Cambodia Section Home Mekong Network Home
Mekong.Net
About this site...

Kampuchea: A Demographic Catastrophe

Editor's Introduction: About This Document

The report that follows was prepared by the United States Central Intelligence Agency (CIA) in 1980. It is significant in that it was probably the first methodical attempt to analyze the statistical impact of the Khmer Rouge reign.

Time has shown that this report has serious flaws, most notably the complete lack of any discussion of the purges of 1977 and 1978. Some scholars (particularly Michael Vickery) have suggested that this omission represented a deliberate attempt to whitewash the Khmer Rouge, to justify U.S. opposition to the government installed by the Vietnamese invasion. Readers of the report may judge for themselves whether or not this is a plausible explanation. In any event, the report should not be construed as the final word on the demographics of communist Cambodia. However, it may prove useful as a point for further research and discussion.

More recent analyses of the demographic impact of the Khmer Rouge regime can be found in the articles Counting Hell and The Number: Quantifying Crimes Against Humanity in Cambodia.

This copy of the document was re-created from a paper copy which was reproduced by the National Technical Information Service, U.S. Department of Commerce, Springfield, VA 22161. As the report does not appear to be copyrighted, it has been converted to HTML for inclusion on this site. Since the text of the report was retyped manually, there may be some typos or transcription errors. Also, it should be noted that the format of the document has been altered somewhat to simplify the HTML. Because of the width of some of the tables, you may find it necessary to maximize your browser window.

A paper copy of the original document can be purchased from the CIA. For details, see the CIA's Publications page, at http://www.cia.gov/cia/publications/pubs.html. To locate documents pertaining to Cambodia, go to the Search page (http://www.cia.gov/search) and search for Kampuchea or Cambodia.

Many thanks to Nathan Folkert for transcribing the original document.

The text of the report, with graphs, follows.

 


Central Intelligence Agency: National Foreign Assessment Center Logo

Kampuchea:
A Demographic Catastrophe

 

A Research Paper

Research for this report was completed on 17 January 1980.

Comments and queries on this report are welcome and may be directed to:

Director of Public Affairs
Central Intelligence Agency
Washington, D.C. 20505
(703) 351-7676

For information on obtaining additional copies, see the inside of front cover.

GC 80-10019U
May 1980

 

Summary

The estimated decline in the Kampuchean population between 1970 and 1979 is unprecedented in any national population since World War II. A decade of war, social revolution, and invasion has thrown the country into chaos. Precise population figures are not available. To measure the demographic impact of the social upheavals on the Kampuchean population during the 1970s, we prepared population estimates for the period 1970 to 1979, using the 1962 census as a base. The December 1979 population is estimated to range between 4.7 million and 5.5 million persons, with the most likely estimate 5.2 million -- down from 7.1 million in 1970. Under normal demographic circumstances the population would have totaled roughly 9 million by yearend 1979.

After four years of sharply reduced growth, population numbers plunged following the takeover by the Cambodian Communist regime in April 1975; the decline continued through the nearly four years of Khmer Rouge dominance. By our estimates, the savagery of that regime caused an actual drop of between 1.2 million and 1.8 million people. The downward trend continued under the Vietnamese-supported Heng Samrin regime, with an estimated decline of about 700,000 persons by December 1979. The elevated number of deaths from inadequate diet and disease, the small number of births, and the flight of refugees ensure a continued decline.

The grim demographic outlook is for a Kampuchean population of a few children, few elderly people, and many prematurely old people whose lifespan has been drastically shortened by events. Decimated by disease, famine, and war, and bereft of its leaders and its labor force, the Kampuchean society will need decades to come back, if it survives at all.

 

Discussion

Ten years of war, social revolution, and invasion have played havoc with the Kampuchean population. Over the centuries the Khmers have repeatedly weathered expansionist incursions by neighboring states and European colonialism, but now they may be threatened with virtual extinction as a people.

The situation is chaotic, and precise population figures are not available. The only hard demographic data are from the 1962 census. The lack of a recent reliable population estimate supported by systematic assumptions relating to its preparation has led to the use of a wide range of population figures from a host of differing sources. Published 1979 population estimates range from 4 million to 9 million. The lower figure is the unsubstantiated estimate used by the Heng Samrin regime; the latter is a recent UN projection that gives little or no consideration to the effects of the social upheavals of the 1970s. (Footnote 1)

CIA Estimation Series
In an attempt to measure the impact of events during the 1970s on the Kampuchean population, we prepared population estimates for the period 1970 to 1979 (table I-1). Using the UN 1970 population estimate of 7.1 million as a base and incorporating the possible demographic impact of events during the 1970s, we estimate the 1 December 1979 population at somewhere between 4.7 million and 5.5 million. Our most likely estimate is 5.2 million. (Footnote 2) Therefore, since 1970, the Kampuchean population has declined by 1.5 million to 2.4 million people, or by 22-34 percent; the drop has been even more pronounced since the Communist takeover in April 1975. The difference between the CIA estimate and the UN estimate -- 3.5 million or more people --illustrates the full extent of the demographic disaster visited upon the Kampuchean people during this period (figure I-1).

Figure I-1 Population Estimates Graph

This report is in two parts. Part I briefly traces the demographic events of the past decade, highlighting the estimated ranges in population during the three regimes between July 1970 and December 1979, and discusses the outlook for the Kampuchean population. Part II presents the methodology used in obtaining our estimates.

We acknowledge that our assumptions are highly speculative given the uncertainty surrounding events in Kampuchea and their demographic impact; therefore, all estimates are presented in three series. The Medium Series portrays, in our view, the most likely pattern of demographic events. The High and Low Series, which incorporate more favorable and more adverse assumptions, respectively, set reasonable limits within which the population statistics are likely to fall. Population estimates cited in part I are from the Medium Series. (A detailed chronology of events, their estimated demographic impact, estimation assumptions, and total populations by period can be found in table II-4, enclosed herewith.


