The Real Economics of Kaesong

The February debate on 38 North between Ruediger Frank and William Brown highlighted opposing views about the Kaesong Industrial Zone (KIZ), and it most likely helped readers to better understand the ins and outs of the KIZ.

(Photo: Getty Images)
(Photo: Getty Images)

Nonetheless, their debate missed some important aspects of the KIZ, and included arguments based on misplaced assumptions about the reality of Kaesong. The most conspicuous of these dubious points was Brown’s assertion that Kaesong’s workers were “slaves.”

This article will attempt to provide a more accurate picture of the KIZ, first by addressing the characterization of Kaesong workers as slaves, and then by considering Kaesong’s significance to both Koreas as well as its shutdown and possible future.

Were Kaesong’s Workers “Slaves”?

Kaesong’s workers cannot necessarily be considered slaves, as Brown claims, for three reasons: 1) they were paid although not in dollars, but in goods; 2) they responded to price signals in a manner similar to workers in capitalist economies; and 3) the constant labor scarcity in Kaesong likely enabled them to avoid the most brutal abuses typically associated with slavery.

Brown argues that each KIZ worker received minuscule compensation—on the order of $5 per month—because the North Korean government effectively garnished their wages through distortion of the exchange rate.[1] According to Brown, South Korean companies provided all Kaesong salaries to the government,[2] which in turn conveyed wages to workers at the very low official exchange rate of about 170 North Korean won per dollar. He suggested that workers had to re-exchange their won for US dollars at an extremely high black-market rate (about 8,000 North Korean won per dollar) to buy rice and other daily necessities at black markets.[3]

This description is only partially true. While the North Korean government indeed took 30 percent of the total labor compensation as a de facto tax, the remainder (70 percent) went to each worker, mostly in the form of ration coupons redeemable for rice and other goods at Kaesong’s state-run commissaries. Employees reportedly preferred the state-run commissaries to black markets because, with the coupons, they could purchase goods at more favorable prices.[4]

According to Song Yong-deung, a Korean-Australian whose company managed Kaesong’s commissaries from 2004 to 2006, the dollar value of commissary goods purchased by North Korean workers amounted to roughly 90 percent of their total wages after tax.[5] That is, they actually consumed most of their disposable income over the three years that his company ran the commissaries. His estimate was based on the fact that workers were paid in the form of coupons, not actual cash, issued from the North Korean government’s General Central Bureau of Development and Supervision for the Special Zone (CGB), a subsidiary organ of the North Korean Cabinet, which manages the KIZ and which pays the workers. Workers would then use these coupons to purchase goods in the KIZ commissaries. Additionally, in order to purchase commissary inventory from countries like China and Malaysia, Song had to trade these coupons received from KIZ workers for actual US dollars also at the CGB. This recycling of coupons through the commissary provided a way to track how the coupons were being used and what portion was going toward consumption.

Song’s experience provides only a limited picture, since he turned over management of the commissaries to the North Korean government in 2007. Since then, the North Korean government may have emptied the shelves of Kaesong’s commissaries to the extent of store inventories elsewhere in the country, rendering the coupons useless. For this reason, the recent real wages of KIZ workers require confirmation.

In a recent OhMyNews report, Kim Chin-hyang, who worked in the KIZ from 2008 to 2011, explained that since 2006, few changes had been made to the salary payment system of the KIZ, which was based on coupons and commissaries.[6] Another South Korean official, who worked in the KIZ from 2013 until the February 2016 shutdown, explained that Kaesong’s workers “appeared to receive adequate nutrition through the commissaries.”[7]

Second, “slaves” do not typically receive extra pay in the form of incentives, bonuses and overtime premiums. The base monthly salary for each KIZ worker before the shutdown was $73.87 for 48 labor-hours per week,[8] but they generally earned far more than that through incentives and bonuses: they earned a 50 percent premium over their base hourly pay for each hour of overtime work, and earned double when working on holidays.[9] As a result, some workers earned up to $240 per month while others earned as little as $150.[10] For some, working extra hours was possible, in part, because physical conditions in the KIZ far surpassed those of other North Korean workplaces or of Kaesong residential areas.[11]

This pay disparity indicates that many Kaesong workers were willing to work longer hours in order to receive additional wages.[12] They knew exactly how much they had earned each month, and to confirm that they received the proper amount on payday, each employee signed a ledger—shared by the North Korean government and South Korean companies—that contained the labor-hours of every worker.[13] A kind of a price mechanism was therefore at work in the KIZ, though its signal was measured in hours and coupons rather than wages.[14]

