Introduction

Causes of the Crisis

The Economics of the Famine

Death Rates

International Food and the Famine

Consequences of the Famine

Political and Security Consequences

Coup Plot

Population Movements

Communicable Disease

Conclusions and Recommendations

The central questions for U.S. policymakers

Timeline

About the United States Institute of Peace

About the Report

SPECIAL REPORT 51

The Politics of Famine in North Korea

Andrew Natsios

Briefly...

• Because of the withdrawal of USSR and Chinese food subsidies in the early 1990s and the cumulative effect of collective farming, food availability in North Korea declined steadily and then plummeted between 1995 and 1997 when flooding followed by drought struck the country.
  
  
• From 1994 to 1998, 2-3 million people died of starvation and hunger-related illnesses, and the famine has generated a range of social and political effects.
  
  
• Beginning in 1994, the central authorities appear to have triaged the northeast region of the country by shutting down the public distribution system. In 1996, they appear to have begun selective food distributions to people in the capital city, workers in critical industries, and party cadres, leaving the rest of the population to fend for itself in the private markets.
  
  
• Refugees report that the famine has undermined popular support for the current political leadership. In 1995, a planned coup by military officers was uncovered by secret police in Hamhung, the city most devastated by the famine. Public anger is more typically reflected in growing corruption, black market activities, and other anti-system behavior.
  
  
• International food aid has stimulated private markets, reduced the price of food in the markets 25-35 percent, and undermined central government propaganda concerning South Korea and the United States.
  
  
• Those who have died or suffered most during the famine have been those unable to adjust to the economic reality of these new markets either by growing their own food or by producing some other marketable product, labor, or service to exchange for food.
  
  
• Reduced purchases of Chinese maize in 1999 will force higher prices in the private markets. Given the fragility of the private food system, and absent international aid, this situation may plunge the mountainous regions of the country into a new round of famine-related deaths.

Introduction

While it is widely acknowledged that the collapse of the North Korean economy has caused a severe food crisis, the severity and political implications of the crisis have been a source of considerable dispute. A substantial body of new evidence indicates that the country has been experiencing a major famine with abnormally high mortality rates since 1994.

The food crisis did not begin with the floods in August 1995, as has been commonly understood, but with the sharp reduction in heavily subsidized food, equipment, and crude oil from the Soviet Union and China in the early 1990s. This reduction precipitated an agricultural and industrial decline of enormous magnitude. As output fell, the central government initiated a "Let's eat two meals a day" campaign to ration diminishing food supplies. A family of defectors to South Korea in 1994 reported that elderly people were going out to the fields to die to relieve their families of the burden of feeding them (Samuel Kim, The Foreign Policy of North Korea, 1998). In their study of North Korea's demographics, Nicholas Eberstadt and Judith Bannister speculate that the underreporting of elderly female deaths in official population figures may have been a function of families wishing to continue to receive their relatives' rations even after they had died (The Population of North Korea, 1992).

The massive floods during August 1995 led to a central government appeal to the World Food Program (WFP) for food aid. This natural disaster and a series of successive droughts and floods over the next three years are responsible for about 15-20 percent of the food deficit facing the country, the rest being attributable to collectivist agricultural policies [WFP/Food and Agriculture Organization (FAO) crop assessment, 1995].

Most of the food production estimates of the North Korean harvest fail to include post-harvest losses attributable to poor storage, rot, rodents, and insect infestation. WFP/FAO reports have gradually increased this estimate from 0 percent in 1995 to 6 percent in 1996 and 15 percent in 1998. Russian scholar Marina Yi Tribuyenko estimates the North Korean loss rate to be comparable to the annual Soviet loss rate of 30 percent ("Economic Characteristics and Prospects for Development" in North Korea: Ideology, Politics and Economy, edited by S. Han Park, 1995).

An acrimonious policy debate has been taking place within humanitarian organizations about the severity of the famine--indeed, its very existence--and the role of international food assistance in ending it. The questions being raised in the debate are not new; they reflect legitimate concerns about the effect of food aid to a country where those with political authority may have objectives very different from those of humanitarian agencies trying to reduce death rates.

North Korea is notable, even among its former eastern bloc allies, for being the most controlled and reclusive society on earth. Discerning what is actually happening in such a society is no easy matter. Much of the reporting on the famine has been based on visual observation by humanitarian aid workers who have visited or worked in the country. While these visits do provide some information, neither field visits nor the data provided by the North Korean government about the food situation are conclusive evidence of anything because they present conditions as the central authorities wish them to appear to the outside world.