Table I-1: Population Estimates, 1970 - 1979
Midyear High Series Medium Series Low Series
  Total Population (thousands) Growth Rate (percent) Total Population (thousands) Growth Rate (percent) Total Population (thousands) Growth Rate (percent)
1970 7,060   7,060   7,060  
1971 7,133 1.03 7,133 1.03 7,111 0.72
1972 7,201 0.95 7,201 0.95 7,157 0.64
1973 7,270 0.95 7,270 0.95 7,204 0.65
1974 7,334 0.88 7,334 0.88 7,245 0.57
1975 6,877 -6.43 6,726 -8.65 6,358 -13.06
1976 6,544 -4.96 6,191 -8.29 5,812 -8.98
1977 6,407 -2.12 6,012 -2.93 5,626 -3.25
1978 6,284 -1.94 5,899 -1.9 5,507 -2.14
1979 6,141 -2.3 5,767 -2.26 5,337 -3.14
1 December 1979 5,521 -25.56 5,160 -26.71 4,684 -31.34

Part I. Estimates and Conditions by Period

Lon Nol Regime
(1 July 1970 to 17 April 1975)
The war between the Lon Nol regime and the Communist Khmer Rouge set the stage for the cataclysmic decline of the Kampuchean population that was to follow. Families were separated, food networks disrupted, and health services thrown into chaos as a result of the war emergency. Malnutrition, long recognized as a major problem for the poorest Cambodian families, rapidly engulfed other segments of the population. For this period, we assumed that birth rates declined as conditions worsened. Death rates, high since the 1960s, soared with the addition of an estimated 600,000 to 700,000 war-related deaths. During the Lon Nol regime the population increased by only an estimated 324,000 as the population growth rate fell from 2.8 percent (1965-70) to less than 1.0 percent (table I-2).

An even more consequential impact of this period of strife was the change in the makeup of the population, from a traditional rural society to one that was largely urban based. Large numbers of people fleeing the war moved from the countryside to the perceived safety of villages and cities, and from city to city as the war enveloped ever greater areas of the country. As the fighting intensified in the early months of 1975, more people crowded into the cities. By the time the Lon Nol regime fell in April 1975 the urban population probably was on the order of 4 million, more than half the total population. (UN projection in 1973-74 had assumed that only about one-fourth of the total population would be urban by 1975.)

Table I-2: Population Estimates
By Period (thousands):
Estimation Series 1 Jul 70 17 Apr 75 1 Jan 79 1 Dec 79
High Series 7,060 7,384 6,225 5,521
Medium Series (Most likely) 7,060 7,384 5,845 5,160
Low Series 7,060 7,278 5,452 4,684
Period Change:
Estimation Series Lon Nol Regime
1 Jul 70-17 Apr 75
Pol Pot Regime
17 Apr 75 - 1 Jan 79
Heng/Samrin/Vietnamese Regime
1 Jan 79 - 1 Dec 79
  Absolute (thousands) Percent Absolute (thousands) Percent Absolute (thousands) Percent
High series +324 +5 -1,159 -16 -704 -11
Medium series (Most likely) +324 +5 -1,539 -21 -685 -12
Low series +218 +3 -1,826 -25 -768 -14

Pol Pot Regime
(17 Aprl [sic] 1975 to 1 January 1979)
The almost four years of brutal rule by the Pol Pot regime drastically accelerated the disintegration of Kampuchean society. The destructive designs for the now largely urban population became apparent within hours after the Pol Pot takeover of Phnom Penh on 17th April 1975. Hundreds of thousands fled the nations immediately. Various sources put the number of ethnic Vietnamese who left between 200,000 and 500,000. For the remaining Kampucheans, it became obvious that the "decadent, Westernized" city dweller would be subjected to very different conditions and policies than the idealized peasant. The rural population, then totaling roughly 3 million people, was to be the foundation for the new Khmer Rouge revolutionary society.

Within two weeks of the Communist takeover, the entire urban population had been expelled from the cities and forcibly sent to the countryside. Given no provisions en route, they sustained themselves on foraged food and surface water. Upon arrival at designated rural areas, the displaced urban population was classified as new people and tasked with tilling fields and building shelters.

Conditions were harsh. Labor was performed by hand and foraging for food and water was still necessary. Food intake provided no more than 800 to 1,200 calories per day. (Footnote 3) Because of inadequate food, no medical care, and a lack of natural immunity to diseases peculiar to rural areas, the new people became sick and epidemic disease spread. Then, between September and December, hundreds of thousands of this already weakened populace, along with the majority of the rural population living in areas bordering Vietnam, were again relocated.

The forced marches and debilitating conditions, coupled with the excessively hard labor under strict discipline, resulted in an extremely high number of deaths among the new people. We estimate that three times as many died as would have been expected under normal conditions. In addition to the deaths from "natural" causes, 50,000 to 100,000 former military personnel, bureaucrats, teachers, and educated people may have been executed. The population was further depleted by the flight of up to 55,000 people to Thailand, half of whom died during the journey. Because evidence was so fragmentary, refugee flight to Vietnam was included only in the Low Series. Under the extreme conditions that prevailed during this period, births among the new people offset only a small fraction of the population losses from death and migration.

The old people -- those living in rural areas when Pol Pot gained control -- fared somewhat better. They too, however, suffered from food shortages, lack of medical care, and hard labor. We estimate that their births slightly outnumbered their deaths and that the old people population increased negligibly during this period.

The conditions imposed upon the people by the Pol Pot regime resulted in an absolute decline in the population between April 1975 and January 1979 of an estimated 1.2 million to 1.8 million people (figure I-2). Educated persons, military personnel, and the skilled were disproportionately represented in the numbers lost. The physical condition of the survivors was greatly weakened.

Figure I-2: Population Estimates Graph

Heng Samrin/Vietnamese Regime
(1 January to 1 December 1979)

The fall of Kampuchea to the Vietnamese may ultimately spell the demise of the Khmer as a people. The first six months of the Heng Samrin regime were characterized by the destruction of the agricultural system and another reordering of the population. The people were urged to leave communes into which they had been moved by the Khmer Rouge and to return to their former villages to grow food. Those who returned found that the Heng Samrin/Vietnamese forces were seldom able to protect them from sporadic attacks by the Pol Pot guerrillas, who (along with a sizable civilian population) had retreated to mountain strongholds along the Thai border. Fearful, the people deserted the farms for the more secure roads and towns. Most of those who returned to the cities were denied entry and restricted to makeshift camps in the immediate environs. Thousands of people chose to flee the country and the increasingly chaotic conditions. By the end of June about 17,000 more had fled to refugee camps in Thailand.