Finally, the labor market in Kaesong was a “seller’s market” that benefited North Korean workers in key respects. As of October 2015, the number of North Korean workers in the KIZ was 54,357,[15] while the population of the city of Kaesong was 192,574, based on the most recent available data.[16] It means roughly 28 percent of Kaesong’s population worked in the KIZ. The South Korean companies demanded more labor, but supply was limited.[17] Kaesong lacked dorms or apartments to accommodate more labor, and with the exception of roughly 200 commuter buses that connected Kaesong city to the KIZ, there was no long-distance transportation to bring in more laborers. Given this labor scarcity, workers who faced significant abuses, if any, on the shop floor presumably could move to other workplaces.

Such conditions may not sound heavenly to residents of the free world, but they appeared wonderful to many North Koreans and are sufficient to cast doubt on Brown’s characterization of KIZ workers as “slaves.” While South Korean companies in the KIZ plainly benefited from employing a low-cost workforce with deeply limited opportunities outside their factories, Kaesong workers did enjoy a degree of independence through the site’s payment scheme and through the relative freedom of movement afforded by its labor market.

Did KIZ Funds Support Weapons Development?

Observers have often asked whether the KIZ helped to fund North Korea’s nuclear weapons and ballistic missile development. While the answer is not entirely clear, the KIZ could not have provided significant support to these programs.

According to the announcement of the South Korean Ministry of Unification, a total of $120 million went to North Korea through the KIZ in 2015. [18] This total can be broken down into three parts: 1) after-tax salary; 2) tax or what North Korea calls “socio-cultural policy cost (SCPC)”; and 3) social insurance. If we assume that the total labor compensation is 100, for example, then, 70 will be after-tax salary, and 30 will be SCPC, and to this total labor compensation, social insurance premium of 15 is added (see Table 1).[19]

Table 1. Structure of the Total Dollar Transfer to North Korea[20]

*Equals base monthly salary times 12 months, multiplied by the number of North Korean workers in the KIZ ($48 million = $73.87 X 12 months X 54,357).

As we analyzed above, however, most of the after-tax salary are actually consumed by the workers. Therefore, it is roughly $47 million or 39 percent of the total dollar transfer (=$47m/$120m) that is left to the discretion of the North Korean government.

According to the agreements signed between North and South Korea, the North Korean government is supposed to spend these funds to cover the costs of education, medicine and other public services including social infrastructure, industrial accident compensation, retirement annuity and social security for the KIZ workers and their families in Kaesong.[21]

In theory, it is possible the North Korean regime could have misused these funds for weapons. In practice, however, it would have been difficult given the public demand for use of those funds. Kaesong must have schools and hospitals, however low their quality may be. There must be police and fire stations, too. All these institutions must have employees who depend on government ration coupons to survive. There should also be utilities, roads and bridges that require at least minimum maintenance.

What is the size of these actual demands? We do not know for sure how much this city of 200,000 needs for these purposes, but we can estimate.

The South Korean city closest to Kaesong in size is Yangju, whose population as of 2015 was 204,907.[22] Yangju’s total annual budget in 2014 was 493 billion South Korean won,[23] meaning that its annual budget per citizen was 2.4 million South Korean won, or about $2,000. This per capita maintenance cost is not much different in the United States, where the city of Palo Alto budgeted $185.7 million annual general expenses in 2016 (excluding capital improvement and utilities).[24] Using the city’s estimated population of 66,642 in 2013,[25] the city’s per capita cost for the maintenance appears to be roughly $2,787 (=$185.7 million/66,642).

If we suppose that the city of Kaesong spends one-tenth the amount of Yangju per citizen on maintenance, the annual cost required for Kaesong is approximately $39 million (=$200 X 192,574), which happens to approximate the combined annual amount of the SCPC and social insurance premium ($43 million). If our assumptions about the estimated maintenance cost of Kaesong are true, then not much money would go to Pyongyang because most of the “discretionary funds” would be consumed in Kaesong.[26]

While we cannot exclude the possibility that at least some KIZ funds may have ended up in Pyongyang—therefore possibly used for weapons—the city of Kaesong probably spent a significant portion of funds taxed from KIZ workers to meet the fundamental demands of the population.

Did the KIZ Amount to Unilateral Aid for North Korea?

