Peter Hayes


November 16/93

paper to the symposium

“United States and North Korea: What Next?”

Carnegie Endowment for International Peace

Washington DC

Copyright Nautilus Institute


The transfer of light water reactor (LWR) technology to North
Korea (DPRK) emerged as an important issue at the third round of
high-level talks between North Korea and the United States held
in Geneva in July 1993. In section I of this paper, I provide
some background to the negotiations to date over this issue. In
section II, I analyze the relative proliferation intensity of the
DPRK developing its present nuclear fuel cycle versus “trading it
in” for a light water reactor fuel cycle. In section III, I
appraise nuclear power technology in terms of the DPRK’s energy
economy. In section IV, I review the implications of the likely
poor economics of nuclear power in the DPRK, and examine various
constraints to transferring LWR technology to the DPRK. In
section V, I discuss critical outstanding issues that must be
resolved before LWR technology is transferred to the DPRK.


The DPRK has developed its nuclear fuel cycle capability for many
years and has obtained substantial assistance from the
international community (via the IAEA/UNDP) to this end,
especially for uranium prospecting. The specific issue of DPRK
cooperation with South Korea (ROK) on nuclear research and
development has been raised also in the Korean bilateral
commissions pursuant to the 1991 non-aggression declaration,
albeit with little progress.

The North Koreans denounced a South Korean proposal to build a
nuclear power plant on or near the Demilitarised Zone to be run
jointly. But in June 1992, they revealed an interest in light
water reactors in discussions with the Director General of the
International Atomic Energy Agency, Hans Blix. Blix had told the
North Koreans that their reactors were outmoded and uneconomic.
In response, North Korean officials recognized the economic
advantage of shifting to light water reactors (1).

After the DPRK announced its intention to withdraw from the
Nuclear Non Proliferation Treaty (NPT) in March 1993, interest
intensified in this possibility. In my discussions with senior
North Korean officials in May 1993, I asked three questions:

1) Would North Korea cooperate with South Korea on joint
development of peaceful nuclear power technology?

2) Would North Korea agree to putting its plutonium (along with
that of South Korea) under joint North-South Korean control?

3) Would North Korea change to light water reactors (LWR) if
South Korea or the international community provided the

Senior party foreign policymaker Kim Yong Sun prefaced his
response by stating that science and technology traverse
political boundaries and ideology. He continued as follows:

About the possibility of nuclear cooperation, whatever the form
and size of such
cooperation for peaceful purposes, it should be studied and
researched. Science
surpasses ideology and borders. There are several additional
documents on exchanges
and cooperation in which cooperation is scientific, not only
political and cultural. If
we seek broad scientific exchanges, why not nuclear cooperation;
but not only
nuclear, we should cooperate in all fields. In the 10 point
program [for reunification,
announced in April 1993], we also mention this issue where it
refers to everyone
making their own contribution with power, knowledge and money.
When we say
knowledge, this contains fields such as scientific cooperation
including nuclear
cooperation for peaceful purposes and not only between North and
South Korea, but
also with the international community (2).

Thus, it was no surprise that the North Koreans raised the issue
of shifting to a light water reactor technology at the second
round of high level talks in New York in June 1993. In response,
the American negotiators indicated that the United States would
support such a move as light water reactor technology is
inherently less proliferation prone than the graphite reactors
under construction in North Korea. But they suggested that the
issue was moot until the DPRK complies fully with its full scope
safeguards commitment under the NPT. Moreover, they informed the
North Koreans that the appropriate way to pursue this possibility
was to discuss it with South Korea and with Russia which has
already agreed to supply four such reactors (when the North
complies with its NPT obligations and finds a way to pay for the
transfer). There the matter rested until Geneva.

In Geneva, the North Koreans raised the reactor technology
transfer issue on July 16th after an initial round of discussions
had already been completed. The North Koreans stated that the
real source of problem in nuclear issue, they said, is their
inferior graphite nuclear reactors which they were forced to
adopt because no one would help them with anything else. They
suggested that the only way to solve the nuclear problem is for
the DPRK to adopt and to obtain light water reactor technology.

The Americans promptly agreed. They also stated, however, that
only after the immediate problem was solved in relation to
implementing the safeguards agreement, would the United States
explore ways for North Korea to obtain light water reactors.
They cautioned the North Koreans to keep in mind that the US
government does not sell power reactors. Moreover, North Korea
would have to arrange finance with private corporate suppliers.

Although the North Koreans sought (and did not obtain) an
American commitment that the DPRK should be supplied with light
water reactors, they also referred to the Russian deal to supply
them four reactors. They appeared at the Geneva meeting to be
satisfied with Russian LWR technology so long as the United
States (or someone else) finances it. In one aside, the
Americans suggested that as South Korea has light water reactors,
the North Koreans should raise the issue of finance with Seoul.

