cites is that North Korea would start a “diversionary” war with South Korea to distract discontented factions from fighting the system, possibly using sleeper agents carrying out terrorist attacks in South Korea or Japan.

In other words, a North Korean collapse presents a significant risk of a long, dreary counterinsurgency, a war with China, or both, but good planning, diplomacy, and information operations could do much to mitigate those risks.

The ROK Army can’t stabilize North Korea alone

Bennett doesn’t believe the South Koreans have nearly a large enough army to restore order in North Korea, in part because they continue to reduce its size by reducing the term of conscription. Bennett sees the answer as improving the size and quality of the ROKA reserves.

(Me: another part is building a professional ROK Army. Unlike Bennett, I’m dead-set against more than a very brief and limited U.S. role in restabilizing North Korea. Yes, go in with generators, tents, blankets, and medical clinics. Grab the WMD before anyone call sell them off. Liberate the camps before the prisoners are killed off. Then get out. Obviously, without the hundreds of thousands of Americans the ROKs are planning on having by their side, the manpower problem is a lot bigger.)

Bennett doesn’t see how anyone can re-stabilize North Korea without the help of whatever North Korean military and police units remain intact. That brings us back to the importance of information operations.

Some recommendations:

– “To deal with China’s concerns, the ROK could explain to China its plans for unification, and the United States could commit to China that U.S. forces would not be based on North Korean territory after unification. The ROK and the United States could use an information operations campaign to communicate this information to China. But note that such an information operations campaign needs to begin now, well before a government collapse, and then be adjusted as circumstances develop.”

– “Currently, there is also no bilateral, interagency organization to prepare and implement a strategy to address these challenges.” (Me: Logically, that organization should be USAID.)

– “Former U.S. Secretary of Defense Robert Gates indicated that he would send primarily air and naval forces to handle Korean contingencies, instead of the U.S. ground forces that would normally be involved in handling a collapse.”  (Me: I wish Gates was still SecDef, for a lot of reasons. The answer to any personnel shortfalls, as the author indicates, is improving the ROK reserve forces, coopting intact portions of regime security forces, and doing info ops to prepare the NK population to support the reestablishment of order.)

My thoughts:

Bennett’s analysis makes a lot of sense to me. It’s exactly the kind of thing that people in Washington read, agree with, fail to do anything about, and wish they’d taken more seriously years later, after it’s too late. It would be easier to identify the few things Bennett says that I don’t agree with and work from there.

Unlike Bennett, I’m not sure a civil war in North Korea would be worse than the status quo. Reform is a dead dream, and a bloodless coup is unlikely. A split in the security forces, followed by a civil war leading to regime collapse, is probably North Korea’s only way out of a tragic situation that has probably killed millions of people. How bloody that war is could depend on whether U.S. and South Korean governments give early and strong support to an opposition that’s willing to replace the current system with something more to our liking. This administration got that mostly right in Libya, but completely blew it in Iran and Syria, which means that both situations are now much worse than they had to be.

I don’t concede that a Chinese entry into North Korea can ever be a good thing, under any circumstances. Sure, it would be awfully swell if Xi Jinping agreed to occupy, stabilize, and rebuild half of the world’s most broken country free of charge, and then leave as soon as he’s asked politely, but it isn’t very likely.

I agree that existing ROK forces may be insufficient to stabilize North Korea, but I don’t follow that to agreement that U.S. forces should occupy the North. The national mood in the United States wouldn’t support it, and putting foreign troops on North Korean soil would greatly amplify popular opposition to the new regime, which would require even more troops. South Korea’s population is twice that of the North. The ROK Army had better get ready to draft enough South Koreans to do the job. The good news is that at this point, a lot of North Korean soldiers and police would probably work for anyone who gave them decent and steady pay.

The cost estimates of occupying the North run into the trillions of dollars. For comparison’s sake, the annual budget of the South Korean government is $250 billion. (Me: Yes, it’s probably worth some expenditure by U.S. taxpayers to help a successful reintegration of North Korea, but other nations with interests will have to help, too. Time to work on mending fences with Japan.)

Although it’s certainly possible that “North Korea’s WMD capabilities make WMD proliferation to a terrorist group a potential consequence that could lead to another potential consequence, terrorists using WMD against the United States,” to me, that consequence is more likely with the regime and its increasingly reckless proliferation networks intact.

One other point of emphatic agreement relates to the importance of diplomacy with China. With China, however, you need to bring both positive and not-so-positive incentives to the conversation. We have plenty of leverage, if we decide to use it, including recognition or non-recognition of Chinese investments and debt holdings, a commitment to contain refugee flows inside North Korea, access to North Korean ports and railways in Hamgyeong Buk-do, a commitment to keep U.S. forces and bases out of North Korea, or a commitment to the eventual removal of U.S. ground forces from South Korea. (The United States should not support Chinese demands that Korea be neutral or non-nuclear. Those issues should be left up to Koreans to decide.) And of course, if China just decided to occupy and digest part of North Korea anyway, South Koreans would probably be mad enough to support some fairly inflammatory anti-Chinese information operations, or even to provide material support for North Korean resistance to a Chinese occupation.