To the question of whether the KIZ was merely a form of unilateral aid to North Korea, the answer is a definite “no.” The South Korean companies in Kaesong, the South Korean employees of those companies, and the South Korean consumers who bought KIZ products at low prices all benefited from the KIZ.

As of October 2015, there were 124 South Korean companies in the KIZ. Their projected output for 2015, based on the sum of the values added in the KIZ, was roughly $500 million, about five times more than the wages they paid to workers.[27] In addition, if we calculate the total output based on the actual market prices of the final products, then the value of the total production may soar to $1.5 billion or more.[28]

Moreover, the KIZ created jobs for South Koreans. As of October 2015, 809 South Koreans also worked in the KIZ.[29] If we include the number of workers in the South Korean offices of those companies as well as those employed by their subcontractors, the number of people employed because of Kaesong may have been much larger.

South Korean consumers benefited from the KIZ, too. Its accumulated total production as of October 2015 was $3.14 billion. Of this total, South Korea exported products worth $268 million, or 8.5 percent of the total output. The remainder of the total output, worth $2.87 billion, was consumed by South Korean end-users.[30] As a result, for example, about 70 percent of all underwear and 30 percent of all clothing sold in South Korea came from the KIZ.[31] In addition, a total of 679 middle and high schools in South Korea planned to procure school uniforms from KIZ factories prior to its February shutdown, and the students of those schools consequently had no uniforms when the new semester began in early March.[32]

Kaesong’s economic benefits were mutual, not a form of unilateral aid. In fact, the South reaped greater economic benefit from the KIZ than the North.

Why the KIZ was Shut Down

Why then, despite all these benefits, did President Park close the KIZ? The reason is simple: realism overwhelmed functionalism. The KIZ was founded on functionalist thinking: that accumulated functional cooperation across borders will lead to peace and integration. A realist perspective, on the other hand, called for South Korea to stop any cooperation that tipped the balance of power in the North’s favor, no matter how large the South’s gains.

North Korea’s recent nuclear test and satellite launch changed the South’s perception of Kaesong, regardless of whether Seoul’s cost-benefit analysis had shifted. In light of Pyongyang’s heightened threats, Seoul could no longer tolerate the possibility that even small amounts of KIZ funds were supporting the North Korean military. The lack of transparency in North Korea’s allocation of dollars from Kaesong contributed to this perception change.

The Future of Kaesong

The KIZ’s instability was theoretically predictable even before it began operations. The late Seoul National University professor Ku Young-rok once told me, before he passed away in 2001, that if such inter-Korean cooperative projects succeeded, then political scientists would have to award President Kim Dae-jung for breaking new ground: previously, functionalism had worked only between parties that were not antagonistic toward one another. If functional cooperation created peace and integration between military opponents like North and South Korea, the success would have opened a new theoretical horizon in the study of international relations.

The shutdown of Kaesong proved that the Korean peninsula remains far from that horizon. However, its closure must not mark the end of inter-Korean economic cooperation. Kaesong’s profitability revealed the potential of such cooperation. Its revenue model brought firsthand benefits to South Korean investors and consumers just as it supported North Korean leaders and workers, and all sides now understand the boon it represented. As long as memories of easy profits, cheap products, regular cash and hot water remain alive among Kaesong’s stakeholders, its economic model may yet be revived.


[1] In fact, the central control of all foreign exchange transactions is not an invention of North Korea. Developing nations often use this method to strengthen its control over the economy and the allocation of resources. For example, Japan enacted “the Foreign Exchange Control Law in 1933, which made all overseas transactions subject to the approval and licensing of the minister of finance.” This “control over the convertibility of yen lasted uninterruptedly until April 1, 1964.” See Chalmers Johnson, MITI and the Japanese Miracle, Stanford, CA: Stanford University Press, 1982, 119. In South Korea, the U.S. military government prohibited use of foreign exchange including dollars for domestic transactions in 1946, and this prohibition continued in one way or another until 1999 when the Foreign Exchange Management Law was abolished. See Tae-hun Kim,우리나라 외환거래제도의 이해 [Understanding the Foreign Exchange Transaction System of Our Country], Bank of Korea, August 31, 2012, Under that law, South Koreans could not hold dollars. They had to sell or deposit their dollars in the Bank of Korea. See also the history of the Foreign Exchange Central Depositing System (외국환예치집중제) at Thus, for example, when, the South Korean workers in the Middle East remitted their salaries back home in the 1970s, the dollars they sent were immediately converted into Korean won at the official exchange rate. Therefore, what ended up in the hands of their family in South Korea were in won, not dollars. See Young-hwan Park, “39호실 달러 유입 방식…70년대 한국은행도 했다” [The way the Office 39 controlled dollars … The Bank of Korea did the same in the ’70s], Kyonghyang Shinmun, February 16, 2016,

