Famine in North Korea: Markets, Aid, and Reform
Marcus Noland, Devin T. Stewart

Although the famine conditions have eased ten years on, we estimate that roughly a third to perhaps a half of the North Korean population remains chronically "food insecure." And we argue that the marketization of the North Korean economy, which has occurred over the last fifteen years, is less a product of a top-down attempt by the North Korean government to achieve particular economic or political goals, but was rather an unplanned, unintended development that arose out of the trauma of the famine and fundamentally out of state failure.

When the state, which had organized both the production of food and its distribution, was not able to fulfill its obligation, small-scale social units—households, workplaces, local party organs, small-scale military units—were all forced into entrepreneurial coping behavior, and the marketization that began with the food economy then spread throughout the economy. The result is that the famine, which started out as a kind of classic socialist famine, where access to food was determined by political status, has now morphed into a chronic food emergency, similar to what one would observe in a market economy in which access to food is determined by one's ability to command resources in the market.

The state's response to these developments has been fundamentally ambivalent—in some ways legitimating them and ratifying them, in some ways trying to channel them, and even in some ways trying to reverse them. Indeed, it is that last one, reckless actions undertaken over the last two years, that have brought North Korea possibly to the brink of another humanitarian disaster. Indeed, at the end of March, Anthony Banbury, the Asia Bureau Chief of the World Food Programme, announced that North Korea was a million metric tons short of food and millions of people could go hungry unless donors increased their donations. I will discuss that possibility at the end of my remarks.

But what I would like to do is go back and basically now just walk through the book and those three parts: first, the origins of the famine; secondly, the ethical dilemmas for the humanitarian community; and finally, this legacy and what it may mean.

The origins of the North Korean famine lay in the state's misguided attempt to achieve an understandable goal of national food security through a misguided attempt at self-sufficiency. The state banned private markets. It controlled both the production and the importation of food, as well as its distribution, through a quantity rationing system called the Public Distribution System, or PDS.

Now, given the inauspicious growing conditions in North Korea—very high ratio of population to arable land, relatively high northerly latitudes, short growing seasons, limited opportunities for double cropping, and so on—the only way the North Koreans could maximize output was to maximize yield over a given acre. In order to do that, they constructed a very industrial input-intensive form of agriculture. It is a form of agriculture that used a lot of chemical fertilizer, a lot of insecticides, pesticides, and electrically driven irrigation. So, it was an agricultural system that was highly dependent on the health of the industrial economy.

When the industrial economy began to falter in the 1980s, outputs and yields began falling, as one can see in this chart on the left. So, depending on which source of data you look at, production probably peaked around 1988 or 1990.

Now, faced with the dwindling output, the secular decline that we see in this chart, the state fundamentally had three choices: one was to export more, earn foreign exchange, and then purchase grain; second was try to borrow money on the international market to purchase grain; or the third was to compress domestic consumption. Regrettably, the state chose the third option, inaugurating a "let's eat two meals a day" campaign and cutting state rations.

In the spring of 1995, with the country experiencing famine, North Korea appealed initially to Japan, then South Korea, and eventually the United Nations for support and assistance. The first shipment of aid was headed towards North Korea in May of 1995. Later that summer the country was hit by floods. This proved to be politically advantageous both to the government of North Korea as well as for some in the donor community, who could portray the famine as a product of bad weather rather than a product of bad policy. This chronology, then, is very important. The decline in output and the famine preceded the floods rather than was caused by the floods. Now, the floods were certainly not helpful, but they were a minor contributing factor at best.

One of the most disturbing things we document in the book was that in the mid-1990s—1995, 1996, 1997, 1998—as aid began ramping up, North Korea systematically cut the amount of grain it imported on a commercial basis.

This is a chart where we have indexed total imports and food imports. As you can see in the right-hand panel, food imports fall disproportionately, they fall more than total imports; and then, as the economy begins recovering in 1999—and today aggregate imports are higher than they were in 1993—food imports never come back, and they remain minimal today. In effect, rather than using aid as a supplement to the supply of food, North Korea used it, in effect, as balance-of-payments support.

Now, as I said, there are basically two parts to the description of famine and food availability. The first is aggregate availability and the second one is distribution. In terms of aggregate availability, we show in the book that if North Korea had simply maintained its imports at that level of 1993, then normal human demand could have been met throughout the period. We have a whole series of charts in the book of this sort, where we have various sorts of counterfactual calculations.

This line shows normal human demand. This shows minimum human needs. This is the line you would have gotten if they had just kept the imports at 1993 levels rather than cutting them. Now, even accepting the fact that they cut imports, the actual available food line always lies above the minimum human needs. So if there are people who are perishing from a famine, distribution has something to do with it.

Now, the way that all urban North Koreans received their food was through the Public Distribution System. We can see from the data that, even on paper, from 1995 on the public distribution never delivered the minimum human need. In fact, in recent years the Public Distribution System has been delivering about 300 grams a day.

Now, some of you may wonder how much 300 grams of rice or corn is. Let me show you. That's 300 grams. I don't believe there is anybody in this room that could last very long on this ration—and we all have desk jobs. Anyone doing manual labor, who is actually using their body to work, cannot survive on this ration of food. And indeed, sadly, many did not.

The result was a famine with pronounced geographic, socioeconomic, and demographic components. For those of you who may be listening through the podcast, I have put up a slide that has two images. One is a multicolored map of the country and the other is an image of an emaciated North Korean child being measured by UN workers in 1997.