The North Koreans also stated that the best way to proceed would
be to implement their safeguards obligations step by step with
progress in achieving light water reactor technology transfer,
culminating in access to sites (they did not refer to special
inspections specifically although referring to “sites” implies
the latter). The American side promptly disabused them of this
notion, insisting that substantive discussion and measures to
transfer light water reactor technology could come only after the
DPRK was in compliance with the safeguards accord.

The text of the joint US-DPRK statement issued on July 19 in
Geneva refers obliquely to all of these issues (see Appendix 1).
One phrase states: “on the premise that a solution related to the
provision of light water moderated reactors (LWRs) is achievable”
(which refers to the variety of obstacles that have to be
overcome for the United States or any other supplier to transfer
LWR technology to the DPRK) including COCOM controls, and US
legislation on terrorism and trading with enemy states.

For all these reasons, the statement that “the USA is prepared to
support the introduction of LWRs” and “to explore with the DPRK
ways in which LWRs could be obtained” is qualified with the
phrase “including technical questions related to the introduction
of LWRs.” This phrase refers in turn to these difficult legal
and practical questions outlined above which will be discussed in
the next round of talks–should they occur.

Thus, the DPRK’s line in Geneva was new and potentially
significant. The DPRK shifted blame from US policy to the fact
that the North has inferior nuclear technology which, it
suggested, inadvertently implies that it is interested in nuclear
weapons. It signifies that the leadership in Pyongyang may have
tilted away from its anti-NPT hardline, at least temporarily. In
short, the approach taken in Geneva appears designed to keep open
a face- saving way out of the nuclear impasse created by
Pyongyang while sustaining its DPRK’s nuclear weapons option for
the moment. The LWR issue gives the DPRK a tactical advantage in
on-going negotiations as it maintains ambiguity as to its
ultimate intentions while giving the appearance of being a
confidence building measure that might increase the transparency
of the DPRK’s nuclear program., (3) Kang Sok Ju (head of the
North Korean delegation in the Geneva talks) said, for example,
that his government proposed switching to more modern reactors to
“prove the point” it does not want nuclear weapons (4).

Undoubtedly, the DPRK also aspires to match South Korea and Japan
in terms of perceived technological prowess and prestige
associated with nuclear power programs although (as I will argue
in section IV) they can ill afford to pursue this objective.

Some American officials at Geneva observed that it is easy for
the DPRK to make this move knowing that the many obstacles to
transferring light water reactor technology cannot be overcome,
at least not in a time frame that is meaningful to the nuclear
issue. Others believe that the DPRK is setting its price for
compliance at a level which requires the American side to clear
the way for upgrading trade and investment relations between the
two countries, and thus, with the rest of the world. In this
sense, nuclear technology transfer impelled by the threat of
nuclear proliferation is an excellent battering ram to pound
against the American closed door policy toward the DPRK.


The DPRK has developed the basic infrastructure for a nuclear
fuel cycle with a view to constructing and operating a nuclear
power plant. In 1991, Kim Chol Ki, Director of Science and
Technology Bureau of the DPRK Ministry of Atomic Energy Industry
told me that North Korea plans to build a 1.76 GWe nuclear power
plant as part of the third Seven Year Plan for the DPRK. He
anticipated that the plant would have four 440 MWe units
operating on a 2-on, 2-off shift to provide back up against
outage (5).

Recently, the South Korean Atomic Energy Research Institute
released a report entitled “The Present Status of Atomic Energy
Development in North Korea” according to which the DPRK has
operated a 5 MW reactor at Yongbyon since 1986; and has a 50 MWe
reactor under construction at Yongbyon due to operate in 1995,
and a 200 MWe power reactor under construction at Taechon due to
operate in 1996. The report also stated that the DPRK plans to
build a 635 MW power reactor at Sinpo on the Northeast coast (6).
An American analyst has reported a different range of reactor
sizes and locations in the DPRK than those listed in the more
recent South Korean report (7). I have assumed that the former
South Korean data is more accurate as it is consistent with the
facilities declared to the International Atomic Energy Agency
(see Figure 1) (8).

Source: Nuclear News, “North Korea’s Nuclear Power Programme
Revealed,” July 1992, p. 2.

In May 1993, I visited the Heavy Industry Sector exhibit in
Pyongyang which features a display of the DPRK’s nuclear fuel
cycle facilities. It included a scale cut-away model of the 200
MWe reactor which revealed primary and secondary heat exchange
systems for the gas coolant, and two generators. From the SPOT
satellite photographs of Yongbyon released by the Tokai Research
Image Center in Tokyo, it is evident that the Yongbyon reactors
are not intended for electricity production, as no power lines
exist to or from the reactor sites.