Conditions were deteriorating swiftly for the civilian population by late summer. The people had been unable to produce sufficient food and the Heng Samrin regime was no longer dispensing food rations. In July 1979 about three and one-half million people were estimated to be in danger of starvation. In the face of the impending disaster a great westward movement of people began in early fall, with many seeking haven in camps and holding centers in Thailand and along the border. Thousands of those who chose not to flee died of starvation and disease. A conservative estimate is that for every Kampuchean born during July-December 1979, 10 died. All told, the first year of the Heng Samrin rule had brought an additional drop of 700,000 in the population, to an estimated 5.2 million.

Outlook
The social upheavals in Kampuchea during the 1970s will continue to have an adverse impact on the Khmers not only in the short term, but also well into the next century. Even if great improvements were to occur in food supply, living conditions, and health standards, the catastrophic reduction of Khmer numbers will constrain recovery. Although there is no specific information on the age, sex, or occupational categories of the remaining population, the reduction in size of individual groups is bound to have a lasting effect on the future Khmer population.

Indeed, as things stand, short-term prospects are grim. After the incountry relief food and the rice from the meager 1979-80 harvest are gone -- probably by early spring -- severe hunger at best, famine at worst, could again strike and possibly extend well into 1981. Internal obstructions virtually preclude the timely distribution of externally supplied rice for food or seed. It also seems unlikely that there will be a bountiful 1980-81 harvest. Irrigation systems are in disrepair, many fields are damaged, and shortages of seed, fertilizer, draft animals, and manpower all point to a much reduced harvest.

Even were food and health conditions to improve markedly, a rebuilding of Kampuchean society would be a long process. The executions under Pol Pot effectively wiped out the leadership echelon, and the extreme conditions suffered during the last five years have decimated the adult labor population. There are few Khmers to replace the dynamic segments of society. The ranks of those over 20 years old are thin, and the numbers within those ranks possessing skills are few. With a life expectancy shortened drastically by events of the next decade, the adult population will decrease further over the next two decades. It could only be strengthened by the return of the Khmer now in Thailand; even so, revitalization of the country would be a lengthy and difficult undertaking.

Young people in their teen years are not readying themselves to fill normal adult roles. Conditions militate against their assuming leadership and labor positions. Their health is fragile, and their training has been curtailed. Survival is their goal.

As for the very young, few children were born during the Pol Pot years. The children who were born and survived the first crucial months and years are now suffering from disease and serious malnutrition. Many are orphaned. All children are living under extreme circumstances that will appreciably reduce their numbers. Whether they will be healthy and vigorous enough -- or whether there will be enough of them -- to eventually fill the jobs and occupations needed for a functioning society is highly questionable.

Furthermore, not many babies will be born in the next few years. The fertility of unhealthy mothers and fathers is low. The mortality level will undoubtedly be high for infants, for children under 5 years of age, as well as for mothers. The prospects for regeneration of the population by the end of the century thus are poor.

Part II. Methodology

The estimated population numbers for Kampuchea were obtained through the use of a simple bookkeeping method. Births occurring during the period were added to the population at the beginning of the period, deaths and refugees were subtracted. The resulting number was assumed to be the population at the beginning of the next period.

The estimates in this paper were developed as follows: Subjective reporting was studied to determine its validity; analysts of Kampuchean affairs were consulted to verify the occurrence of historical events; and assumptions were made about the probable impact of these events on the demographic parameters of population change --fertility, mortality, and migration. Since the estimates are based on available information and the expert interpretation of events, they should be considered reasonable; but because they are founded on incomplete historical data and attempt to evaluate the results of nearly 10 years of war and oppression on population change, they should be considered speculative.

Definitions

Fertility: refers to the actual reproductive performance of a population.

Mortality: refers to deaths that occur in a population.

Vital Rates: refers to the measurement of the births and deaths (the vital events) that occur in a population. In the following estimates the vital events are estimated using crude birth rates, crude death rates, and percentages dying (which are subsequently calculated as crude death rates).

Crude Birth Rates, Crude Death Rates: refers to the number of births or deaths during one year per 1,000 persons based on the midyear population. In the following estimates, however, the crude rates are defined as the number of births or deaths occurring during a specified period per 1,000 persons based on the population at the beginning of the period.

Data Base
The only demographic data available for Kampuchea are from the national census conducted in 1962. One important analysis of the census, including fertility and mortality trends and estimates and projections of the total population, was published in 1970 by George Siampos, The Population of Cambodia, 1945-1980. The United Nations has published estimates and projections for Kampuchea from 1950 to the year 2000 in a provisional report, Demographic Estimates and Projections for the World, Regions and Countries as Assessed in 1978. The US Bureau of the Census published estimates of demographic parameters and projections for Kampuchea in 1977, which are now being updated. For the Chinese population, William Willmott's The Chinese in Cambodia was a key source; his estimates used the 1962 census, surveys, and migration data and were published in 1968. None of these analyses allow for the effects of the war, and they make only very general estimates for 1975 to 1980.

There are estimates by US officials for the urban population, for refugees, and for war dead for 1970 to 1975, but many of these are contradictory and highly suspect. From 1975 to 1979 Kampuchea was closed to Western observers. Reporting was sparse and much of it was propaganda. Since the Vietnamese invasion in January 1979 and the increased flow of refugees to Thailand, a more complete picture of conditions since 1975 has begun to emerge. Still, most of the information available comes from Kampuchean refugees and thus is piecemeal in nature and possibly biased.

Assumptions and Estimates

Lon Nol Regime
July 1970 to 17 April 1975

Base Population. The UN-estimated 1970 population of 7,060,000 was accepted as the base population for all series in this estimation set. The United Nations derived this estimate by projecting population trends evident in the 1962 census.