These two images are supposed to convey two points. The first is that the effects of the famine were not uniform geographically. In general, the southwest of the country was the best off. The rust-belt industrial cities of the northeast were the worst off. This data is from 1997, and we can't see that because the data for all these provinces is missing. I'll come to that in a second. At the time, the North Korean authorities would not allow the UN workers into these provinces. Subsequently, they gained access. In fact, we could document that this was indeed the worst affected. And then, of course, in demographic terms, the very young and the very old were the worst hit.

As I mentioned at the outset, we estimate that about 600,000 to a million people died. That would be about 3 to 5 percent of the pre-crisis population. More broadly, the World Food Programme (WFP) data implies that, for instance, in 1998 North Korean seven-year-olds were 20 percent shorter and 40 percent lighter than their South Korean counterparts. So it's not just death; there is widespread malnutrition as well.

Indeed, one of the things we do in the book is we actually look at the data going all the way back to the Japanese colonial period. If you accept these data as being accurate, they suggest that this group of North Korean seven-year-olds measured in 1998 were smaller than any cohort of Korean seven-year-olds in recorded history, going all the way back to 1910 when the Japanese colonial authority started keeping these records.

This disaster posed an enormous challenge for the international humanitarian community. In trying to ameliorate this emergency, the humanitarian community faced a fundamentally hostile environment. The North Koreans would not allow normal assessment and monitoring activities. As a consequence, the World Food Programme and the private NGOs that effectively piggybacked on their protocols were forced into a kind of second-best solution of targeting institutions where they thought there would be people in need—so targeting orphanages, elementary schools, hospitals, and so on. Indeed, they ended up with a list of more than 40,000 such institutions throughout the country.

Unfortunately, more than ten years into this program, the North Korean government has never provided the WFP with a comprehensive list of all these institutions. Nor is the WFP allowed to use Korean speakers. No Korean speakers, no ethnic Koreans, are allowed to be in the country. And generally, pre-notification was required to visit any of these institutions, most of the time around one week. So the WFP was reduced to using less than 50 people to run a food program that was supposed to be targeting 40,000 institutions in a land area the size of New York state or the state of Louisiana, none of whom could speak Korean, all of whom were dependent on government-supplied drivers and interpreters, despite the fact that at its peak the international humanitarian aid program, in principle, was feeding a third of the population.

Now, under such conditions, with such large volumes of aid going in—think about it, enough food to feed a third of the population, less than 50 people working under very difficult circumstances—there was ample opportunity to divert aid away from its intended recipients.

Let me say two things about this. Normally, people think of diversion of aid as kind of being a centralized conspiracy in which the aid is diverted from less-deserving people to some other group, typically the military. That may have happened, although we argue in the book that the incentives and opportunities for diversion were probably much greater lower down the food chain, to use a bad metaphor, that it was local officials and local actors who really had both the opportunity and the incentive to divert.

The other thing is oftentimes people when they think about the aid being diverted, they stop thinking at that point. But the aid didn't vanish into the ether. It was consumed somehow. One of the things we argue is that in this context, in the context of an economy where markets have been eradicated, the availability of food aid, the weak monitoring, and the opportunities for diversion actually had a profound and ambiguous impact.

The implicit value of this aid was astronomical, because remember this was a country experiencing a famine, so the real price of food is very high. If one could divert the aid and sell it, one could get rich. But in order to recognize or realize those profits, you need markets.

So, ironically, powerful people within North Korean society had an incentive not only to divert aid but to encourage the development of markets. And indeed, we estimate that by the latter part of the 1990s and the early part of this decade, the typical North Korean household was actually getting most of their food through the market. That was the institutional mechanism by which people were actually getting food. With the state having trouble procuring food in the countryside, the Public Distribution System had basically become a mechanism for distributing aid. The marketization that began with food spread to other products. We spend a fair amount of time detailing the institutional mechanism through which this marketization occurred, much of it oriented toward China.

The system, fraying in growing desperation over this period, was also associated with an intensification of illicit activities, which have received a fair amount of press in recent weeks, especially in connection with the Banco Delta Asia case and the six-party talks on nuclear disarmament.

In 2002, the government, in essence, acknowledged the marketization of the economy which had begun a decade earlier, in some ways ratifying and decriminalizing much of the coping behavior that had occurred, but in other ways trying to control or even reverse it.

The inexpert ways that the reforms were implemented, however, have contributed to high levels of inflation and widening inequality within North Korea. The industrial proletariat of those old industrial cities, especially on the east coast, have been the hardest hit, with World Food Programme household surveys indicating that some families were spending 40, 50, 60, 70, as much as 80, percent of their household budgets on food.

This is reflected in changing pathways to status and power. As one person said to me, it used to be that party cadres and state officials were the preferred sons-in-law; now it's military officers and entrepreneurs.

So in the context of several years of improving harvests, in the fall of 2005 the government of North Korea undertook a series of actions that could only be described as reckless. First, they banned the private trade in grain, in effect criminalizing the primary institutional mechanism through which most North Korean households actually obtained food. Second, they engaged in confiscatory seizures of grain in the rural areas.