In addition to field reports, seven other sources provide valuable anecdotal and empirical evidence about North Korea's food crisis: United States Institute of Peace Special Reports by Scott Snyder ("A Coming Crisis on the Korean Peninsula?" and "North Korea's Decline and China's Strategic Dilemmas"); defector interviews; a book by the preeminent defector Hwang Jong Yop (North Korea: Truth or Lies? 1998); research by scholars of North Korea; the collected reports of Jasper Becker, reporter for the South China Morning Post, and of Hilary Mackenzie, a Canadian journalist with unusual internal access; four studies based on refugee interviews (two very large surveys of more than 2,000 interviews and two smaller nongovernmental organization surveys); and the speeches of Kim Jong Il, particularly his speech of December 1996, which provides exceptional insight into the dynamics of the famine. I visited North Korea in June 1997 and the Chinese border with North Korea in September 1998 to interview 20 food refugees through meetings arranged with the assistance of a South Korean nongovernmental organization (NGO). I have attempted to cross-check information with at least three independent sources before concluding that it is accurate.

Causes of the Crisis

The Economics of the Famine

Traditionally, famines have been principally economic phenomena with political and public health consequences, not vice versa. In totalitarian regimes in which economics is subordinated to ideology, famines can be politically driven. Despite manipulation and control, who lives and who dies is ultimately determined by microeconomic forces affecting specific regions, ages, incomes, and job groups differently, complicated by local food market prices. Macroeconomic perspectives offer less insight.

As food becomes scarcer in a famine, its marginal value increases exponentially. Beginning in 1995, North Korea's central authorities reduced the grain ration for farm families from 167 kilograms per person per year to 107 kilograms, which was insufficient to live on. Reducing farm family rations proved disastrous to the food distribution system because it instantly changed the economic incentive for farmers. This reduction had three pathological consequences for the North Korean food system:

• It broke the social contract between farmers and the rest of society that farmers would grow food in exchange for industrial production from the urban and mining areas. Under the traditional arrangement, the state and collective farms took control of the harvest each year and then gave an allotted ration back to the farmer and his family. Because of the economic downturn after the withdrawal of Soviet oil and food subsidies, the system could not produce the pesticides, fertilizer, and herbicides needed to supply the farms, further reducing agricultural production. With the farmers' rations reduced, any incentive to provide food to industrial areas disappeared.
  
  
• It encouraged farmers to divert production from the agricultural system before the harvest. The sudden drastic reduction in the food ration meant that farmers were given the choice of letting their families starve or secretly preharvesting and saving food to build up family stocks before the harvest was actually taken. Kim Jong Il complained bitterly about this secret preharvesting in a speech (December 1996). One defector said that he had seen reports of the roofs of many farmers' homes collapsing under the weight of hidden grain. The fall 1996 WFP/FAO agricultural assessment acknowledges that half the corn harvest was missing--nearly 1.3 million metric tons (MT). This hoarding began an undeclared war between the central authorities and the individual farmers. According to defector and refugee interviews, soldiers--called corn guards--were dispatched to protect the fields as the harvest matured in an effort to prevent this enormous diversion. However, this command and control tactic failed when farmers simply bribed the soldiers, hungry themselves because of a breakdown in the military distribution system, to join them in the diversion.
  
  
• It encouraged farmers in mountainous maize-growing areas to spend their time, expertise, and energy cultivating the private plots of land the government gave them decades ago to grow vegetables for household use, and on cultivating secret plots of land in the mountains outside the control of the collective agricultural system. Though technically illegal, these secret plots, called fire fields, can be seen across the country on mountaintops so steep and infertile that it is difficult to imagine that crops can grow there. Soldiers have traditionally helped plant and harvest crops. In the maize-growing areas, soldiers now appear to have taken the place of farmers, because farmers are spending so much time on their private plots (refugee interviews, September 1998; Natsios interview with Democratic People's Republic of Korea official, June 1997).

The rise of farmer's markets and the shutdown of the public distribution system for the nonfarm population made food inaccessible to families that had no way of paying higher prices, particularly when the economic downturn reduced the purchasing power of families. As industrial production plummeted, salaries of urban and mining workers were either reduced or stopped entirely, as was the heavily subsidized food from the public distribution system on which these workers had relied for 50 years. The public distribution system was not a social service system, but a means of workforce compensation. The nonsubsidized price in the farmer's markets for a kilogram of maize (not enough to feed a family for a day) equaled an average industrial worker's monthly salary, while under the subsidized rate a month's industrial salary easily bought the maximum ration allowed through the public distribution system. Thus, these urban families sustained a major reduction in family purchasing power at the same time the price of food increased exponentially.

According to a Johns Hopkins University study of 440 refugees, 39 percent of the people in the far northern region of North Hamgyong province rely on farmer's markets as their principal source of food (through either barter or cash purchases), while only 5.7 percent rely on the public distribution system ("Rising Mortality in North Korean Households Reported by Migrants to China" by W. Courtland Robinson, Myung Ken Lee, Kenneth Hill, and Gilbert Burnham, Lancet, July 1999). Surveys by the Korean Buddhist Sharing Movement (KBSM), a South Korean NGO working in the North Korean-Chinese border area, show a similar shift from the public distribution system to markets in other provinces. A Republic of Korea (ROK) Unification Ministry study using defector information indicates that there are 300-350 farmer's markets in North Korea and that people get approximately 60-70 percent of their food from them.