Crude Birth Rate Estimates. The adjusted 1970 UN crude birth rate served as the base fertility measure for all estimation series. The 1970 rate was decreased by 6.0 percent in all series for mid-1970 to April 1975. The estimated rate of decline in the crude birth rate for Kampuchea was based on reported declines in Bangladesh. A recent study of war-impacted rural area in Bangladesh showed that crude death rates increased in the civilian population as a result of the 1971 War of Independence; malnutrition levels and incidence of disease also increased rapidly. While in no way implying htat similar conditions prevailed in the Bangladesh study area as in Kampuchea in 1970 to 1975, the assumed 6.0-percent decline in the birth rate is a reasonable assumption and possibly a conservative one. The impact of war conditions in Kampuchea may have intensified over the five years, and a faster rate of decline near the end of the period may have been justified. However, living conditions varied greatly and a conservative outlook was adopted. The rounded crude birth rates assumed are 46 per 1,000 in 1970, 45 for 1971 and 1972, 44 for 1973 and 1974, and 43 for 1 January 1975 to 17 April 1975.

Crude Death Rate Estimates. The estimated UN 1970 crude death rate of 18 per 1,000 was used in all estimation series and was held constant throughout the period. The decision to hold the rate constant implies that no deterioration in the standard of living occurred between 1970 and 1975. As this most likely was not the case and since the war eventually enveloped the entire population, an assumption about war-related deaths was necessary.

War-related deaths totaling 600,000 were added to those resulting from applying the 18-per-1,000 death rate to the population in the High and Medium Series, and 700,000 were added in the Low Series. (The deaths were prorated throughout the period as a demographic convenience. The level of such deaths in any one period is not known.) When deaths resulting from the estimated death rate are added to war-related deaths, the crude death rates in the High and Medium Series are 36 per 1,000 for 1970 and 35 per 1,000 for each year from 1971 to the end of the period. The rates for the Low Series are 39 per 1,000 for 1970 and 1971 and 38 per 1,000 for each year from 1972 to the end of the period.

The assumed number of war-related deaths (600,000 to 700,000) is debatable; US government sources put the figure unofficially at 600,000 to 700,000; authorities of Democratic Kampuchea say 600,000 to 800,000; Prince Sihanouk is quoted as saying 600,000. None of these estimates is well founded. However, by accepting 600,000 to 700,000 additional deaths, the death rate is about double the UN-estimated 18-per-1,000 rate. From the rural Bangladesh study, it is seen that death rates rose by 40 percent in the civilian population after a war of only a few months; therefore, a doubling of the rates for a war of five years duration does not seem unreasonable.

Pol Pot Regime
Phase 1: 17 April to 1 July 1975

Total Population. Estimates for this period were driven largely by two events: (1) the people of Vietnamese origin and Kampucheans with ties to them were repatriated to Vietnam immediately after the Khmer Rouge takeover; and (2) the remaining Kampuchean population was divided into two distinct groups and each group treated differently, resulting in a unique demographic impact on each.

The estimated number of repatriates to Vietnam varies widely; sources put the numbers between 200,000 and 500,000. In the High and Medium Series an estimated 250,000 repatriates were subtracted from the 17 April 1975 population of 7,384,000, leaving a Kampuchean population of 7,134,000. In the Low Series 500,000 were subtracted from a total of 7,279,000, leaving 6,779,000 Kampucheans.

The 17 April 1975 Kampuchean population (total population minus those repatriated to Vietnam) was divided into two groups, the old people and the new people. The urban (new people) population base is estimated at 4.0 million. The 4.0 million subtracted from the total Kampuchean population leaves the rural (old people) population at 3,134,000 in the High and Medium Series, and 2,779,000 in the Low Series (tables II-1A and II-1B).

Vital Rates for Old People. The crude birth and death rates during this period were assumed to be the same as the rates for the total population in the preceding period. The assumption is that conditions did not appreciably worsen for them immediately after the Khmer Rouge victory.

Vital Rates for New People. During this 10-week period about 10 percent of this population was assumed as having died in the Medium and Low Series, and 7 percent in the High Series. These percentages compare with the estimated 4 percent of the total population who died in each year during the Lon Nol period. Our assumption about the high number of deaths in the early Pol Pot period rests largely on refugee reports and other eyewitness accounts. Given the brutal circumstances surrounding the evacuation of about 4 million people from the cities and their eventual dispersal into the countryside, it is certain that many of the old, very young, weak or sick people died in far greater numbers than would be expected.

The Khmer Rouge methodically hunted down former members of the Lon Nol military, civil servants, and the intelligentsia. Nearly 100,000 new people were assumed executed in the Medium and Low Series -- 50,000 in this period, 30,000 from July 1975 to January 1976, 10,000 from January 1976 to July 1976, and 10,000 from July 1976 to January 1977. Only half as many were assumed executed in the High Series -- 25,000 in this period, 10,000 in each of the next two six-month periods from July 1975 to July 1976, and 5,000 from July 1976 to January 1977.

The crude birth rate of 43 per 1,000 from the preceding period was used to estimate expected births. It was assumed that the deaths of pregnant women would increase over expected levels under the harsh conditions and lack of medical care, thereby reducing the expected number of births. The births were reduced by about 10 percent in the High Series, yielding a crude birth rate of 39 per 1,000; by 25 percent in the Medium Series, yielding a rate of 32 per 1,000; and by 50 percent in the Low Series, yielding 22 per 1,000.

Refugees to Vietnam and to Thailand. The evidence of Khmer refugee flight to Vietnam is fragmentary. The number of such refugees is even more questionable. Because of the great uncertainty involved, the assumption of Khmer flight to Vietnam was included in the Low Series only. The number 30,000 was chosen arbitrarily for the entire Pol Pot period, and these refugees were assumed to come from the new people population.