Let me stop for a moment and do a brief aside on this because I think it is very important. There was a history of confiscatory seizures in North Korea. They have happened periodically in response to urban shortages. In 1995, the government went through the normal procurement process during the fall. What happened was the farmers on the cooperative farms were given an allocation which they get to keep, which is supposed to feed them and their families for the remainder of the year. So the farm families get an annual in-kind allocation, and then basically the rest of the harvest is sold to the state at this derisory procurement price.

What happened in 1995 was the state went through the procurement process and then realized it didn't have enough food to make the PDS work. So it sent the army back into the countryside to essentially try to extract from the farmers the food that had already been allocated to them for their own consumption for the next calendar year.

In 1996, fully one-half of the corn harvest disappeared. Now, the initial explanation was flooding, and indeed there was flooding in North Korea in the summer of 1996. But if you look at a map of North Korea showing the counties where there was flooding and the counties where they grow corn, flooding can't explain the disappearance of half the corn harvest, just like flooding in 2006 can't explain the fall in recorded output.

Steph Haggard and I actually wrote an op-ed in the International Herald Tribune in December of 2005 saying that the North Korean government was now possibly on a trajectory towards a renewed humanitarian disaster in the spring of 2007 because the farmers were going to protect themselves in the next harvest cycle.

To make matters worse, at the same time they were trying to ban the market, revive the Public Distribution System, and engage in predation against the cultivators. They, in effect, tried to throw out the humanitarian aid groups.

The World Food Programme, which had had something on the order of 45 to 50 people in the country, not only in the capital city of Pyongyang but also in a series of five regional sub-offices, had those offices closed; the number of staff reduced to, I think, seven—less than ten people—and confined to the capital city of Pyongyang, who were only allowed to leave the capital city once every three months on supervised field trips. So, to use an American idiomatic expression, they took the canary out of the mineshaft. They took away our early-warning system because this is how we had a lot of information about actual conditions outside of Pyongyang in those industrial cities and in the countryside.

So those three things—inaction on the demand side, inaction on the supply side, and then the removal of information—set up this possibility of renewed emergency this spring.

Indeed, as I mentioned at the outset, last month Tony Banbury had a press conference and announced that the North Koreans were a million metric tons short and began sounding the alarm about a possible new food emergency. I say "possible" because the government of North Korea restricts information to such an extent we really don't know how bad it is. Some observers, such as the WFP, and the South Korean NGO Good Friends, take the position that this is a true crisis.

Yet, to the extent that we can imperfectly—and I want to underline imperfectly—observe prices in the markets that have in fact come back, prices for grain do not appear to be skyrocketing, which suggests that there are supplies, presumably either from China or through stocks that the farmers have hoarded, that are being sold into the market.

The real issue—and I don't want it to be lost in this welter of detail—is that the government of North Korea systematically impedes access and prevents the world community from assessing the degree of need and designing an appropriate response. It is fundamentally about the government of North Korea's policy.

So what do we take away from all this? Well, first, the famine and its aftermath are inseparable from the authoritarian nature of the political regime. Only a regime that systematically restricts all human civil and political rights, prevents the spread of information, prevents debate over policy, criticism of public officials, and basically is unaccountable to the populace could have acted with such culpable slowness in the 1990s and maintained such disastrous policies in the face of a humanitarian emergency, regrettably something that they may well be doing again today. In the end, only a regime that is accountable to the people, protects their basic rights to produce and to exchange, allows them to do what's best for themselves, will enable a permanent resolution of the food emergency.

Now, the economic form of that resolution would not be for North Korea to achieve agricultural self-sufficiency. That's a dream. The way that this crisis will be resolved in the long run is if North Korea's industrial economy is revived so that North Korea can export manufactured and mineral products, earn foreign exchange, and import bulk grains from commercial firms, just like South Korea does, just like Japan does, and just like China does.

The third issue, which I haven't discussed much, is that the political interests of the donors are divergent. Let me just draw one inference from that. The inference we take, that it is a strategy of attempting to strangle the North Korean regime and provoke regime change or collapse, is unlikely to work. The reason is quite simple: if the United States or the Japanese or, perhaps, the Europeans cut off aid in an attempt to do that, then North Korea's immediate neighbors, South Korea and China, which for understandable geographical reasons are more risk-averse to the possibility of chaos and instability within North Korea than countries that are further afield, will simply move to offset that by increasing aid. So there is a fundamental coordination problem here that we think undermines a policy of attempting to provoke regime change and solve the problem that way, by economic embargoes or sanctions.

As a consequence, we conclude that there is no real ethical choice other than to engage. But we should be clear-eyed about the terms on which that engagement proceeds and we should seek to further our own values whenever possible.

I would like to close with a metaphor, which is, like all good metaphors, at once both trite and profound. It was offered by Kang Chol-Huan, who himself is an escapee from the North Korean gulag. What Kang said was: "If you hold a cow by its hoof, it will starve. If you allow it to roam, it will find grass and eat." Ultimately, the solution to the North Korean food problem is a political regime in North Korea that allows the North Korean people to do what is best for themselves.

Thank you very much for your attention. I would be happy to try to answer any questions that you might have.

Questions and Answers

DEVIN STEWART: I totally agree with Marcus that engagement is probably the way to go. As I opened up, I think that it is the ethical path.

I was wondering if you could comment a little bit about perhaps the nature of engagement that would seem to work. And also, what's going on with South Korea and North Korea in terms of economic exchange and activity? If you ever follow Asian news, you read these interesting tidbits that get hidden from the mainstream press about roads being built and train tracks being laid and factories being opened in the border there. I'm just wondering if this is having any kind of positive effect.