From this information, I infer that the DPRK’s power reactor
program commences with the 200 MWe gas cooled reactor, and not
with the reactors at Yongbyon. The proposal to shift the DPRK to
LWR technology therefore relates to this and any other nuclear
power plants that the DPRK might construct. Cases for
Comparison: The rationale for proposing to shift the DPRK from
its graphite- moderated, gas-cooled reactor program to LWR
technology is the latter’s relatively lower proliferation
proneness. Assuming that the DPRK will have to abandon its
indigenous 200 MWe reactor in order to obtain LWR technology, the
two fuel cycles must be compared with respect to two criteria
(see Table 1). First, the DPRK could be inside or outside of the
NPT and the IAEA’s full-scope safeguards system will or will not
be applied to its nuclear facilities. Second, it could have its
own or LWR technology. These possibilities produce four possible
outcomes as follows:

1. DPRK is in NPT and has only 200 MWe reactor operating in
power, not weapons grade plutonium mode, under full-scope

2. DPRK is in NPT, has only an LWR operating in power, not
weapons grade plutonium production mode, under full-scope

3. DPRK is not in NPT and has only 200 MWe reactor operating in
weapons grade plutonium production mode (worst case scenario),
without safeguards.

4. DPRK leaves NPT after obtaining an LWR, and operates it in
weapons grade plutonium production mode (worst case scenario),
without safeguards.

In this study, I will conduct the comparison of proliferation
intensity by comparing only two of the four possible cases,
namely, the DPRK outside the NPT running a 200 MWe indigenous
reactor (case B1 in Table 1) versus the DPRK inside the NPT
running an 1 GWe MWe LWR under full-scope IAEA safeguards (case
A2 in Table 1).

Table 1: Possible Reference Cases

DPRK in NPT with
DPRK out of NPT
full-scope IAEA
with no IAEA

indigenous In NPT
Out of NPT

200 MWe reactor 200 MWe indigenous 200 MWe

only reactor

2 Light water A2
reactor only In NPT Out
of NPT

LWR transferred
LWR transferred

To simplify the analysis, therefore, I assume that the United
States will hold out for the following “package” before it
seriously entertains LWR technology transfer to the DPRK:

1) the “radiochemical” laboratory or reprocessing facility will
be dismantled along with any other plutonium separation
facilities, hot cells, etc;

2) the IAEA will be permitted to resolve discrepancies between
North Korean operating records and actual plutonium separation
activities as indicated by sampling, inspection of disputed
sites, etc;

3) the IAEA Board of Governors will have determined that North
Korea is in compliance with its safeguards agreement under the
NPT which will be applied fully to the existing reactors at
Yongbyon (Alternatively, the DPRK will be persuaded to
decommission these plants in return for shifting to LWRs, but
this possibility is left open in my scenarios);

4) North and South Korea will agree to and implement an
inspection arrangement in accordance with the bilateral
denuclearisation declaration.

5) North Korea will abandon construction of its 200 MWe graphite-
moderated, gas-cooled reactor in anticipation of receipt of LWR

6) North Korean spent fuel from an LWR will be kept in holding
ponds at the reactor site or at a dedicated facility; and
plutonium in it will not be separated in offshore reprocessing
plants for recycling into LWR MOX fuel or into an eventual fast
reactor program in the DPRK;

7) North Korea will rely on external suppliers of enriched
uranium LWR fuel. I assume also that a 1 GWe LWR reactor is
supplied by South Korea (or that South Korea bankrolls Russia
which already has contracts to supply LWRs to North Korea).
Relative Proliferation Propensity: At the end of the Geneva
talks, international media reported that US officials prefer that
the DPRK adopt LWR technology because it is inherently less
suited for making nuclear weapons.

In reality, determining the relative proliferation intensity of
different fuel cycles is a complex matter. John Holdren has
suggested four factors against which different fuel cycles can be
judged for their susceptibility to diversion of fissile materials
(see Appendix 2). These factors are:

1) Quality of fissionable materials–the degree of enrichment of
uranium and the ratio of fissionable to non fissionable plutonium

2) Quantity of fissionable materials–the number of critical
masses per GWe-year of operation;

3) Barriers–the chemical barriers to the diversion and use of
fissile materials such as form and dilutants of uranium and
plutonium; and the radiological barriers associated with spent
fuel of low or high burn-up;

4) Detectability–the degree to which the fuel cycle requires new
operations or significant modifications, and/or entails
radiological releases which can be monitored effectively.

It is evident that the once-through LWR (in the case presented
by Holdren, a pressurised or PWR) and CANDU fuel cycles are
significantly less susceptible to diversion of fissile materials
than other power reactor fuel cycles., (9) It is not easy to
directly compare the DPRK’s 200 MWe reactor (even after scaling
down to account for the difference in plant size between the DPRK
plant and that assumed by Holdren) because the DPRK has not
released detailed design information for the 200 MWe reactor. It
is necessary, therefore, to define a “reference” DPRK power plant
to juxtapose to an LWR in terms of their relative proliferation
proneness. DPRK Reference Reactor: In this section, I describe
the basic physical parameters of the British plutonium production
reactors in order to “design” a reference DPRK reactor to compare
with LWR technology.