Table II-1A: Population Estimates for New People
Date/Period High Series Medium Series (Most Likely) Low Series
  Population (thousands) Period Growth Rate (percent) Period Vital Rates Population (thousands) Period Growth Rate (percent) Period Vital Rates Population (thousands) Period Growth Rate (percent) Period Vital Rates
      Estimated Crude Birth Rate (per thousand) Estimated Crude Death Rate (per thousand)     Estimated Crude Birth Rate (per thousand) Estimated Crude Death Rate (per thousand)     Estimated Crude Birth Rate (per thousand) Estimated Crude Death Rate (per thousand)
17 April 1975
17 April - 1 July 1975
4,000 -34.4 39.0 371.0 4,000 -54.5 32.0 547.0 4,000 -56.0 22.0 547.0
1 July 1975
1 July 1975 - 1 January 1976
3,727 -14.8 15.0 155.0 3,576 -25.1 15.0 240.0 3,565 -25.5 15.0 241.0
1 January 1976
1 January - 1 July 1976
3,460 -4.8 10.0 56.0 3,153 -8.3 10.0 87.0 3,137 -8.6 10.0 86.0
1 July 1976
1 July 1976 - 1 January 1977
3,378 -4.5 10.0 53.0 3,025 -8.3 10.0 86.0 3,005 -8.6 10.0 86.0
1 January 1977
1 January - 1 July 1977
3,302 -4.2 10.0 50.0 2,902 -4.5 10.0 50.0 2,878 -4.7 10.0 50.0
1 July 1977
1 July 1977 - 1 January 1978
3,233 -4.2 10.0 50.0 2,838 -4.4 10.0 50.0 2,811 -4.7 10.0 50.0
1 January 1978
1 January - 1 July 1978
3,166 -4.2 10.0 50.0 2,776 -4.4 10.0 50.0 2,746 -4.7 10.0 50.0
1 July 1978
1 July 1978 - 1 January 1979
3,101 -4.2 10.0 50.0 2,716 -4.4 10.0 50.0 2,683 -4.5 10.0 50.0
1 January 1979 3,037       2,657       2,623      

 

Table II-1B: Population Estimates for Old People
Date/Period High and Medium Series Low Series
  Population (thousands) Period Growth Rate (percent) Estimated Period Vital Rates Population (thousands) Period Growth Rate (percent) Estimated Period Vital Rates
      Estimated Crude Birth Rate (per thousand) Estimated Crude Death Rate (per thousand)     Estimated Crude Birth Rate (per thousand) Estimated Crude Death Rate (per thousand)
17 April 1975
17 April - 1 July 1975
3,134 2.5 43.0 18.0 2,778 2.6 43.0 18.0
1 July 1975
1 July 1975 - 1 January 1976
3,150 0.5 30.0 25.0 2,793 0.5 30.0 25.0
1 January 1976
1 January - 1 July 1976
3,158 0.5 30.0 25.0 2,800 0.5 30.0 25.0
1 July 1976
1 July 1976 - 1 January 1977
3,166 0.3 28.0 25.0 2,807 0.3 28.0 25.0
1 January 1977
1 January - 1 July 1977
3,170 0.3 28.0 25.0 2,811 0.3 28.0 25.0
1 July 1977
1 July 1977 - 1 January 1978
3,174 0.3 28.0 25.0 2,815 0.4 28.0 25.0
1 January 1978
1 January - 1 July 1978
3,179 0.3 28.0 25.0 2,820 0.3 28.0 25.0
1 July 1978
1 July 1978 - 1 January 1979
3,183 0.3 28.0 25.0 2,824 0.4 28.0 25.0
1 January 1979 3,188       2,829      

The number of refugees that arrived in Thailand during the Pol Pot years is also very much in question. On the basis of journalists' reports, subjective refugee stories, and probable conditions within the country, we estimate refugee flight to Thailand from mid-1975 through 1978 at 55,000 in the Medium and Low Series, and at 20,000 in the High Series.

Phase 2: July 1975 to January 1976

A second forced population displacement, this time affecting both old people and new people, occurred during September to December 1975. Although more orderly and accomplished by vehicle rather than on foot, again there was no provision made for food or water during the journey or for adequate shelter at the destination.

Famine affected many areas from July 1975 to the end of the year. Epidemic diseases, malaria, and food deficiency maladies were rampant among the already weakened populace. Although the entire population was adversely affected by the deteriorating health conditions, the new people population was assumed to have suffered most.

Vital Rates for Old People. Crude birth rates were estimated to have declined by about 30 percent, from 43 per 1,000 to 30 per 1,000. A 30-percent decline is similar to the reduction observed in one year following the 1974 famine in Bangladesh. The United Nations also decreased birth rates by some 30 percent in its projections for the 1970-75 and 1975-80 periods. The decision to reduce birth rates in our estimates was taken with some degree of assurance, given the impact of bad health conditions and social dislocations on women in childbearing ages.

Estimated crude death rates were increased from 18 per 1,000 to 25 per 1,000. The rate of 25 per 1,000 during this period is the same rate estimated in UN projections for earlier stages of development, for example, 1950 to 1955. A return to this high level is a reasonable assumption, given the forced marches, disrupted medical care, and inadequate diet.

Vital Rates for New People. Crude birth rates for all series were estimated at 15 per 1,000, one-half the old people's rate. Harsh living conditions, psychological stress, bad diet, and no medical care would cause a high incidence of spontaneous abortions and, indeed, the death of many pregnant women. The assumed drastic decline in the birth rates seems a likely possibility under the prevailing conditions.

In all series the number of deaths from July 1975 to January 1976 was estimated to be equal to the number of deaths sustained from April to July 1975. In part because of the weakened condition of the populace, the second forced march was as detrimental as was the April 1975 exodus from the cities. Deaths resulting from the second march were estimated at 400,000 in the Medium and Low Series and 280,000 in the High Series.

Phase 3: January 1976 to January 1977.

Food was more plentiful for the first half of 1976, but still rationed and inadequate given the severe physical demands placed on the population. Food shortages, disease, and poor living conditions kept birth rates low and death rates high.

Vital Rates for Old People. The crude birth rate was held constant at 30 per 1,000 for the first six months in all series. An additional 6-percent decline (bringing the rate down to 28 per 1,000) was assumed for the last six months. The crude death rate was assumed to remain constant at 25 per 1,000 in all series for the entire period.

Vital Rates for New People. The crude birth rate was assumed to decline to 10 per 1,000 in the first six months for all series. It was then held constant in the second six months.