The phenomenal increase in the frequency, selection of products, and size of these markets over the past four years has been noted by United Nations (UN) and NGO workers. Kim Jong Il has publicly attacked these markets as unsocialist (December 1996). He tried to shut these markets down after his father's death but was forced to rescind the order because of urban unrest (defector interview, September 1998). He tolerates them now because they are essential to the survival of the cities. His father, Kim Il Sung, had wisely sanctioned the farmer's markets, saying that they would exist one way or the other and that it was better to have them out in the open so they could be regulated, though it is clear even Kim Il Sung never envisioned that these markets would become this large and essential (Kim Il Sung, "On Some Theoretical Problems of the Socialist Economy," March 1969). Without these markets, the urban areas would be even more depopulated than they are now. The rise of these farmer's markets amounts to a de facto privatization of an important part of what remains of the North Korean economy. The regime's embarrassment may be one reason why officials have prohibited any expatriate visits to observe these markets or to study food prices in them.

The people who have died in the famine or who have suffered the most deprivation are those who were unable to adjust to the economic reality of these new markets either by growing their own food or by producing some marketable product, labor, or service that they could exchange for food.

Famines usually evolve in phases along a timeline that affects different socioeconomic groups and geographic regions at different points depending on access to harvest surplus and the strength of coping mechanisms. Food distributions from the public system in the northeast became intermittent in 1992-1994, effectively stopping during the summer of 1994, though two to three days' worth of rations are still distributed six times a year on national holidays (Jasper Becker, Hungry Ghosts, 1998; KBSM, "Survey of North Korean Refugees,"1997; and Robinson, Lee, Hill, and Burnham, July 1999). In 1994, the central authorities coped with the sharp decline in food availability by triaging the four eastern provinces (North Hamgyong, South Hamgyong, Rangang, and Kangwon), which are politically and militarily less important to the survival of the central government than the western provinces, and by shutting down any food shipments there from other regions (defector interviews, Sept. 1998; merchant and refugee interviews, Sept. 1998; UN staff interviews, Jan. 1999). This triaging also occurred as the regional economic system, under which the northeastern industrial cities and mines had sent their products to the rice-growing areas of the west in exchange for food, fell apart. Thus, the famine began in the northeastern provinces two years earlier than in the western provinces.

In 1996, the central authorities made a momentous decision to deal with the famine by decentralizing authority for feeding the population from national bureaucracies in Pyongyang to county administrators. If county administrators were particularly skillful and energetic, fewer people died; if bureaucratic and lethargic, the effect of the famine was acute (Sue Lautze, "Independent Food Observer to the PRC & DPRK Final Report," June 1996). In January 1998 a decision was announced through the government apparatus that each individual family was henceforth responsible for feeding itself, rather than relying on the traditional public distribution as before the famine or on the county administrators since 1996 (defector interview, September 1998). This decision may have simply recognized the evolving reality rather than deliberately decentralizing decision making. Earlier, Kim Jong Il deplored decentralization and the idea of individual responsibility, which he said would make people less dependent on the state and party, endangering socialism in North Korea (December 1996). Despite his misgivings, he seems to have decided in favor of this new system by 1998, probably out of administrative necessity.

The economics of the famine may be entering a new phase in 1999. The Chinese maize crop in provinces north of the border was quite poor because of exceedingly wet and overcast weather, with reductions in harvest of 70-90 percent (Korean-Chinese merchant and agronomist interviews, September 1998). Extensive flooding damaged the general harvest in other regions of China. These developments have already begun to affect barter exchange rates on the border. In September 1998, North Korean timber was trading for reduced amounts of Chinese corn. The remaining timber is less accessible for cutting and scrap metal from cannibalized industrial machinery which could be bartered for corn by the North Koreans is exhausted. Thus, in 1999, commercial purchases of Chinese maize will decline, increasing the price of grain in the farmer's markets in the northern provinces. Without international aid, this situation, combined with the fragility of the new private food system, may plunge the mountainous regions of the country into a new round of deaths.

Death Rates

The outside world will likely not have access to conclusive mortality rate data until the regime falls or internal changes cause embarrassing information to be released to the public. However, we do have enough information now from very different sources to make some rough estimates of the severity of the crisis.