[2] This is indeed indirect payment. Brown seems to emphasize this fact as the main reason why he regards the Kaesong workers as “slaves.” In fact, Human Rights Watch has also pointed to this problem since its first report on the labor condition in the KIZ was published in October 2006. Human Rights Watch. North Korea: Workers’ Rights at the Kaesong Industrial Complex, October 2006, 3, 6, 10. See also a recent report by Phil Robertson, Deputy Director of its Asia Division. Robertson, Phil. Dispatches: Abuses in Kaesong Industrial Complex. Human Rights Watch. April 22, 2015. However, how can we directly pay the workers in the KIZ given the foreign exchange concentration system of North Korea? There are two ways: one, give the dollars directly to the workers. Then, the workers have to exchange the dollars at the official exchange rate because, legally, use of dollars for a purchase is prohibited. Two, give North Korean won to the workers. For this purpose, the South Korean companies have to exchange the dollar-denominated wages for North Korean won at the official exchange rate because they cannot get access to the black market for dollars inside North Korea. Either way, therefore, the workers lose a substantial amount of the real value of their wages in the exchanging process. This must be partly the reason why they prefer state-run commissaries because, there, they can use their ration coupons to buy goods at relatively low state-designated prices.

[3] Actually, Brown’s calculation is not new. A very similar reasoning is found from this Dong-A Ilbo article published in 2006. This article claims that the North Korean government extorted 96 percent of the average $76 monthly salary to each Kaesong worker through the exchange rate distortion. But, this article does not say anything about how the workers got access to rice and other daily necessities.

[4] Chin-hyang Kim, “개성 공단 사람들” [People of the Kaesong Industrial Zone), Naeileul Yeoneun Chaek, 2015, 59.

[5] According to Song’s calculation, in March 2006, his company spent $233,400 to import goods to refill the inventory while, in the same period, the workers in the KIZ received $264,000 after tax. In other words, the workers consumed 88.4 percent of their after-tax income at the commissaries. Jiyong Choi, “북 노동자, 자기 월급 철저히 따져, 개성공단이 핵개발 출처? 난센스” [North Korean workers calculate their salaries meticulously. Is KIZ the monetary source for nuclear development? Nonsense]. OhMyNews, February 17, 2016, See also Minseung Jeong, “개성 임금, 北 정권에 70% 상납 있을 수 없다” [70% of the Kaesong wages going to the North Korean regime? Impossible). Hankook Ilbo, February 18, 2016,

See also Digital News Team, (Footage) (영상)이해찬, 홍용표 장관에 “차라리 그만둬라” 질타 [Lee Haechan reprimands Unification Minister Hong, telling him to resign), Hankook Ilbo, February 16, 2016,

[6] Jiyong Choi, op. cit. Kim Chin-hyang says that the North Korean workers consume most of their wages in coupons and use the small remainder to purchase other daily necessities and for gifts for one another, like wedding gifts or condolence money. Again, see Kaesŏng Kongdan Saramdŭl [People of the Kaesong Industrial Zone], 59. Kim Chin-hyang was Director of Corporate Support at the KIZ Management Committee, an inter-Korean governmental organization that manages the KIZ. Before joining the KIZ in 2008, he spent five years as a director and later a senior director in the Office of the President during the Roh Moo-hyun administration, taking care of the inter-Korean relations and the Korean Peninsula Peace Regime affairs.

[7] This official worked for the KIZ Support Foundation, which works as a liaison between the South Korean government and the South Korean companies in the KIZ. He says that if KIZ employees had not received enough food, the South Korean managers in the KIZ, who worked closely with the workers day and night, would have noticed from their faces and bodies. He added that the quality of the food given to the North Koreans was evident in the lunch boxes they brought to the KIZ. “Had the commissaries been empty, how could they have made their own lunch boxes?” he asked. He referenced the babies at the KIZ daycare center as another nutritional barometer of Kaesong residents, noting that their faces “looked okay.” Conversations with this official took place over the phone.