MARCUS NOLAND: Those are two good questions. In fact, they actually lead into each other in a very nice way.

Let me give you, first of all, a little technocratic answer to the first question, engagement that works. I think in the case of North Korea we have to face the fact that we are dealing with a highly vulnerable population that has no real voice, that has no real control over the action of its governance. So it seems to me what we would like to do is try to ameliorate the suffering of that population. But, we want to try to do it in a way that, number one, is efficient and, number two, promotes our values and not the values of a government which I believe is at root the problem.

Let me give you an example, and it's an example from the Bush Administration. I would like to indicate that I am a former employee of the Clinton Administration, so this is not boosterism for the Bush Administration. When the Bush Administration came to office, it appointed a man named Andrew Natsios as administrator of USAID. Andrew had a deep interest in famine issues, and indeed had written a book on the North Korean famine. One of the useful things that the Bush Administration did under Andrew Natsios's leadership is put a requirement on U.S. donations to the WFP. The United States is the primary donor to the WFP program, or has been historically.

What Andrew said was that a significant share—I believe three-quarters—of the food aid needed to go into ports on the northeast coast where the worst problems were. The idea was, even if that food is stolen and diverted the moment it leaves the ship, markets in North Korea are not perfectly integrated, they're fragmented, and the fact of the matter is that food would be sold in the catchment area around that area.

So the first-best solution would obviously be that food aid go to the intended recipients, whether they be widows or orphans or children or whoever, but the second-best is that the poor families in that area buy the food in the market. So it is not necessarily the best allocation of food, but it is a kind of second-best way of dealing with it.

Another thing one can do is provide food aid in forms that the elite don't like to consume. So provide barley or millet instead of rice. Again, it reduces the likelihood of this aid being stolen or diverted into elite consumption.

So one can do things of that sort that increase the effectiveness of the humanitarian aid program even within these extremely difficult political constraints. I would point out that these sorts of tactics differ fundamentally from the food aid policies pursued by the governments of China and South Korea, which basically simply hand rice to the central government. This is one of the reasons that Steph and I have consistently urged the governments of South Korea and China to donate more of their aid via the WFP, because we believe the WFP protocols are fundamentally more effective.

We also note that if South Korea became a major donor to the WFP, it would be increasingly untenable for the government of North Korea to maintain its restriction of not allowing ethnic Koreans or Korean speakers to serve in-country. So you could increase the effectiveness of the program even further.

Now, the second question is about South Korea's engagement with North Korea. Indeed, I have a bunch of slides about this, although not as part of this PowerPoint, but if you're really interested, I guess I could probably put them up. Increasingly, North Korea is economically integrating with two countries, China and South Korea, but the nature of that economic integration is significantly different between the two.

Being only slightly facetious, when we look at the nature of economic integration between China and North Korea, we see that the firms and enterprises from the communist country are behaving like capitalists. But when we look at the economic integration between South Korea and North Korea, we see that the firms and enterprises from the capitalist country are behaving like the socialist tools of foreign policy.

What do I mean by this? I mean that while China does provide North Korea with aid, and while there are certainly forms of economic integration between the two countries that are politically determined, it appears that increasingly economic integration between China and North Korea is occurring on market-conforming terms.

The Chinese firms and enterprises that are going into North Korea are not charitable enterprises and they demand to be paid. So, it is a market. It is a Chinese-style market—maybe not a World Bank-style market, but it is a market that is developing.

In contrast, if you look very carefully at the economic integration between South Korea and North Korea, it has a very large official transfer component to it, whether it be outright subsidies, it be various forms of investment guarantees and insurance, and so on. So that when you strip out what is either straight aid or what is subsidy and so on, it appears that the amount of economic integration between North and South Korea that is really on market terms is rather small.

Now, I want to be clear—I have criticized both the governments of China and South Korea already this afternoon—that this is not purely because of South Korean policy. For understandable political reasons, it would appear that the government of North Korea is actually more comfortable with Chinese firms wandering around the North Korean countryside than it is with South Korean firms wandering around the North Korean countryside. So the South Korean economic initiatives have really been literally confined in a physical sense to these sorts of enclaves. Now, in the end this may be a first step, it may broaden out and so on, but right now I would have to say in some ways China is leading this process of globalization, so to speak, of North Korea, not South Korea.

DEVIN STEWART: Why is North Korea more comfortable with the Chinese firms?

MARCUS NOLAND: Because they are Chinese, they're not Koreans. Having South Koreans, who are going to be physically larger, obviously more robust, much better dressed, carrying fancy cell phones and all this kind of stuff, could have a degree of political impact in North Korea that perhaps Chinese in the countryside would not. Remember, North Korea still tells its people that South Korea is worse off than it is. This presumably increased exposure of the typical North Korean person with South Korean counterparts presumably would create a significant cognitive dissonance that might ultimately have significant political ramifications.

DEVIN STEWART: I'd like to open it up.

QUESTION: I have a couple of questions. Just to start, are the Chinese firms generally small scale, coming in across that northern border, rather than state-owned enterprises, I would assume?

MARCUS NOLAND: That's a very interesting question, and in fact it is a question that Steph and I are about to begin to look at. We are literally going to do survey work on exactly this topic. Our impression from our visits and our interviews and what data we've been able to gather thus far is that the economic integration takes a variety of forms. There is some straight aid. There is some what I would call politically determined transactions, where a large Chinese state-owned enterprise is told "go build a glass factory," right? But much of this trade appears to take the following form. I'll take a minute because it is an interesting though somewhat complicated story.