Hwang Jong Yop, former party ideologue and philosopher of juche, published North Korea: Truth or Lies in June 1998 following his defection to South Korea in February 1997. In it he writes the following:

In November 1996, I was very concerned about the economy and asked a top official in charge of agricultural statistics and food how many people had starved to death. He replied, "In 1995, about 500,000 people starved to death including 50,000 party cadres. In 1996, about one million people are estimated to have starved to death." He continued, "In 1997, about 2 million people would starve to death if no international aid were provided" (Chapter 20, p.15).

In interviews with the South Korean media, Hwang has subsequently estimated that the death toll has reached 2.5 million people since 1995. (He suggests that one million died in 1997.)

A second source of information on death rates comes from the KBSM, which between September 30, 1997, and November 1, 1998, interviewed 1,679 food refugees who moved to China to escape the famine. The survey employed the same interview techniques used by international human rights organizations to gather information from refugees escaping abuse. The study found that the mortality rates among the family and ban members who had not moved to China (the ban is the lowest level of North Korean society and is composed of 30-50 families) were 26-28 percent. These rates cannot be applied against the entire population because everyone is not equally at risk (indeed, many are not at risk at all because their political power allows them to accumulate resources to ensure their survival) and because this is not a random sampling of the population. However, because of the very large number of interviews, their depth, and the collection of data over time, the KBSM survey does offer telling insights.

The refugees interviewed in the KBSM survey are overwhelmingly urban workers from factories, the transportation sector, or the mines. Farmers are poorly represented in the population of refugees crossing the border, though they represent 25-35 percent of the population. These data confirm the reports of Pyongyang officials, NGO and UN reports, and defector interviews that the famine is centered in urban and mining areas (other than the capital city of Pyongyang, where a minimum food ration has been maintained). Some maize-growing mountainous areas suffering localized crop failures from flooding (1995 and 1996) or drought (1997) have also experienced high death rates. Hwang's report that 10 percent of the deaths in 1995 were of party cadres [who, according to Eberstadt and Bannister (1992), make up 15 percent of the total population], indicates that middle and lower-level cadres are also suffering.

A team from the Johns Hopkins University School of Public Health has completed a second study of famine death and birth rates in North Korea based on interviews in China with 440 food refugees from North Hamgyong province (Robinson, Lee, Hill, and Burnham, July 1999). The study shows that from 1995-1997 death rates rose eightfold over what they had been in pre-famine North Korea, rising from 0.55 percent in the 1993 census to an annual average of 4.3 percent during each of the three years of the study. During this same period, the number of people reporting the public distribution system as their principal source of food declined from 61 percent in 1994 to 6 percent in 1997, confirming earlier findings by KBSM, Jasper Becker, and my own interviews. Births declined by 50 percent from 21.8 per thousand in 1995 to 11 per thousand in 1997-another famine indicator. Even though the authors are reluctant to extrapolate their figures to the general population of North Hamgyong, they do estimate that 245,000 out of a population of 2 million people may have starved to death in that province alone.

How severe has the famine been in North Hamgyong province compared with other provinces? North Hamgyong has three advantages not enjoyed by other provinces. First, travel and trade restrictions across the North Korean border with the ethnic Korean region of China are markedly weaker than in other regions, allowing hungry North Koreans to migrate in order to work and to beg and borrow food from relatives in China. Second, barter for food between North Hamgyong and China is extensive. Third, the province traditionally produces larger agricultural surpluses on the farmlands along the Tumen River than do the rest of the mountainous eastern provinces, according to a United States Agency for International Development/Office of Foreign Disaster Assistance (USAID/OFDA) map of agricultural production. As a result, North Hamgyong province likely experienced lower, not higher, death rates than other mountainous provinces without these coping mechanisms.

The death rates in North Hamgyong may be conservative on another count as well. The Hopkins study, like the KBSM study, considers only family members of interviewees who died at home in North Korea. The study did not account for deaths of family members who left home before the interviewees themselves fled to China, because the interviewees could not know the fate of relatives who left home before them. This likely means that mortality rates have been underestimated, because experience from other famines shows that the death rates for people displaced by a famine rise precipitously compared to the death rates for famine victims who remain at home, if those displaced cannot find work or humanitarian assistance. In these cases, the food refugees are more vulnerable because they lose the protection of their home, family, and neighbors. In North Korea, economic collapse and the highly centralized distribution of humanitarian assistance have left most food refugees with few options except fleeing the country. No one can be certain how many food refugees have died along the way. Thus, a large number of deaths of hungry migrating people are not recorded in either the Hopkins or KBSM surveys.