The DPRK reportedly told the International Atomic Energy Agency
that their reactors are modelled after the British Calder Hall
reactors built to produce plutonium for nuclear weapons (10).
They were graphite-moderated, CO2-gas-cooled reactors fuelled
with natural uranium metal rods clad in a magnesium allow
(“Magnox”). The second generation of four Magnox reactors were
known as Chapelcross. Both generations produced plutonium but
generated electricity as a byproduct. All eight reactors were
nominally rated at 50 MWe (net) (11). Another source rates the
early Calder Hall reactors at 225 MWth, and 41 MWe (net);, (12)
I adopt 50 MWe in this study.

When operated primarily to produce electricity, the Magnox
reactor operators typically set fuel burnup at 3-4,000 megawatt-
days/tonne of uranium fuel. The core measured about 14 meters
wide by about 8 meters high. Each fuel channel in the reactor
contained a stack of six fuel elements, each of which in turn
consisted of massive, solid rods of natural uranium metal about a
meter long and 3 cm wide. Each stack of six fuel elements
weighed about 77 kg. Each core contained about 1,691 fuel
channels for a total of assembly of about 10,146 fuel elements.
The total uranium fuel contained in the core was about 112 tonnes
of natural uranium (excluding cladding).

The fuel could be replaced in later, civilian Magnox reactors
while producing electric power by using on-line, continuous
refuelling techniques, and about three fuel channels were
refuelled per week. Spent fuel from gas cooled Magnox reactors
cannot be stored indefinitely in water because the Magnox alloy
(magnesium alloy containing 0.8 percent aluminium, 0.002-005
percent beryllium, 0.008 percent cadmium, and 0.006 percent iron)
corrodes slowly in water. (Dry storage, however, is feasible
although difficult.) Each tonne of Magnox fuel irradiated for a
1,000 megawatt days contained about 998 kg of unconverted uranium
and 0.8 kg of plutonium (13).

When operated to produce weapons grade plutonium, as they were
between 1956 and 1964, the Calder Hall and the next generation
four Cross reactors were run rather differently. Instead of
continuous refuelling, the whole core was irradiated and removed
about twice a year (allowing for about three months repair and
maintenance work). To produce very pure plutonium without the
bothersome isotopes that impede weapons production, the burnup
rate was reduced to about 400 MWdt/tonne of fuel, at which

Table 2: Relative Proliferation Intensity of LWR vs DPRK
Indigenous Reactor
PWR Once Through DPRK Indigenous
Fuel Cycle Reactor
Fuel Cycle
Per GWe-year Per 0.2 GWe-

Operated to Maximize

Plutonium Production
_______________________ ______________________
enriched spent fuel natural
spent fuel
uranium storage uranium
__________ __________ _________ __________

1. Quantity of 855 kg U235 250 kg of 336 kg of 315 kg of
fissile material in 28500 kg Pu (69%U235 in weapons grade
and main dilutants U238, 3 %fissile) in223,664 kgplutonium in
at this point enrichment26,000+ kg of U238 approx.

uranium andzero % 223,000 kg of

fission enrichmentU238 and fission

products products

2. Further extensive chemical enrichment chemical
processing neededfurther separation from scratchseparation
from this pointisotopic from uranium required from uranium to
use in nuclearenrichmentand fission and fission
products explosives required products

required for use in nuclear

explosives; storage

may require reproc-

essing of wastes

3. Proliferation Susceptibility Indices (5 = worst, 1 = best)
As is 1 3 1 4
Enrichment 5 4 5 4
Quantity 4 4 1 4
Chemical 4 2 4 2
Radiological 5 1-2 5 2
Detection 3 1 5 1
____________Source: J. Holdren, “Civilian Nuclear Technologies
and Nuclear Weapons Proliferation,” in C. Schaerf et al, New
Technologies and the Arms Race, St. Martin’s Press, New York,
1989, pp. 182-185; text for DPRK reactor. Note: see Appendix 2
for definitions of numerical weights.

rate about 79 kg of weapons grade plutonium was produced per
reactor year., (14)

On this basis, what can be said about the proliferation
propensity of a 200 MWe scale up of the early graphite-moderated,
gas-cooled reactors compared with an LWR when measured against
the factors listed above (see Table 2)?

In terms of quality, replacing the DPRK reactor with an LWR would
increase the international community’s leverage over the front
end of the DPRK’s fuel cycle by virtue of the latter’s resultant
dependency on imported uranium enrichment services.