Let's start with the Chinese side of the border. There is a provision in Chinese law that gives certain firms in the border region the ability to transact what is called "border trade," which is basically tariff-free, tax-free. My understanding of the origin of the law is it was originally just to take care of all the normal, small-scale cross-border exchanges.

So what will happen is, say, a steel mill in Manchuria will identify a firm in the border region, say in Tumen City, and they will form a joint venture in order to get this legal status to do "border trade." Then, they want to import low-grade coal from North Korea. What North Korea is exporting to China at this point is mainly natural resource products, mainly minerals, and then some niche natural resource products, such as ginseng, sea urchins, things of that sort.

Now, on the North Korean side you have, let's say, a state-owned enterprise that used to make tractors, tractor factory number three. They haven't made tractors, at least in volume, in years. Nevertheless, that enterprise manager is told that he is responsible for covering his wage bill. Now, the thing that he has going for him is that as a large state enterprise he either has his own official trading company or he can do official trading through the ministry. So essentially what he does is write down a wish list: cheap televisions and radios, second-hand refrigerators, just all kinds of cheap and secondhand Chinese consumer goods.

He has his purchasing agent go into China. The deals are actually done in China so the Chinese counterpart has recourse to Chinese dispute settlement. He has his guy go into China and round up the cheap televisions and secondhand radios and refrigerators. This market is actually big enough that there are specialized transportation firms on the Chinese side. So you load it up on the truck, it drives to the border, there's a land bridge across the river, he drives to the North Korean side and stops. The truck is then offloaded. I talked to a Chinese owner. He would not allow his truck to go into North Korea. He would never get it back. He turns around and the North Koreans are responsible for hauling this stuff off themselves.

So then, having acquired these televisions and so on, he takes the former workers at tractor factory number three and sends them off to sell them. What I call this is the "Wal-Martization" of North Korea. The former industrial proletariat, which used to have high wages, good benefits, they have been reduced to marketing cheap Chinese imported consumer goods.

Now, how does he pay for it? The way he pays for it is he goes to his counterpart, who is the manager of the coal mine, and he basically does a deal with the coal mine guy to get the coal, which is then either literally bartered or used to generate cash which is then used to pay for the televisions and stuff. That stuff is all cash in advance.

So it is a complicated set of relationships and developments that are working out. Some of it is "two steps forward, one step back." You talk to the Chinese about this, you can hear some pretty bad stories about things that have happened to them in North Korea. But it's of that nature. It is market conforming. The North Koreans are having to pay them. They are having to give something up.

I was in southern Africa in the summer. I was working on a project involving diamonds. I used some of your [Devin Stewart's] work in that project. I was having lunch with the management staff and the union leaders from the world's largest diamond mine in Jwaneng, Botswana. These guys were complaining, as did their counterparts in South Africa, about their trouble keeping not only engineers but skilled workers, the guys who could repair big diesel engines and so on. The reason was that the Chinese-driven commodity boom, the worldwide commodity boom, was so strong that those skilled artisans, blue-collar workers, and mine engineers were being bid away for double and triple wages in projects in Australia and Canada.

I know that was a very long answer. I'll try to be shorter.

DEVIN STEWART: In the meantime, let me just do a quick follow-up so Marcus can rest his brain for a second. Let's talk about the book. I think we have copies back there.

MARCUS NOLAND: And I'll be here to sign them.

DEVIN STEWART: Very gracious.

We always hear these crazy stories about North Korea. I mean North Korea is great for crazy stories. I think there are some journalists here. I used to live in Tokyo and I worked for the Yomiuri Shimbun. One of our favorite things was to watch the ferry that would go back and forth bringing Asahi super-dry beer directly to Kim Jong-Il. So we would just keep track of this ferry going back and forth. The movie Team America is not that far from the truth.

Any crazy stories you could relate from your research or data on North Korea?

MARCUS NOLAND: Devin, my guiding philosophy on this has always been that the reality of North Korea is bad enough that one doesn't need to exaggerate, that in fact one should err on the side of being conservative. So, for example, let me take one specific example from the book, famine deaths. Estimates of famine deaths range from about a quarter-million people, which is the North Korean quasi-official figure, up to, I think, 3 to 3.5 million—somebody may have even claimed 4 million. When we analyzed the data, the most sober academic analysis one can do suggests 600,000 to a million people. That's 3 to 5 percent of the pre-crisis population. If that were to happen in the United States, that would be 15 million people.

In that situation, it may well be that someday, when in some sense the rock is lifted up and we can see what's scurrying around underneath it, it may well be the reality is worse than what we depicted. That is often the case in closed societies. But I always tell people there's no reason to exaggerate; the reality is plenty bad enough. Let me give you an example. It's not a crazy story. Actually, I can't resist. I will tell you one crazy story. Okay, here's a crazy story, but you get two for one in this story.

DEVIN STEWART: Very economical.

MARCUS NOLAND: I was in North Korea, and I got my minder alone, where what we were saying to each other could not possibly be monitored. We had a wide-ranging conversation that lasted several hours, a pretty frank conversation. At one point in the conversation, he mentioned in passing that Kim Il-Sung was god. I said, "Well, I'll grant you that, but how about Kim Jong-Il?"