Other studies show that the famine is not limited to North Hamgyong. The WFP and UNICEF nutritional surveys show malnutrition rates similar to those of the Hopkins study across the country. The first nutritional study of children under five years old, done in September-October 1998 by UN agencies, the European Union (ECHO), and Save the Children Fund/UK, shows acute malnutrition and wasting (of body mass) of 18 percent and stunting of 62 percent. These are alarmingly high rates, particularly considering four other factors: (1) for the past year WFP has been concentrating its feeding program on children under age 7; thus, their condition should have improved; (2) the survey did not include abandoned, internally displaced, or refugee children who are much more vulnerable to malnutrition; (3) central authorities excluded 30 percent of the population from access by the survey teams (in what I surmise are militarily sensitive areas); and (4) the peak of the famine took place during the last half of 1996 and the first half of 1997 (a year before the children were measured), at which time many of the most vulnerable children died. This study appears to have underestimated the severity of the crisis again.

The KBSM study plotted the home villages of the 1,679 refugees interviewed on a map of the 211 counties and then averaged the death rates by county, which were once again similar. The only exception is Pyongyang, which in the KBSM study showed a death rate one-third lower than the other counties--the lowest in the country. In both the Hopkins and KBSM surveys, death rates varied by age, gender, and profession rather than by geography.

The Hopkins death rates, therefore, are a consistent, though conservative, estimate of the severity of the famine across the country. If anything, extrapolating the Hopkins death rates to the country as a whole understates rather than exaggerates the famine's severity. By deducting the 2 million people living in Pyongyang (the senior party cadre who have suffered least in the famine are heavily concentrated in the capital) and the 1.2 million soldiers, and then applying the 12 percent Hopkins rate of mortality to the 20 million people remaining, the total number of famine deaths would approach 2.5 million, the same number of deaths claimed by Hwang Jong Yop. Thus, we have several independent studies that give credence to Hwang Jong Yop's estimate.

International Food Aid and the Famine

Sizable amounts of food have been imported into North Korea either as food aid, subsidized commercial sales, or cross-border barter trade between 1995 and 1999 primarily from China, Japan, South Korea, United States, and the European Union. As reported by WFP, food imports from all sources totaled the following amounts:

Why have so many people died after so much food has been delivered? While it is true that major contributions or sales were made by the Chinese, donor governments, and private organizations, the timing of the pledges and deliveries did not match the peak period of the famine. Second, the central authorities seem to have panicked at the poor crop in the fall of 1996 and the lack of donor governments' pledges or delivery of food until late spring and early summer 1997, which resulted in a shutdown of the public distribution system. Third, the central government saw to it that triaged populations did not have access to imported food because they were not seen as critical to the survival of the state. The drop in death rates appears to have taken place because the vulnerable urban and mining population had either died down to a sustainable level or moved to agricultural areas or to China. Some people found new income sources to purchase higher-priced grain in farmer's markets.

The international food aid provided privately through NGOs, the Red Cross, bilaterally from the European Union, and through WFP seems to have had several unintended but ameliorative outcomes.

• Privatization of markets. In some areas, food aid, diverted from the aid program by corrupt officials who sold it on the urban markets, has stimulated the size and robustness of the urban markets, a form of unplanned and unintended privatization. People are no longer dependent on the state for their food supply, thus undermining a principal means of control. Donor food aid has appeared for sale in many of the markets in South Hamgyong provincial cities. (The maize for sale in the northern border city markets is Chinese, while the rice in the southwestern urban markets appears to be locally grown, according to refugee interviews.) Refugees who have seen the food in South Hamgyong markets remark that the maize and rice for sale are in donor government bags (from the United States, European Union, and ROK) and that the grain in them is not grown in China or North Korea (refugee interviews, September 1998; Medicins Sans Frontieres interviews, 1998).
  
  
• Food price stabilization and reduction. The volume of international food aid and Chinese imports is such that according to refugee interviews, the price of food in the private markets diminished by 25-35 percent between March and September 1998--the last half of the harvest year--just when it should have been increasing. The volume of food on the markets has reduced the price sufficiently so that families with limited assets can purchase or barter for food; they can get one-third more food for the same price.
  
  
• Diversions undermine regime support. Refugees from South Hamgyong know that countries they have been taught are their enemies are giving food aid. As one refugee from Hamhung City told me, "We were taught all these years that the South Koreans and Americans were our enemies. Now we see they are trying to feed us. We are wondering who our real enemies are." When asked why they had not received the food aid free through the public distribution system, refugees repeatedly replied, "The corrupt cadres [or bureaucracy] are stealing the food and selling it on the markets for their own profit while we starve. We see it there for sale." Medicins Sans Frontieres refugee interviews provide similar testimony as does the Washington Post investigation based on refugee interviews (February 11, 1999). Thus, the inability of the regime to feed its population and the presence of food aid in the markets are undermining popular support among those without political power. Given the nature of the regime, this public dissatisfaction is not reflected in overt opposition but in growing corruption, black market activities, sabotage, and other antisystem behavior that reflects public cynicism and anger.

Food aid has reduced the malnutrition rates among some of the population under seven years old in school settings, the more recent focus of WFP distributions along with a food-for-work program.