On the back end of the fuel cycle, it would also reduce the
quality of the plutonium available from spent fuel by increasing
the amount of plutonium isotopes which may prematurely initiate
nuclear chain reaction in a weapon (unless the LWR were removed
from the NPT regime and operated to maximize the production of
weapons grade plutonium).

In terms of quantity, a 1 GWe LWR would produce about 250 kg of
plutonium per year. A DPRK 200 MWe reactor scaled up from Calder
Hall technology and operated in plutonium production mode could
produce about 315 kg of weapons grade plutonium. Thus, LWR
transfer would decrease the quantity of plutonium to be
controlled under safeguards, although only marginally. In
neither case, however, would diversion of 1 percent per year
yield a “bomb” quantity of plutonium (5 kg for weapons grade

In terms of chemical barriers, LWR technology is fairly resistant
on the front end in that the fissile material is in oxide form,
albeit not mixed with an effective dilutant. However, the gas
cooled reactor would use natural uranium fuel which would be even
more difficult to utilize for weapons purposes than low enriched
uranium oxide for LWR fuel. So long as both fuel cycles do not
introduce plutonium recycling, they are equivalent in terms of
chemical and radiological barriers to diverting spent fuel from
storage to weapons activities. Unfortunately, due to the
difficulty of storing spent MAGNOX fuel in water for long
periods, North Korea has argued that it may be obliged to
reprocess the fuel for safety reasons and has already cited
precedents to this effect in Britain, France and Japan (15). Some
experts contend that dry storage is feasible, however.

In terms of detectability of diversion, an LWR fuel cycle appears
to offer significant advantages. If we assume that the DPRK
operates its reprocessing plant in case B1 (go-it- alone with its
own 200 MWe plant outside of the NPT system) but would abandon it
along with the 200 MWe reactor in case A2 (rely on LWR
technology), then the LWR would reduce the opportunities for
diversion at various points in the reprocessing and recycling
portions of the fuel cycle from relatively high to essentially
zero. The LWR is inherently easier to safeguard as shutdown is
obvious and required for removal of any fuel rods (although the
fact that an LWR is relatively easier to control in this respect
is not relevant to the comparison with the DPRK indigenous plant
because I assume that this reactor would only operate outside the
NPT whereby diversion detectability becomes moot).

Overall, therefore, the major reduction in proliferation
intensity associated with switching to LWR technology would be 1)
the increased dependency of the DPRK on the international
community for enrichment services; and 2), the reduced
opportunity for and enhanced detectability of diversion of
plutonium from LWR spent fuel under safeguards versus an
indigenous reactor operating outside the NPT. Finally, inducing
the DPRK to abandon the 200 MWe reactor would lay to rest any
possible rationale for completing and operating its reprocessing
facility in order to safely store spent fuel. Other than these
advantages, the LWR is only marginally less proliferation prone
than the indigenous plant from a technical perspective. Other
Considerations: Six other factors offset or reinforce these
marginal technical advantages of an LWR over an indigenous DPRK

First, an LWR in North Korea could legitimate continued
accumulation of weapons- relevant skills that could be mobilised
at short notice to produce nuclear weapons from a large stock of
accumulated plutonium in spent fuel. Thus, the acquisition of an
LWR is consistent with the DPRK maintaining a posture of studied
ambiguity as to its ultimate intentions with respect to nuclear

Second, the DPRK could reduce the leverage implicit in its
reliance on imported enriched uranium fuel by stockpiling this
material (assuming it could afford to do so, and that this step
passed unnoticed by the international community).

Third, LWR or “reactor grade” fuel containing excessive amounts
of the plutonium isotopes Pu 240 and Pu 242 is still useable for
a nuclear weapon at a cost to expected yield and certainty of
yield than weapons using “weapons grade” material. Moreover, it
is not appreciably more difficult to design a weapon using
reactor rather than weapons grade plutonium (16).

Fourth, the DPRK could operate an LWR (presumably after departing
from the NPT) to minimize the production of these inconvenient
isotopes by shutting down the reactor more frequently to remove
irradiated fuel (but at a cost to electricity production) (17).

A “modernised” DPRK that is rendered capable of running (or even
constructing) an LWR could also become a more active and
disruptive exporter of nuclear technologies than it would if it
only has access to its own relatively primitive nuclear
technology. Weighing against this disadvantage of an LWR is the
fact that although the DPRK could become a more capable and
potentially disruptive supplier of nuclear fuel cycle
technologies, materials (such as graphite) and techniques, it
would be less likely to have developed and transfer nuclear
weapons capabilities under the political conditions in which an
LWR might be transferred to the DPRK. Conversely, it might
develop and share nuclear weapons-related expertise with other
states in the near-term if left to its own devices; whereas it
would take many years (up to fifteen years for advanced reactor
core components) for the DPRK to develop exportable expertise in
LWR manufacture (18).