I will never forget his response. The response was, "Kim Jong-Il is 75 percent god." The conversation moved on and I never clarified whether he meant that Kim Jong-Il was some kind of demigod or if that 75 percent of the people thought Kim Jong-Il was god and 25 percent thought he was something other than god. You know, this is a pretty odd story.

DEVIN STEWART: Or god is 175 percent of something?

MARCUS NOLAND: Several years later, I was at the George Herbert Walker Bush School of Government at Texas A&M University. As you know, the former President Bush was also director of the Central Intelligence Agency. There is a significant CIA presence at the Bush School of Government. In fact, I think they literally have a position called CIA Officer in Residence.

The President was hosting a conference on North Korea and I was one of the participants. One evening he hosted a public dinner. He recruited a man named John McLaughlin, who at that time was the number two man at the CIA, to give the after-dinner speech. The last time I checked, you can verify what I'm about to say is true, because it's on the CIA website, or at least it was last time I looked.

So John McLaughlin gets up, the number two guy at the CIA, to give this after-dinner speech. We're sitting there at these round banquet tables. There's several participants there. There's the man who's the head of the Blue Bell Dairy, which is a big ice cream company in Brenham, Texas.

John McLaughlin gets up and he says, "Ladies and gentlemen, tonight I'm going to talk to you about North Korea. North Korea is a very strange society. Let me give you an indication of just how strange North Korea is. Once a foreign traveler was traveling in North Korea and he got into a conversation with his North Korean counterpart. The North Korean counterpart said that Kim Il-Sung was god. 'That may be so, the foreigner replied, but how about Kim Jong-Il?' 'Seventy-five percent god' came the response."

I turned to the other people at the table and I said, "That's my story." But here's the scary part. Did I tell this story to people who may be working at the CIA? I'm sure I did. Did any of them ever come back to me and ask me if I was telling the truth or is this beer talk? I mean no one was asking, "When you talk about the guy saying Kim Il-Sung was god, were you really telling the truth?" This kind of a joking conversation somehow goes up the ladder and ends up in a public address by the number two man in the Central Intelligence Agency and posted on the website.

So when it comes to funny stories, like I said, this one's a two-for-one, it's a twofer. You know, I'm not sure who it says more about, North Korea or the United States.

QUESTION: The question I have is that following through and checking on the performance and whether the food is being distributed to these poor people in North Korea is a challenge, and it will always be a challenge. From what you say, it seems like at the state level, at least the NGO level, there hasn't been success. Have you seen a success level with some private enterprises, maybe small, that it seems like they are actually helping people and keeping the government accountable for distribution?

MARCUS NOLAND: That's a really good question. One of the things we mention in our book is that—let me step back for a second and give some context for this.

What happened in the case of North Korea was you had this famine emerge, it was an extremely closed society, very difficult engagement of the humanitarian aid community. What happened was eventually, in the mid-1990s, you had a situation where the primary in-country group providing food aid was the World Food Programme of the United Nations. There were a bunch of private NGOs that in quantitative terms were much, much smaller. More than 90 percent of the food was coming through the WFP, but there was this group of small NGOs from various countries, various types, some religious, some secular, and so on. They essentially piggybacked on the World Food Programme protocols. They actually formed something called FALU, or the Food Aid Liaison Unit, which was a way of aggregating the activities of these many small NGOs. So in that sense the NGOs didn't get any better deal than the WFP did in terms of access, the ones that were basically just working through the WFP channels.

However, there were other NGOs who managed to negotiate in some sense better terms of access, oftentimes by trying to maximize their contact with local-level officials, whose incentives in terms of monitoring, in terms of trying to keep up a smokescreen and so on, might have been different than the central government officials.

One of the things we conclude in our book is that some of these small private NGOs who had very small operations may have actually had more effective monitoring regimes in reality, even if they did not on paper, that they had in fact established enough essentially allies at the ground level among these local officials who were involved in a kind of repeated game with them that, although some of this was informal and maybe not down on paper, in effect they had a better idea of where their aid was actually going than some of the larger-scale official groups as well as NGOs.

Now, Devin asked for funny stories. I'll tell you this story, which is both funny and frightening. This is a quote from Marx or Oliver Wendell Holmes, depending on your politics: "History repeats itself; first it's tragedy, then it's farce."

I had more than one conversation, independent conversations, with NGO workers, some official, some from private NGOs, who told the same story. Remember North Korea is a mountainous country. These people don't speak Korean. They asked to be taken to a particular school or they asked to be taken to a particular public distribution center. The North Korean government gives them a car and a driver and an interpreter. They drive them around the mountains and they take them to this place. The next day they are supposed to go to another school or another public distribution center or another orphanage. They drive around in the same car. They drive around in the mountains and they take them to the place. I had more than one of these workers swear to me it was the same place they had been taken the day before, but were told that it was a different location. We laugh, but this is serious.

One of these people told me about just being horrified, realizing how effectively powerless she was to do any sort of real assessment or monitoring of her organization's activities. When she left North Korea and went back to her home country, she literally got one of those "how to speak Korean" books and simply concentrated on learning written Korean so that she could at least read the road signs when she was taken out the next time she was sent into the country.

She said they literally had no idea what county they were in, they had no idea what province they were in. They just had been driven all around the mountains. Especially if it's overcast, you can't see the sun, you don't even really know what direction you're going, and then being taken to what they thought were sort of "Potemkin village"-type setups and really being taken to the same one. It's like Groundhog Day—"No, really, this is a different school. You weren't here yesterday, believe me."