[8] KIZ Support Foundation,평화와 상생의 개성공단 [KIZ: A Place for Peace and Coexistence], January 2016, 16. This presentation material was created by the KIZ Support Foundation to explain the current activity and the investment environment of the KIZ:

[9] This incentive system is stipulated in Article 30 and 31 of the Labor Regulations of the KIZ 개성공업지구 노동규정, which was agreed upon by North and South Korea in 2003. See also the KIZ Support Foundation, op. cit., 16.

[10] Conversation with the South Korean official from the KIZ Support Foundation.

[11] In the factories, KIZ employees worked in a relatively warm and clean environment and had access to electricity, hot water, and even clean toilets. Their homes received limited electricity and water and offered little to do, whereas the KIZ offered showers, laundry facilities and equipment for playing volleyball and other sports during breaks. The South Korean managers even provided hot soup for lunch. See Chin-hyang Kim, 153, 185.

[12] This willingness to work longer hours in turn strongly suggests the existence of the desire to save or consume more than others, the usual stimuli that drive the workers in capitalist economies to work long hours. In other words, despite its egalitarian guise, the North Korean government may be allowing the workers in the KIZ to save or consume part of the income that exceeds the base level at their disposal.

[13] Jiyong Choi, op. cit. and Chin-hyang Kim, op. cit., 58.

[14] It is worth noting that the North Korean workers did not show much interest in enhancing the productivity as measured by the number of products per hour. Instead, they sensitively responded to the signal measured by labor hours. Chin-hyang Kim, op. cit., 141. This response is reasonable because the ledgers they sign every month do not record the number of products, but the hours for which they worked. Their behavior, in this sense, is as rational as that of the workers in capitalist economies where productivity matters. The only difference is in the unit of measurement.

[15] KIZ Support Foundation, op. cit., 11.

[16] “개성특급시 [Kaesong], Wikipedia, last modified March 1, 2016,개성특급시.

[17] Chin-hyang Kim, op. cit., 60.

[18] Upon the shutdown of the KIZ on February 10, 2016, South Korean Unification Minister Hong Yong-pyo announced in a statement that, in 2015, 132 billion Korean won or roughly $110 million went to North Korea through the KIZ and that it appears that those funds were used to advance the technology for nuclear weapons and long-range missiles. Hyung-gu Kim and Su-jin Chun, “개성공단 올 스톱 … 1,320억 김정은 돈줄 끊는다” [KIZ All Stop … Cutting off the cash pipeline of 132 billion for Kim Jong-un], Joong Ang Ilbo, February 11, 2016.

[19] Jehoon Lee, “70%가 북 노동자 몫인데 … 개성공단 돈으로 핵개발은 억측” [70% goes to workers … KIZ funds for Nuclear Development? Baseless Speculation], Hankyoreh, February 11, 2016, Conceptually, the difference between the SCPC and the social insurance premium is that the former is paid by the workers out of their labor compensation while the latter is paid by South Korean companies before paying the labor compensation. In the end, however, both go to the North Korean government.

[20] I created this table based on the description given by Je-hoon Lee, op. cit.

[21] Chapter 6 of the Labor Regulations of the KIZ, where the purposes of the funds from SCPC and social insurance premium are specifically stipulated. See also Lee, Jehoon, op. cit.

[22] “양주시 [Yangju],” Wikipedia, last modified February 17, 2016,양주시.

[23] 2014년도 양주시 예산서 [Yangzhou 2014 budget]. September 19, 2014,기획예산과.

[24] City of Palo Alto. Budget in Brief for Fiscal Year 2016.

[25] “Palo Alto, California,” Wikipedia, last modified March 14, 2016,,_California.

[26] Yongdeung Song, who managed the commissaries in Kaesong, agrees on this estimate. In 2006, he testified in an interview with the Hankyoreh newspaper that the taxes that the North Korean government took off from the total labor compensation were mostly spent for the city of Kaesong. See “세금꼴 30%’는 대부분 개성시 경비로 써,” Hankyoreh, November 7, 2006,

[27] KIZ Support Foundation, 8, 10.

[28] Chin-hyang Kim, 54-55.

[29] KIZ Support Foundation, 8.

[30] KIZ Support Foundation, 8.

[31] Chin-hyang Kim, 101.

[32] “입학식인데 교복이 없어요” 개성공단 폐쇄로 3월 교복대란” [“No uniforms at entrance ceremony,” Kaesong shutdown brought ‘uniforms crisis’ in March], Chosun Ilbo, March 1, 2016,

Stay informed about our latest
news, publications, & uploads:
I'm interested in...
38 North: News and Analysis on North Korea