The larger question is whether this food program has kept the North Korean government in power. Kim Jong Il clearly does not like the aid program and has attacked it publicly in an official speech, "On Preserving the Juche Character and National Character of the Revolution and Construction" (June 19, 1997): "The imperialist's aid is a noose of plunder and subjugation aimed at robbing ten and even a hundred things for one thing that is given." The food aid program is visible evidence of the failure of juche, the governing state ideology; it has undermined state propaganda about the outside capitalist world; and it has accelerated the privatization of the economy. Perhaps this is one reason the military opposed the initial food program in 1995 and forced the temporary shutdown of the program in 1996. Thus, the food aid program is undermining state ideology rather than propping the system up, just as Paul Bracken suggested outside aid would do in his "poisoned carrot" article ("Nuclear Weapons and State Survival in North Korea," Survival, Vol. 35 No. 3, Autumn 1993).

Consequences of the Famine

Political and Security Consequences

North Korea has sustained more destabilizing change over the past five years than it has over the previous 40 years combined. The arrival of 100 expatriate humanitarian relief workers at a high point from mid-1997 to mid-1998 and, with them, the exposure of the internal problems of the country to the international news media, have made the central authorities very uncomfortable. For the most part, these changes cannot be revoked by the central authorities, and they have shaken an already teetering system.

If the death rates are as substantial as suggested, a significant portion of urban families have seen members die. Nearly 40 percent of the 16-24-year-olds in the country (or 6 percent of the total population of the country) are in the military (Eberstadt and Bannister, 1992). These are exceptionally high percentages, comparable only to a general mobilization during all-out war. Under other circumstances, this percentage of people under arms would mean a strong base of popular support for the military and a high level of political mobilization in the society. Under famine conditions, the reverse would appear to be true. A popularly based military of such enormous size means that a large proportion of the military will have seen their parents, brothers, and sisters die. Thus, the regime has a large number of young men with weapons, albeit in a highly controlled and disciplined organizational structure, who are likely to be unhappy about deaths in their families. Such anger in the ranks in China in 1963 contributed to the end of the disastrous Great Leap Forward famine (Becker, 1998).

Coup Plot

In The Two Koreas, Don Oberdorfer reports that in early fall 1995 in the northeastern region of the country, "the Sixth Corps of the North Korean Army ... was disbanded, its officers purged, and units submerged into others" (p. 375). A defector confirmed this incident as a coup plot (planned but never attempted) by a corps-level army unit in Hamhung City in fall 1995. Hwang Jong Yop reports that 500,000 people died from the famine in 1995, most of whom would have died in early fall at the end of the agricultural harvest year as food stocks ran out. According to refugee accounts, the worst famine-affected city in North Korea was Hamhung City, the country's largest industrialized city with the highest proportion of factory workers, and coincidentally the headquarters of the corps-level unit that planned the coup at the peak of the famine. The coup plot thus appears to have been driven by the famine sweeping across the northeast.

During the peak of the famine in late 1996 and early 1997, Kim Jong Il began purging general officers who had dominated the military establishment for decades and replacing them with younger officers who presumably were more loyal. Unconfirmable media reports claim that martial law was imposed in early 1997, again as the famine was peaking. Perhaps Kim Jong Il realized that he was at risk because of the coup plot in late 1995, which in turn was a result of the chaos and death caused by the famine.

The purges and the coup plot suggest that the famine has already had a convulsive effect on the political system. While the newly promoted general officer corps may be loyal to Kim Jong Il, it is not clear that field and company grade officers share the same loyalty. The KBSM survey, though clearly of a disaffected population, shows that refugees attribute the famine far more to poor political leadership than to natural disasters.

Beginning in September 1998, the regime began to reimpose order in the country beginning with the transportation system (refugee interviews, September 1998). Internally displaced people were no longer allowed onto trains without travel permits and paid tickets, which had not been the case since 1995. One refugee who had been across the border five times said that security was "six times more severe" than what he had experienced since the start of the crisis. In addition, authorities imposed new fines on unauthorized population movements, issued new internal identification cards in June, and instituted new measures to increase security in the "927" detention centers for displaced people.

Critics argue that the United States and other donor nations should starve the North Korean regime into collapse. Yet the famine has already taken a large number of lives, and the regime, though shaken, is still in power. It is important to remember that no Communist government has been overturned during or after a famine, though coups or popular revolts have frequently replaced authoritarian governments after famines. However, the Communist record in this century may not be applicable in the case of North Korea, because other Communist regimes experienced famine early in their histories when revolutionary fervor was at its peak, and they maintained control of the food distribution system so that dissenters would starve if they were not put down by internal security forces. The North Korean regime has de facto privatized the food distribution system; its revolutionary fervor has diminished over time and by the famine itself; and it has no allies left to save it in the event of a military coup, which is the only serious threat to the survival of the regime. Given the likely anger in the military over famine deaths, Kim Jong Il must now fear the military as his greatest threat, which is perhaps why he avoids visiting military units that are engaged in exercises using live ammunition (U.S. government sources).