QUESTION: You had a very interesting line about the desirable son-in-law going from being party cadre to military or entrepreneurs, implying that the famine has helped shift more power towards the military. If you'd speak a little bit more about this, has it weakened Kim Jong-Il's political base, has it strengthened the military, and in which ways? If you would, talk a little bit more about that, please.

MARCUS NOLAND: That's actually a really deep issue. At the risk of being shamelessly self-promoting, I actually have a small monograph that addresses exactly this issue, called Korea After Kim Jong-Il. I think you can buy it used on Amazon for like $1.98. It makes a great cure for jet lag if you're flying to Asia.

One can conceptualize the famine having two effects. One is the economic effect, which I really addressed in my remarks, that it encouraged a marketization of the economy and eroded direct state power and control, helped create a kind of new class of entrepreneurs, and so on. So that's one part of the equation.

The other part of the equation was that period of the 1990s was also a period in which the state failed. It basically began failing in some fundamental ways. The Korean Workers Party, which had been a serious mass-membership communist party, began to wither in terms of its actual functioning as a communist party, as a mechanism for, for example, transmitting information from the ground up to the central leadership and being a mechanism to ensure discipline from the top down.

In that vacuum of governmental organs that simply failed to operate, a party that was withering away, the military, almost by default—almost in relief, as everything else receded—the military was the last institution standing. It had organization, it had resources, it had coherence.

As the economy deteriorated, the military, which had been operating its own economy—it had its own farms, its own factories, its own mines; it had its own trading companies, preferential access to technology, all sorts of things—it in some ways got sucked into the civilian economy because as things began falling apart, the government increasingly turned to the military to literally fix things—fix roads, fix bridges, harvest the grain, and so on.

That shift was then made manifest in shifting propaganda. The national ideology had been Juche, normally translated as "Self-reliance." But in the 1990s increasingly what we heard was "Military First." "Military First" was used as an adjective, it was used as an adverb, it was used as a verb, it was used as a noun. It was just increasingly the Songun ["Military First"] policy that became the banner.

Since you are from the Japan Center for International Exchange—and I know there are other people, including Devin, who have backgrounds in Japan, as I do—one can imagine this as being the way that North Korea in some sense could get itself off the Juche hook, that the syllogism would go something like this: We must have a strong army for national survival; a strong army depends on a strong economic and technological base; hence, any sort of reform justified in terms of improving the functioning of the economy or technological advance, no matter what kind of departure it was from past practice, could be justified under the "Military First" policy of national survival.

If you look at the propaganda and you look at the phraseology in the slogans, they bear striking resemblance to Japan during the Meiji Restoration. In effect, what Kim Jong-Il seemed to be wanting to do, or seems to want to do, is have a similar political revolution, where a sort of technocratic modernizers want to overturn past practices justified by the need to have military capacity in a world that is so threatening that it could actually threaten national survival.

The other example I would use is modern Turkey under Mustafa Kamal, where political revolution is basically done by the military and justified as national survival.

But in the Kim Jong-Il campaign—he has a theatrical bent, as we all know—he wants to play both the Tokugawa shogun and the Emperor Meiji in this production. He will play both roles and then somehow carry this off.

QUESTION: Thank you very much for such a thorough, in-depth, and vivid picture of the famine of North Korea. But I think that we are also concerned about today's food shortage situation. As far as I know, till now we just have two voices for today's situation in Korea. One is from WFP. One of the directors said that there is a million ton food shortage; and the other one is from you. I think you said that actually, North Korea is facing a severe food shortage, but not as severe as the one in the 1990s. Would you please elaborate a little bit on today's food shortage situation in North Korea? Thank you.

MARCUS NOLAND: To be clear, there are multiple voices. Tony Banbury, the Asia Regional Director of the WFP, did hold a press conference in Beijing last month, and he did say that they are a million tons short, that millions of people would go hungry. He went into some greater detail about birth defects and so on if donor donations did not ramp up. That analysis, that there is a genuine and severe shortage, is also supported by the South Korean NGO Good Friends.

There are also other NGOs from South Korea, such as NK Net, who have contacts within North Korea and do things like try to measure prices in North Korean markets, who are skeptical whether the shortage is real or not.

I think, to be perfectly frank—and I say this with all due respect to the World Food Programme and to Tony Banbury, and I think they do great work under very, very difficult conditions—but the fact of the matter is Tony Banbury doesn't know. He doesn't know how much food North Korea needs. How would he know? He's got seven people confined to Pyongyang. He doesn't know what's going on in the countryside. He doesn't know what's going on in Chongjung.

Good Friends may or may not know. NK Net may or may not know. The fact of the matter is the North Korean government systematically impedes access to the country that would allow anybody to do a persuasive assessment of the situation and design an appropriate response. So it may well be that this is 1993 and 1994 and that we are starting to go into another famine and this is just the early stages.

One of the astonishing things, if one goes back to that period of the 1990s—and we go over some of the history in the book, but we don't beat it into the ground—is just how little people knew at the time. We have personal friends who are very competent people who both visited North Korea at the same time and came away with very different impressions, partly because conditions varied a lot from region to region. They also varied a lot depending on what social class you were in. And so you could have two people there in the same week, but, depending on where you were, you might be seeing very different things.