For each six-month period, in the Medium and Low Series, we assumed that about 4 percent of the population died from more or less natural causes and that an additional 10,000 were executed. Implicit in the lower percentage of deaths is the assumption that a slightly better food supply and marginally more stable living conditions contributed to a higher survivor rate than in previous periods. Presumably the remaining population learned the techniques of survival under harsh conditions.

In the High Series the crude death rate was assumed to be 50 per 1,000 for both six-month periods, or double the old people's crude death rate. (Assumed deaths total about 2.5 percent of the population during each period.) An additional 10,000 were assumed executed in the first six months, and 5,000 in the second six months.

Phase 4: January 1977 to January 1979

Living conditions most likely did not vary during these two years from the conditions during 1976.

Vital Rates for Old People. This population barely grew throughout the period. Estimated crude birth rates were assumed to be 28 per 1,000; crude death rates, 25 per 1,000.

Vital Rates for New People. This population continued to decline. Crude birth rates for all series were held constant at 10 per 1,000. In the Medium and Low Series the percentage dying was held at 2.5 percent resulting in death rates of 50 per 1,000.

Heng Samrin Regime
January 1979 to July 1979

The Heng Samrin regime did not provide additional food rations, health services, or even a semblance of social order. Under these conditions the former old people were assumed to have lost their slightly favored position. They most likely shared with the new people the food taken from warehouses and harvested after the Pol Pot forces left, but they also shared the ensuing food shortages. In the eyes of marauding Pol Pot forces they were just as suspect of collaboration with the Vietnamese as were the new people and suffered the same retaliation. The Heng Samrin government made no distinction between old people and new people; neither received food or services and both were caught between the opposing forces fighting in Kampuchea.

Vital Rates, Total Population. For this period, we assumed that the wide disparity in birth rates between the two populations narrowed slightly. The rate for the former old people was allowed to decline to 26 per 1,000 as their living conditions became less stable; the rate for the new people was increased to 11 per 1,000 on the assumption that with Pol Pot gone conditions improved slightly for them. Crude birth rates for the total population were prepared by combining the number of births from the two populations and by assigning these births to the total population at the beginning of the period.

An assumed 2 percent of the total population died during this period. This percentage implies that not as many new people died as in the preceding period (estimated at 2.5 percent) and that more old people died in this period than previously (estimated at 1.4 percent).

Refugees to Thailand. About 17,000 refugees were in UN camps in Thailand by 30 June 1979. In all series these refugees were assumed to have left Kampuchea during the first half of 1979. In the Low Series, an additional 40,000 refugees were assumed to be in Thailand, but not in camps.

July 1979 to December 1979

Fighting between Pol Pot and the Vietnamese resumed in September, forcing people in the border area to flee to Thailand. The harvest at the end of summer was meager, consisting for the most part of food crops that were grown around the secure roads and towns. Widespread starvation throughout the country was evident by early November. An international relief effort was started, but as of November very little food had been delivered and little of that had been distributed among the civilian population. The people in the countryside did not receive any relief supplies.

A great westward movement of people from all areas of the country toward the Thai border began in late October. The only concentration of Khmers in Kampuchea was around Phnom Penh; 200,000 were reported in the environs of the city, some 70,000 in the city.

Vital Rates, Total Population. Crude birth rates were assumed to decline to the level of 10 per 1,000 for the population as a whole.

The percentage dying during this five-month period was estimated at 4 percent to reflect famine conditions. This is the same percentage as was estimated for the new people during the January-December 1976 food shortage.

Refugees to Thailand. In the High and Medium Series we estimated that 400,000 Khmers had entered camps and holding centers in Thailand by 1 December; the Low Series incorporated a total of 500,000. (These estimates are rough approximations and are subject to abrupt fluctuation because of the fluid nature of the refugee situation along the border.) In the Low Series the 40,000 who were subtracted as refugees in the preceding period were assumed to have voluntarily reentered Kampuchea in August 1979. These 40,000 were added back into the Low Series population.

Chinese in Kampuchea
There were approximately 425,000 Chinese (about 6 percent of the total population) in Kampuchea in 1970 (table II-2). The majority lived in urban areas, employed as rice merchants, shopkeepers, or business-men; those who were in rural areas were engaged in small trading enterprises or farming.

Table II-2            Thousands
Estimated Chinese Population in Kampuchea, 1970 to 1979
Date High Series Medium Series (Most Likely) Low Series
1 July 1970 425 425 425
17 April 1975 444 444 438
1 December 1979 299 259 247

For July 1970 to 17 April 1975, it was assumed that the growth rates estimated for the total population applied to the Chinese population as well. It was further assumed that throughout this period, as conditions deteriorated in the countryside, there was a gradual but constant flow of rural and small-town Chinese to the cities to find refuge. Growth rates for the Chinese population from April 1975 to January 1979 thus were assumed to be the same as those for the new people population in all series. For January to December 1979, the growth rates for the total population were used.

Estimates for People Facing Starvation

The numbers of people in danger of starvation in midsummer 1979 were developed arithmetically using very broad assumptions. (Table II-3 gives the possible combinations considered based on populations as of July 1979.)

Table II-3            Thousands
Estimated Numbers of People in Danger of Starvation, July 1979
Number Facing Starvation Population
Assuming: High Series Medium Series (Most Likely) Low Series
All of the new people and one-fourth of the old people endangered 3,772 3,396 3,250
One-half of the new people and one-fourth of the old people endangered 2,281 2,093 1,973
One-half of the total population endangered 3,070 2,884 2,668
Estimated Populations
Estimation Series Total Population Old People (Pre-1975 rural population) New People (pre-1975 urban population)
High 6,141 3,159 2,982
Medium 5,767 3,162 2,605
Low 5,337 2,783 2,554

 

Table II-4:
Chronology of Demographic Impact of Events in Kampuchea

Total Population (in thousands)
Date High Series Medium Series Low Series
1 Jul 70 7,060 7,060 7,060
17 Apr 75 7,384 7,384 7,278

Date: 1 Jul 70-17 Apr 75

Event: Civil war. Lon Nol government against Khmer Rouge and Vietnamese forces.