One other issue is worth noting. North Korean officials have
noted that South Korea’s nuclear power reactors might be hit
during a war. These reactors present tempting radiological
targets (19). By the same token, a large scale nuclear power
plant in North Korea presents the South with a reciprocal
targeting option. Having a much larger reactor program (twelve
power reactors operating or under construction), the South
proffers the North 10-15 times as much radiological damage
potential as would one reactor in the North to the South. But a
large reactor in the North would make the implicit threat to
attack a radiological target in wartime a risk shared by both
sides, which in principle provides the South with a qualitatively
similar deterrent against such attack. Although an LWR might
contain much more fission products and radioactive materials than
the DPRK’s 200 MWe plant, the switch to LWR technology per se
would make little difference to this factor.

In this section, I have shown that an LWR offers some inherent
advantages over North Korea’s own reactor in terms of the
criteria of quantity and quality of fissile materials, chemical
and radiological barriers, and detectability. I also noted that
six other factors should be considered in relation to the
transfer of an LWR to the North Korea, namely: continued DPRK
ambiguity as to ultimate proliferation intention; fuel
stockpiling; the utility of LWR-grade plutonium for nuclear
weapons; the possibility that an LWR could be used to make
weapons-grade plutonium; North Korea’s export behaviour; and the
issue of radiological targeting in wartime in the Korean

In the next section, I analyze the economic soundness of a
nuclear power plant in the North Korean energy economy.


As of 1991, the DPRK planned to build only one nuclear power
plant. When that is completed successfully, North Korean
officials assert that they will develop further plants “in
accordance with the needs of national economic growth (20).”

There is little doubt that the DPRK is suffering from acute
energy shortages, both of petroleum fuels (especially in the
transport sector, probably in industry, and possibly in
fertiliser production), and of electricity. Energy Sector: As is
well known, the DPRK relies heavily on coal, hydropower, and
imported oil for its energy supplies. Table 3 shows an
approximate energy supply balance for the DPRK. This section
demarks the energy sector which accounts for the bulk of the
DPRK’s greenhouse gas emissions

Table 3: DPRK Energy Supply Balance, 1991 estimate

(10-to-the-15th-power joules)

Oil Gas Coal Electricity
Other* Total

Primary Production – – 1285.4
343.3(#) 37.7 1666 Imports
239 – 75.4 –
– 314.0

Exports – – –
?*** – –

Primary Supply 238 – 1360.8
343.3(#) 37.7 1980.5 Net
Transformation-12.6 – -314.0 –
167.5 – – 494.1 Final Consumption
226.4 – 1046.8 175.9+


Source: Economist Intelligence Unit, China, North Korea Country
Profile 1992-93, 1993, p. 80, citing Energy Data Associates.
Notes: * No accounting for fuelwood and other bioenergy
fuels. # Primary electricity production, imports and
exports are expressed as input equivalents

on an assumed efficiency of 33 percent. *** No account of
small exports of hydroelectricity to China, nor jet bunkers and

international shipping + Output basis.

The institutional arrangements in the energy sector are
complicated and reflect a high degree of functional
fragmentation. The energy sector in the DPRK has no single
specialised institutional authority or ministry responsible for
energy analysis, integrated planning and management. These tasks
are scattered in agencies and ministries as depicted below:

(a) Coal exploration, mining and supply is under the jurisdiction
of the Ministry of

Coal Mining;

(b) The electric power sector development, power generation,
distribution and sales

are the responsibility of the Electric Power Industry Commission
(see below for


(c) Energy statistics and energy planning activities are
performed by the State

Planning Commission incorporating Central Statistics Bureau under
its authority. The

State Commission for Science and Technology acts as a consulting
body in these

activities mainly providing appropriate recommendations and
software for energy plan

formulation and decision making;

(d) Supervision of energy flow and reasonable consumption of the
fuel in the transport

sector is assigned as a function of the State Transport

(e) The Ministry of Atomic Energy is in charge of development,
construction, and

power generation of nuclear power plants, as well as nuclear fuel

(f) The External Economic Affairs Commission is responsible for
purchase of crude

oil and petroleum fuels, and all imported machinery and equipment
for the energy


(g) The Ministry of Machine Building Industry is responsible for
manufacturing and

supply of domestic power equipment. Most of the research and
development work

for the energy sector is performed by the institutes affiliated
with the Academy of

Sciences, although all the above-mentioned Ministries and
Commissions have their

own research institutions;

The non-standing State Committee for Energy, chaired by the Prime

discusses and decides on major issues in the energy sector;