Since the famine years, North Korea has been a system on the defensive, glued together by a combination of fear, awe, pride, habit, and exhaustion. Fear probably still holds for the most part, but fear is a fragile bond and, without the reenforcing fibers of awe and pride, it would likely shatter in the event of a war, mutiny, or popular uprising. Koreans on both sides of the DMZ are nationalistic and proud of their nation and culture, but one would think that this pride is transferrable to whichever Korea government and society that holds the greatest military, economic, and cultural power.

The awe is probably about spent. Many North Koreans still hold a residual faith in Kim Il Sung, the old guerrilla who built a nation that once had (thanks to generous Soviet subsidies) a decent standard of living. I can’t imagine that his morbidly obese twenty-something grandson evokes a similar reaction with his inspection visits to water parks and ski resorts, or with his public friendship with a towering foreign laughingstock like Dennis Rodman. There can’t be much awe under a system where corruption is as rife as it is now. North Koreans will hold Kim Jong Un in contempt if they see him as having abandoned his grandfather’s aura of dignity, national superiority, and rigid principle. If the leaders aren’t faithful to the core principles, why should anyone else respect the system?

Eventually, the stopped clocks of most Korea watchers will be right. The fear will shatter, and that will be the beginning of the very painful death of North Korea.


  1. The Chinese have been stating through academic papers (which drip of Chinese nationalism) that Korea has largely been, since the days of Koguryo, a Chinese vassal. While the factual evidence behind some of these papers is thin, what is apparent to me and most others is that they are attempting to establish a historical precedent for Chinese control of the Korean Peninsula. Even the Chinese name for Seoul, ?? (Hancheng), means something analogous to ‘city of han [Chinese]. The swift movement of Chinese forces to occupy and claim as much as their own, or to establish a puppet state is an inevitability in my mind. I think the Chinese have been planning for this occasion a lot longer than the US and ROK have.


  2. Great post. Good God, how long did it take to write?

    Hypothetical: let’s say war breaks out on the peninsula. Let’s say that Kim Jeong-eun and a few of his cadres flee the country early in the conflict, both to save their own skins and, maybe, to continue prosecuting the war as leaders-in-exile. Do you think KJE would be hunted down, Osama-style, as a war criminal, or is it more likely that, as a head of state, he would enjoy the protection of diplomatic immunity? We went to war with Japan and didn’t execute or imprison Hirohito, for example (of course, Hirohito didn’t try to flee his own country, either, but I suspect KJE might). Osama, on the other hand, was responsible for the deaths of almost 3000 Americans, which immediately made him the object of a manhunt. Saddam, meanwhile, was a head of state but was also treated as a criminal who needed to be brought to justice; we eventually found his ass cowering in a spider-hole. What status do you think KJE would hold in the eyes of the US and the rest of the world? Note that I’m not assuming that all countries would view KJE the same way. My fervent wish, of course, would be for KJE to swing from a lamppost or to be put up against a wall and shot at Christmastime, a la Nicolae Ceausescu. What a great gift to the world that would be.


  3. I’m no expert, but this report by Rand seems speculative to the point of meaninglessness. I read 65 pages and then couldn’t take it anymore. Each speculation results in a covey of many more speculations.

    Obviously, North Korea is an utter enigma and, if the regime/government falls, there will be absolute chaos. Who can prepare for that?


  4. I have read the entire report, and applaud your synopsis. Personally, I think Mr Bennett’s analysis is too hopeful because the DPRK will not just dissolve into civil war and whimper away. Instead, when the conditions at home become intolerable, it will unleash a bang, an atomic bang, somewhere else and seek to take advantage of the external (and anonymous) confusion and distress that the explosion has caused to compel reunion with the South. You don’t need a missile to deliver a bomb, just a boat or a truck.

    There’s far more likelihood of the North taking over the South than the other way round.


  5. As someone with no cultural identity or history in the region, what would be the problem with China taking control of large parts of NK and footing the bill for modernising the place. With the direction China is heading politically and economically would it be such a bad thing to have NK split in two east/west-Germany style as a step towards unification, with the two major competing interests (those centered around USA/SK and those centered around China) sharing the bill?

    USA/SK cant afford to modernise the place alone and Japan’s economy isnt exactly in great shape. I dont see how USA/SK would be able to convince China to share the tab if they were outmanouvered and kept out of the political/military solution. The threat of waves of NK refugees flooding over the border conjures images in my mind of 1’000’000 Chinese AK47s pointed across the border daring someone to try to cross.

    I apologise if im missing something that is obvious to everyone else but it seems like having Chinese money would go a long way to easing the burden.