Estimated Demographic Impact: An estimated 600,000 to 700,000 war-related deaths. Refugees flee to cities; urban population estimated at 4 million.

Estimation Assumptions: Total population: UN estimate of 7,060,000 for 1 July 1970 used for all series.

Births: UN estimated crude birth rate of 46 per 1,000 for 1 July 1970 accepted and decreased by 6.0 percent. Rate of decline nearly three times normal, but in line with type of decline estimated for war-ravaged areas of Bangladesh following 1971 Bangladesh-Pakistan War.

Deaths: UN estimated crude death rate of 18 per 1,000 for 1 July 1970 held constant; added 600,000 war-related deaths (prorated evenly) in the High and Medium Series, and 700,000 in the Low Series.

Rates of Growth: calculated, used estimated populations for beginning and end of period.




Total Population (in thousands)
Date High Series Medium Series Low Series
17 Apr 75 7,384 7,384 7,278
1 Jul 75 6,877 6,726 6,358

Date 17 Apr-1 Jul 75

Event: Lon Nol defeated; Pol Pot and Khmer Rouge gain control of Kampuchea.

Estimated Demographic Impact: Estimated 250,000 to 500,000 flee to Vietnam.

Estimation Assumptions: High and Medium Series: Assumed 250,000 repatriated to Vietnam. Low Series: Assumed 500,000 repatriated to Vietnam.

Event: Population divided into urban and rural groups.

Estimated Demographic Impact: Urban population expelled from cities, forcibly moved to countryside.

Estimation Assumptions: After subtraction of repatriates to Vietnam, assumed new people comprised the estimate 4 million in cities, old people the remainder -- the 3.1 million assumed in rural areas. Old people: crude birth and death rates assumed constant at January-April 1975 level.

Estimated Demographic Impact: Numerous deaths during evacuation. No provisions, food and shelter inadequate. Population weak and listless; high incidence of pregnant women deaths.

Estimation Assumptions: High Series: New people: assumed 90 percent of expected births occurred (crude birth rate assumed same as old people's). Medium Series: Assumed 75 percent of expected births occurred. Low Series: Assumed 50 percent of expected births occurred. High Series: Assumed 7 percent of population died. Medium and Low Series: Assumed 10 percent of population died.

Event: Lon Nol military personnel, civilian bureaucrats, and educated elite purged.

Estimated Demographic Impact: Destruction of leadership element of Kampuchean society and urban-based medical system.

Estimation Assumptions: Assumed about 250,000 new people targeted for execution -- not including families: 200,000 military; 30,000 civil servants; 20,000 of the educated, particularly teachers. Of 250,000: 20 percent died of hunger and disease, 50 percent of remaining executed in Medium and Low Series and 50 percent of that number executed in High Series.

Estimated Demographic Impact: Executions begin.

Estimation Assumptions: High Series: Assumed 25,000 executed. Medium and Low Series: Assumed 50,000 executed.

Estimated Demographic Impact: Kampuchean refugees begin to flee to Vietnam.

Estimation Assumptions: High and Medium Series: Assumed no refugees to Vietnam. Low Series: Assumed 30,000 fled to Vietnam during 1975-79 from new people population. Assumed 3,000 refugees to Vietnam




Total Population (in thousands)
Date High Series Medium Series Low Series
1 Jul 75 6,877 6,726 6,358
1 Jan 76 6,618 6,311 5,937

Date 1 Jul 75-1 Jan 76

Event: Second forced displacement, included both old and new people.

Estimated Demographic Impact: Famine occurs in some areas; no medical care; incidence of malaria, dysentery, and other diseases increasing. Extremely hard work required by all.

Estimation Assumptions: Old people: assumed 30-percent drop in crude birth rate. Assumed crude death rate rose to 25 per 1,000 (UN estimated 1950-55 level). New people: assumed same number died in this period as died in previous period. Crude birth rate 50 percent of old people's rate. High Series: Assumed 20,000 new people fled to Thailand between July 1975 and January 1979. Medium and Low Series: Assumed 55,000 new people -- 21,000 to camps, 8,000 illegals, and 26,000 died in attempt -- fled to Thailand between July 1975 and January 1979.

Estimated Demographic Impact: Refugee flight continues to Vietnam and begins to Thailand.

Estimation Assumptions: High Series: Assumed 5,000 refugees to Thai camps. Medium and Low Series: Assumed 5,000 to Thai camps, 5,000 illegals, 10,000 died in attempt. Low Series: Assumed 5,000 refugees to Vietnam.

Estimated Demographic Impact: More executions.

Estimation Assumptions: High Series: Assumed 10,000 executed. Medium and Low Series: Assumed 30,000 executed.




Total Population (in thousands)
Date High Series Medium Series Low Series
1 Jan 76 6,618 6,311 5,937
1 Jul 76 6,544 6,191 5,812

Date: 1 Jan-1 Jul 76

Event: Meager rice harvest. Khmer Rouge firmly in control of population.

Estimated Demographic Impact: Food rations above starvation level and most people sheltered. Unrelenting hard work on inadequate diet and lack of basic medical care, disease spreads among weakened populace.

Estimation Assumptions: Old people: crude birth and death rates from previous period remain constant. New people: crude birth rate declines a further 33 percent. High Series: Crude death rate double old people's rate, about 2.5 percent of population died. Medium and Low Series: Assumed 4 percent of population died.

Estimated Demographic Impact: Refugee flight continues.

Estimation Assumptions: High Series: Assumed 3,000 refugees to Thai camps. Medium and Low Series: Assumed 3,000 refugees to Thai camps, 2,000 illegals, 3,000 died in attempt to escape.

Estimated Demographic Impact: More executions.

Estimation Assumptions: High Series: Assumed 10,000 executed. Medium and Low Series: Assumed 10,000 executed.




Total Population (in thousands)
Date High Series Medium Series Low Series
1 Jul 76 6,544 6,191 5,812
1 Jan 77 6,472 6,072 5,689

Date: 1 Jul 76-1 Jan 77

Event: Food again in short supply.

Estimated Demographic Impact: Famine again occurs in some areas; disease spreads among total population.