Population Movements

Famines are accompanied by widespread population movements as people attempt to cope with their hunger; these generally occur in the latter stages of the crisis. North Korea is no exception to this pattern of behavior, though the population control system appears to have constrained some of these movements (unlike Africa where the population is more nomadic and national boundaries do not constrain movement). Estimates of the aggregated migration to China from North Korea over the past several years ranges from 100,000 to 400,000, though not all of these people are in China at the same time. The figure refugees quote most often from party cadre sources is 200,000 people (refugee interviews, September 1998; Becker, 1998, p. 330). People move back and forth many times. Given the extraordinary measures taken by the North Korean and Chinese authorities to prevent this refugee migration (they are building a string of detention centers to house captured North Korean refugees before returning them to North Korean authorities) and given the difficulty of moving through the mountain ranges of the northeast to get to the border area, the size of the migration is impressive evidence of the severity of the deprivation in North Korea. Traditionally, anyone caught escaping was executed. The more usual penalty now is a beating, the confiscation of all belongings and money earned in China, and confinement for several months before being sent to more permanent local prisons for former displaced people.

The figure of 200,000 comes from party cadre members, probably based on data collected in the population census in the summer of 1998. The mass population movements, coupled with the national elections held in 1998, caused the government to issue new identification cards for all North Koreans, which meant that anyone in China who harbored any intention of returning home needed to get this new identification card or face the possibility of permanent exile. So the refugees came home and were counted and carded, and in the process the authorities likely could identify who had moved and who had not.

Next to the starvation itself, the most politically insidious consequence of the famine is population movement. Party propaganda has claimed that, despite their problems, people in North Korea were better off than in China, where civil war, epidemic, and famine rage (a credible claim during the Great Leap Forward and the Cultural Revolution). The refugee movements across the border into China have given lie to this propaganda as the prosperous reality is evident. One refugee from Hamhung City put it well: "Our first border crossing is a grammar school degree, the second time you visit China is a high school diploma, and the third and fourth trips are college and graduate degrees in reality. They have been lying to us all these years." The decision to encourage population reentry from China for registration and voting in the national elections damaged the regime. It contributed to the further erosion of its base of popular support as these refugees with graduate degrees in economic reality returned to their home villages to tell their families and friends what they had seen. Refugees confirmed this transfer of information.

Two KBSM refugee interviews provide fascinating data on the voter turnout for the summer 1998 elections, which were depressed by the devastation of the famine. In one county, part urban and part rural (rural areas have far less migration, as people are eating relatively well), the number of people who voted (and everyone must vote) was 68 percent of what it had been in the elections before the famine. In another mining city, the vote was 54 percent of what it had been previously. These disparities likely represent both population movements and famine deaths.

On September 27, 1997, Kim Jong Il ordered all of the county administrators in each of the 210 counties to set up facilities, appropriately called "927 camps" (the date the order was made) to forcibly confine those who were caught outside their village or city without a travel permit. The Stalinist permit system had kept most people in their home villages most of their lives. However, it effectively controlled the population movements: people did not receive their food ration unless they were in their home village. As the public distribution system collapsed, these regulations became far less effective at controlling population movements as people no longer relied on the state for food.

Many refugees in China were captured by police and spent time in these squalid 927 camps (usually located in the county hotel) and then escaped back to China. The camps cram 40-50 people into a room. There is no heat in the Arctic winters, sanitation is terrible, food is sparse, and infectious disease and death rates are high. Refugees report that each camp holds between 300 and 1,500 people, so at any one time between 63,000 and 315,000 people countrywide are housed in these camps before they are forcibly returned to their home villages. Although no comprehensive study has been done of turnover rates in the camps, they are intentionally high. Anecdotal information suggests that average stays range from several weeks to two months. Extrapolating these population figures at a turnover rate of six times a year translates to a minimum of 378,000 and a maximum of 1.9 million people per year passing through them. Given that many internally displaced people avoid capture while others die as they move to areas that may have no more food than where they left, the people in these camps do not reflect all internal population movements. The very existence of the camps is testimony to official concern about displaced people wandering around the country. In his December 1996 speech, Kim Jong Il complained of chaos and anarchy in the countryside, which may be a veiled reference to these population movements.

Communicable Disease

The health of the general population is poor and worsening. All famines are accompanied by epidemics of communicable diseases because the immune system deteriorates as malnutrition becomes more severe. People regularly die of disease well before they die of actual starvation. In North Korea the problem has been complicated by a sharp downturn in the health care, water, and sanitation systems. Hospitals for the general population have no western pharmaceuticals, relying instead on traditional herbal drugs, and have no food to feed patients. According to a provincial survey by Oxfam, much of the public water system, when it functions at all, is of such poor quality that it causes health problems.