Research and development activities related to the energy sector
performed by

institutions affiliated with the various ministries are
coordinated by the State

Commission for Science and Technology. Appendix 3 contains a flow
chart illustrating this organizational arrangement. This
functionally differentiated and fragmented institutional
framework results in poor policy coordination and program
implementation. There is no comprehensive energy policy in the
DPRK. There is no apparent economic rationale to the existing
price structure for different energy forms. There are no even
rudimentary markets to facilitate economically efficient
transactions between energy-related supply and demand entities.
Planning and fuel allocation is also inhibited by the apparent
non-existence of a basic energy supply/demand balance in the
DPRK. Indeed, a UNDP energy efficiency improvement project in
the DPRK is meant to create just such a balance at the proposed
Center for the Rational Use of Energy. Electricity Sector:
North Korea claims to have about 12,000 MWe of installed
capacity, with an available capacity of 10,000 MWe.
Approximately 50 percent of the generating capacity is
hydroelectric, and about 50 percent is thermal, mostly coal-
fired. About 84 percent of the electrical energy is fired by
coal. The annual and dail load curves in 1989 for the DPRK are
shown in Figure 2. Generating Plant: Although there are more than
500 generating plants, only 62 major power plants are linked to
the nationally interconnected transmission system. The latter
system in turn transports about 85 percent of the generated
electrical energy. (The residual 15 percent of the electrical
energy is generated by self-reliant industrial facilities and by
small, isolated and mostly hydroelectric units.) Of the plants
linked to the transmission system, 20 are thermal (18 being coal-
fired, 2 being oil-fired), and 42 are hydroelectric. The largest
thermal unit is at Pukchang with an installed capacity of 1,600
MWe. The largest hydroelectric plant is at Supung and has an
installed capacity of 700 MWe (7 * 100 MWe turbines). The output
of the latter plant is shared by the DPRK and China.

The North Koreans run the thermal, mostly coal-fired plants as
baseload units, and use the hydroelectric plants to meet peak
load demands. When demand exceeds supply, the supply to
consumers is suppressed. The DPRK Electric Power Industry
Commission estimates that it has to accommodate a generating gap
of at least 500 MWe. Blackouts occur and loads are shedded
regularly resulting in large production losses. In the winter
(November-December), load shedding reaches 1,000 MWe due to the
accumulation of snow. In summer–particularly in March through
May–shortage of water at hydroelectric reservoirs forces the
power system operators to shed as much as 2,000 MWe for up to an
hour at a time. Bad weather can worsen the situation as storms,
old and low quality equipment, and incorrect operation of
protective devices cause the transmission system to fail.

Consequently, the quality of electric power in the DPRK is also
poor in terms of frequency (often found at 57-59 Hz well below
the permissible deviation from the standard 60 Hz) and voltage
(which frequently fluctuates). The power factor at load centers
is also low and averages 0.8 which can damage badly end use
equipment. Transmission and Distribution System: The transmission
system is isolated from neighbouring countries (except for a 60
KV line feeding power to a remote area of China). The DPRK uses
220 and 110 KV lines for bulk transmission; 60, 10 and 3.3 KV for
distribution; and 380/220 V at 60 Hz for distribution to
consumers. The Government states that 100 percent of households
and industry are electrified. As not all consumers are metered,
the exact quantity and sectoral distribution of electrical end
use are not known. The Government states that transmission
losses are about 10 percent, and distribution losses are about 6
percent. However, some observers believe that this official
estimate (like generation figures) are optimistic, to say the
least. The transmission and distribution system reportedly
urgently needs to be refurbished (see Figure 3). Generation

The DPRK government claimed that generation in 1989 was about 50-
55 TWhe. Informed observers in Pyongyang estimate that the actual
generation in 1992 was about 31-32 TWhe and that the annual
shortfall is between 10-12 TWhe. This difference reflects all
the problems of generation, load shedding, and transmission and
distribution losses referred to above.

In the DPRK’s generating plants, machinery cannot be maintained
or repaired adequately due to the shortage of spare parts,
testing equipment, and obsolete and incomplete monitoring and
control instrumentation in the power plants. The official
estimate of thermal power generation of the thermal-to-
electricity conversion efficiency of 34 percent is likely a
substantial overestimate. At the Pyongyang Thermal Power
Station, for example, major equipment is deteriorating due to the
limited capabilities to track thermal performance, poor
instrumentation and testing equipment, and the lack of a
comprehensive maintenance program. All these technical problems
are worsened by the shortage of skilled staff able to use what
equipment exists. About 211 GWhe of electricity generated at the
station (or 5 percent of its nominal and 7 percent of its actual
rated output at a 100 percent capacity factor) is lost due to
acute problems such as boiler outage, etc. Coal Shortages: The
power sector is also afflicted by problems originating in the
coal mining industry. Coal shortages (reportedly due to the
classic command-and-control bind of shortage of coal for steel
and power production on the one hand, and transport constraints
on getting coal to end users due to steel shortages on the other)
have constrained the power output at thermal power stations.
Also, the Institute for Coal Selection lacks equipment to
determine the energy content of mined coal. Consequently, power
station operators may not know the quality of fuel loaded into
steam boilers at generation plants. The DPRK lacks a long range
coal mining industry development programme and master plan for
each coalfield and basin to determine the best allocation of
investment resources in coal production in relation to projected
consumption needs. Moreover, that coal which is produced is not
cleaned before it is sent to consumers which imposes operating
and pollution problems (from ash) for power plant operators.
Perhaps 60 percent of the coal used in power plants is wasted in
inefficient combustion.