Estimation Assumptions: Old people: crude birth rate declines 6 percent. Crude death rate remains constant. New people: crude birth rate remains constant. High Series: Crude death rate remains constant. Medium and Low Series: Assumed 4 percent of population died.

Estimated Demographic Impact: Refugee flight continues.

Estimation Assumptions: High Series: Assumed 3,000 refugees to Thai camps. Medium and Low Series: Assumed 3,000 refugees to Thai camps, 1,000 illegals, 3,000 died in attempt to escape. Low Series: Assumed 5,000 refugees to Vietnam.

Estimated Demographic Impact: Final executions.

Estimation Assumptions: High Series: Assumed 5,000 executed. Low Series: Assumed 10,000 executed.




Total Population (in thousands)
Date High Series Medium Series Low Series
1 Jan 77 6,472 6,072 5,689
1 Jul 77 6,407 6,012 5,626
1 Jan 78 6,345 5,955 5,566
1 Jul 78 6,284 5,899 5,507
1 Jan 79 6,225 5,854 5,452

Date: 1 Jan 77-1 Jan 79

Event: Khmer Rouge still in control.

Estimated Demographic Impact: Food remains in short supply. Among the new people, the weak and vulnerable dead. Survivors learning to live under conditions, although still in dire straits.

Estimation Assumptions: Old people: crude birth and death rates remain constant for each half-year of the period. New people: crude birth rates remain constant for each half-year period. High Series: Crude death rate remains constant for each half-year period. Medium and Low Series: About 2.5 percent of population died during each half-year period.

Estimated Demographic Impact: Refugee flight continues.

Estimation Assumptions: High Series: Assumed for each successive half-year period: 3,000, 2,000, 2,000, and 2,000 refugees to Thai camps. Medium and Low Series: Assumed for each successive half-year: 2,500 refugees, 2,500 died in attempt, and no illegals. Low Series: Assumed for each successive half-year of the period: 4,000, 4,000, 3,000, and 1,000 refugees to Vietnam.




Total Population (in thousands)
Date High Series Medium Series Low Series
1 Jan 79 6,225 5,845 5,452
1 Jul 79 6,141 5,767 5,337

Date: 1 Jan-1 Jul 79

Event: On January 1979 Phnom Penh fell to Vietnamese troops. Heng Samrin installed as head of new government.

Estimated Demographic Impact: Harsh labor requirements and oppressive control measures of Pol Pot regime lifted, but new government unable to supply food, shelter, clothing, or health care.

Estimation Assumptions: Assumed crude birth rate for old people declined by 6 percent while rate for new people increases by 6 percent. Assumed about 2.0 percent of total population died during period.

Estimated Demographic Impact: In Pol Pot-controlled areas, general deprivation and strict control still enforced. Physical security for civilian population poor as opposing forces fight for dominance.

Estimated Demographic Impact: Food for civilians, Vietnamese, and Pol Pot forces in short supply.

Estimated Demographic Impact: Refugee flight continues.

Estimation Assumptions: High and Medium Series: Assumed 17,000 refugees to Thai camps. Low Series: Assumed 17,000 refugees to Thai camps and an additional 40,000 in Thailand but not in camps.




Total Population (in thousands)
Date High Series Medium Series Low Series
1 Jul 79 6,141 5,767 5,337
1 Dec 79 5,521 5,160 4,684

Date: 1 Jul-1 Dec 79

Event: Fighting resumed between Heng Samrin/Vietnamese and Pol Pot forces.

Estimation Assumptions: Assumed crude birth rate of 10 per 1,000 for total population.

Event: Meager harvest.

Estimated Demographic Impact: Famine becomes widespread by end of October.

Estimation Assumptions: Assumed about 4.0 percent of total population died.

Event: International relief effort started.

Estimated Demographic Impact: International relief effort not able to alleviate conditions.

Estimated Demographic Impact: Refugees from every part of country join flight.

Estimation Assumptions: High and Medium Series: Assumed 400,000 refugees to Thai camps. Low Series: Assumed 40,000 refugees from previous period returned to Kampuchea. Assumed 500,000 new refugees to Thai camps.


Notes:

Footnote 1: In 1978, the UN published projections that imply a 1 December 1979 Kampuchean population in the range of 8.7 million to 9.0 million people. It estimated a population increase of 15.0 percent between 1970 and 1975, and a 9.0 - 14.0 percent increase in the 1975-80 period. Overall, the UN projections for 1970 to 1980 show what Kampuchea's growth would have been in the absence of emigration and deaths from war, mass executions, and famine. As such, it provides a benchmark for estimating the full population impact of the past decade of social upheaval.

Footnote 2: Ordinarily, population estimates based on incomplete data are made by using demographic parameters from other countries at similar levels of development, or by using demographic models constructed for developing countries. In the case of Kampuchea, and especially for the period 1975-79, the reported events were so bizarre that it was impossible to assume a demographic similarity with any other country or with an existing model. Consequently, demographic estimates were linked to the unique conditions in the country as reconstructed from narrative reports by US Government sources for 1970-75 and by journalists, refugees, and relief workers for 1975-79.

Footnote 3: The average adult body requires a minimum of 1,800 calories per day to perform light labor. Below this minimum intake the body begins to use muscle mass for protein, the person becomes thin, listless, and susceptible to disease.

Front Cover Note:
[Editor's Note: In the interest of completeness, the information below is reproduced from the original document. However, the procedure for obtaining these documents has probably changed. See the CIA's Publications page, at http://www.cia.gov/cia/publications/pubs.html for details.] This publication is prepared for the use of the US Government officials, and the format, coverage, and content are designed to meet their specific requirements. US Government officials may obtain additional copies of this document directly or through liason channels from the Central Intelligence Agency.

Requesters outside the US Government may obtain subscriptions to CIA publications similar to this one by addressing inquiries to:

Document Expediting (DOCEX) Project
Exchange and Gift Division
Library of Congress
Washington, D.C. 20540
or:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161

Requesters outside the US Government not interested in subscription service may purchase specific publications either in paper copy or microform from:

Photoduplication Service
Library of Congress
Washington, D.C. 20540
or:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
(To expedite service call the NTIS Order Desk (703)557-4650)

W3C Validation