Conclusions and Recommendations

North Korean military incidents against Japan, the United States, and South Korea may be an effort by Kim Jong Il to focus the attention of his military, whose loyalty he doubts, on an external threat that he himself regularly provokes. Kim Jong Il opposes economic and agricultural reforms needed to end the famine because he sees the threat these reforms pose to his control and the threat he feels from his own military as greater than the consequences of the famine.

Donor governments, UN agencies, and humanitarian organizations have been too willing to accept the geographic distribution plans of the central government, even though it is increasingly clear that they have been based on political, not humanitarian, objectives. Donor governments should insist that their food be distributed where it is most needed, particularly to the mountainous regions of the country. Given the lack of reliable information coming from inside the country, donor governments should provide support for more studies of the famine based on refugee and defector interviews.

The central questions for U.S. policymakers are these:

• Will Kim Jong Il exercise power in a ruthless enough manner to restore his control over the population, the military, and the cadres to ensure his own survival following the chaos of the famine? The regime is attempting to reimpose order and control after a period of chaos. Political indoctrination campaigns will yield fewer and fewer results because growing numbers of families have watched loved ones die and the continued suffering of survivors has embittered the population. Public cynicism toward the regime is irreversible and is reflected in growing corruption, sabotage, an expanded black market, and other antisystem activities. Central authorities will be forced to rule exclusively by terror and repression, a fear expressed by Kim Jong Il (December 1996).
  
  
• Has the famine peaked and is now receding, having killed off the most vulnerable population, or is it migrating to other areas of the country? The famine may be migrating from the northeastern mountainous provinces to the rice-growing regions to the southwest (merchant and refugee interviews, September 1998). The fragile new food security system may be easily disrupted by price fluctuations, market failure, or the inability to barter or purchase food in these new private markets.

International food aid, however distorted its distribution has been, has had several beneficial consequences that would be lost if it were terminated. Donor governments should consider expanding the food program by initiating a food monetization program in port cities, particularly on the east coast. WFP would sell food aid on the informal markets to reduce prices to an affordable level, but not so low that they would discourage increased production by farmers. The local currency generated could be used for mass employment programs in industrial cities and mining areas to increase urban worker income. Expatriate monitoring of food sales in urban markets, assisted by expatriate translators, should be a required part of the program. Donor governments should avoid using the public distribution system because it is no longer functional in most of the country and serves purposes other than simply feeding the population.

• Even if reducing or terminating international food aid caused the regime to collapse, would the resulting chaos from such a collapse improve or diminish the prospects for a peaceful transition to a unified Korea? The chaos resulting from a collapse of the regime would threaten far more lives than those the famine has thus far claimed and could create an unpredictable military situation as well. Population movement to South Korea or China could prove explosive as both countries simultaneously take military measures to restore order, increasing the risk of conflict between them. The collapse of the regime may be superficially attractive, but it is a dangerous risk.

Timeline

Korean Peninsula

About the United States Institute of Peace

The United States Institute of Peace is an independent, nonpartisan federal institution created by Congress to promote research, education, and training on the peaceful resolution of international conflicts. Established in 1984, the Institute meets its congressional mandate through an array of programs, including research grants, fellowships, professional training programs, conferences and workshops, library services, publications, and other educational activities. The Institute's Board of Directors is appointed by the President of the United States and confirmed by the Senate.

About the Report

This report analyzes the North Korean food crisis, which began in 1994 and evolved into a major famine with high death rates. The report considers the causes of the famine, the central government's response to it, and the popular reaction to the government's inability to stem the steady collapse of the old system, as well as the convulsive effects of the famine on North Korean society, political system, and military. The international humanitarian aid response to the famine has had the unintended affect of weakening the central government's control over the society and has stimulated irrevocable changes in the economy. The report argues that much of the regime's external behavior is driven by the famine and considers the implications of the famine for the future of the country.

Andrew Natsios wrote this report while in residence as a senior fellow at the U.S. Institute of Peace in 1998-99. He is currently Secretary of Administration and Finance for the Commonwealth of Massachusetts. From 1989-93, he served in the Bush Administration as Assistant Administrator of U.S. Agency for International Development's (USAID) Bureau of Food and Humanitarian Assistance. He is the author of the book, American Foreign Policy and the Four Horsemen of the Apocalypse: Humanitarian Relief in Complex Emergencies and of numerous articles concerning complex emergencies and foreign policy.

See the complete list of Institute reports. The views expressed in this report do not necessarily reflect those of the United States Institute of Peace, which does not advocate specific policies.

August 2, 1999