And the same thing is true today. There is not one North Korean reality. There are many North Korean realities, depending on where you are physically, in terms of your social class, your access to foreign exchange, and so on.

Or it could be that we are in a situation like we were in 1995 after the floods, where the government of North Korea swings from downplaying distress to greatly exaggerating distress. Once you have made the strategic decision to say, "Yes, we are really in distress, we need help," then you may make the decision, "Well, once we admit that we are vulnerable, we might as well get all we can out of this" and exaggerate the amount of distress and get more aid. So it could be the case today that North Korea doesn't need a million metric tons of food, that the government is just saying it because it knows it's under pressure.

Who knows what they may be preparing to do in terms of the nuclear program further down the road, which could result in further sanctions further down the road? So this might be a good opportunity to get a million tons of grain, put it in storehouses, and then do something else that you know may result in further sanctions. Nobody knows. With all due respect to Tony Banbury and the WFP and Good Friends and NK Net, nobody has a good grasp of what is actually going on there. The reason is the behavior of the North Korean government.

QUESTION: I was just wondering if it is too farfetched to contend that famine or food shortage actually serves regime maintenance. How else otherwise to explain the timidity with which they have experimented with reforms, such as in Vietnam and in China? Is the paranoia based on that the entire regime is basically propped up by the big lie in a sense, that North Korea is actually more prosperous than South Korea, and that things could crumble very quickly if foreign influences and a different perspective were to creep in?

MARCUS NOLAND: I think that's a very, very good question. There are essentially two aspects of that question that I would like to tease out.

The first one is the sort of purely economic technical part of it, which is you often hear people, especially people from Asia, say that "Well, North Korea will adopt the China model." Usually, that's not defined exactly what that means, but the implication is that North Korea will adopt the China model and then it will start growing at 8 or 10 percent a year.

One of the things we argue in the book is that in structural economic terms North Korea today more closely resembles Belarus or Romania at the time that they started doing reforms than it does China or Vietnam, and that there are reasons to believe that a reform process in North Korea would have a very different economic and political trajectory than the reform process in China or Vietnam, which, after all, had more than 70 percent of their labor forces in the agricultural sector when they formally initiated it in the late 1970s and late 1980s, respectively.

The other issue, of course, is the one that you raise, the political issue and the legitimization challenge that South Korea poses. Again, I think the experiences in China and Vietnam are very instructive in this regard.

Let's start with Vietnam because that's the easier case. Vietnam had a civil war. One side won. They became the monopolist definers of what it meant to be Vietnamese. So in the late 1980s, when they decided to undertake reform, they realized that Hanoi could come up with justifications, more or less tortured, about how market area reformers were what Uncle Ho had in mind. Fortunately, Uncle Ho had been dead long enough that he wasn't around to comment.

The Chinese case is a little different. You've got that government in Taipei, but I think for most people they would say that doesn't pose an ideological threat to the government in Beijing. So in 1978-1979 Chinese ideologues were free to come up with stories about black cats and white cats and slogans and all this to justify that market-oriented reform is what Marx and Mao really had in mind.

The problem for North Korea is much different because of the divided nature of the Korean Peninsula. North Korea is clearly the junior partner in both size and accomplishment. As you suggest, the economic reform it seems will start to move them closer towards South Korea and erode what is distinctive about North Korean society. That could pose a fundamental legitimization challenge for the regime. Why be a third-rate South Korean when you can be the real thing?

In that monograph I mentioned, Korea After Kim Jong-Il, I go into a lot about this, and trying to think through exactly what are the transitional paths that a future non-Kim-Jong-Il-led North Korea could look like. I think that this basic fact of life poses a real problem for would-be reformers in North Korea.

They have one advantage, and it's an ironic advantage. The dynastic nature of the North Korean state gives Kim Jong-Il a lot of leeway to legitimate what he wants to do. He can always say—I mean look at today—the nuclear talks are being justified in terms of Kim Il-Sung's dying wish, that Kim Il-Sung's dying wish was for a nuclear-free Korean Peninsula. Now, one would have thought that simply not building nuclear weapons could have assured Kim Il-Sung's dying wish without having to build them and then give them away. But, you know, that's another talk.

But Kim Jong-Il is in a completely privileged position to say that "We need some sort of policy, we need to do something with the economy. ...Well, one night my father and I were visiting Shinujoo and we were walking along the beach that evening and he said to me, 'Whatever.'" You don't have to say that Jesus came to you in a dream last night and told you to do it. I mean you just say, "Well, in 1984 my father said, 'If you're ever in this situation, this is what you do.'" So he can justify all of this that way and claim continuity with the past and continue to claim legitimization through that source of authority, even if he tries then to shift the basis of the regime.

I think that this is a good place to end. The thing that we draw away from this, or the thing that I draw away from this, is that this is an extremely risk-averse regime. They behave absolutely ruthlessly when it comes to their core political goals narrowly defined. They will make concessions under duress, but then they try to grab those concessions back when conditions change.

If this is the way they behave towards 50 do-gooders handing out free food, can you imagine the kind of paranoia and attention that is going to surround a serious attempt to monitor and verify some sort of denuclearization regime? That's one of the big lessons from this book. If you look at how they behaved towards these people handing out food, can you imagine what they are going to do when we start sending in arms inspectors? It's going to be a very tough row to hoe.

DEVIN STEWART: Thank you, Marcus. Thank you very much.