It has been estimated that the equivalent of at least 6 million
tonnes of coal is wasted in the whole country and that simply
using high temperature waste heat rationally would increase
electricity generating capacity by 400 MWe. Most of the
industrial furnaces and ovens which vent exhaust gases at
temperatures of more than 500o centigrade do not recover the heat
for preheating fuel or other uses. Nor are piping or furnace
walls insulated due to the lack of insulation materials. Almost
no use is made of modern heat exchangers or simple heat pumps.
Expansion Plans: The Government emphasises expansion of the power
sector in its plans and allocated 3 billion won during the most
recent (1987-1993) plan. It aimed to increase power capacity to
19,000 MWe and to generate 100 TWhe in 1993. These plans are
ambitious and highly unrealistic.

To this end, the DPRK is building 12 new hydroelectric plants
amounting to an additional 2,500 MWe (the largest is 800 MWe).
The Government also plans to construct 4,000 MWe of thermal power
plant ranging from 200-1600 MWe. As noted earlier, it also
proposes to add a nuclear power plant the size of which is
indeterminate. Finally, the Government intends to upgrade the
transmission network by expanding it and introducing 330 KV
transmission in the mid-nineties (to increase eventually to 500
KV). Institutional Weakness: The Electric Power Industry
Commission (EPIC) is the key power sector institution which plans
and develops power generation, transmission, distribution, and
end use sales and has ministerial status in the Government. The
organisation chart for EPIC is shown in Appendix 4.

Within EPIC, the Electric Power Dispatching Bureau is responsible
for the Electric Power Production and Dispatching Control Center
(EPPDCC) which in turn monitors and coordinates the functions of
the power system with its fifty strong staff. EPPDCC is
responsible for planning hydroelectric and thermal power plants;
monitoring the status of generating units for efficiency and
reliability of supply; monitoring the system flow of electricity
at voltage levels at or above 110 KV; planning and implementing
repair and maintenance of the system; responding to faults and
contingencies in the power system; and collecting and storing
data on system operation. It also supervises 11 regional power
dispatching centers. It is supported by the Institute of Electric
Power and Telecontrol in the areas of telecommunications and
control, computer equipment, and software. Load Dispatch
Difficulties: Given the complexity of the power system, EPPDCC
requires instant access to accurate and salient information on 62
power plants, 58 substations, and 11 regional transmission and
distribution dispatching centers. The system operators at
EPPCDD, however, rely on phone or telex messages for status
updates on the value of such parameters as voltage, current,
active power, frequency etc. at a load center, or a drop in
system frequency due to a fall in generation. Relatedly, if a
transmission line is tripped out- of-service due to a fault, then
the network configuration must be reconstituted immediately or
whole sections of the system become isolated. The slow pace and
unreliability of the information systems used by EPPDCC virtually
ensure that the system operators cannot restore the system to
working order. As of late 1992, EPPDCC operated one old desk top
personal computer to collect and analyse system performance data,
but it cannot handle the processing of planning and logging

Thus, the power system lacks an modern, automated, and
computerised supervisory and monitoring capability that can
support a load dispatching function in real time. The pilot
project underway with UNDP support to rectify this deficiency
covers four critical power plants and substations only, and will
not resolve this problem at a system level. Vast End Use Energy
Waste: In addition to the problems noted above, the consumption
at point of end use of electricity is also very inefficient in
the DPRK. The Government estimates that industries typically
waste between 30 and 50 percent of energy supplied. In the
building sector, many residential buildings are not insulated.
Typically, heating is by hot water pipes embedded in the floor
with a single on/off valve per apartment. The source of heat is
centralised, and is linked to power plant waste steam output on a
district basis. (Cooking is by bottled gas or kerosene with fuel
stored on balconies) (21). Aside from dramatically increasing
comfort levels in North Korean buildings, properly insulating
walls and windows would reduce the demand for “waste” steam from
power plants which could be used better on-site power plants to
increase the generating efficiency (or reduce fuel usage) of
electricity. The Government has recognised that large
opportunities exist to reduce energy waste and has decided to
establish a Centre for Rational Energy Use.

In short, the main characteristic of the DPRK’s power sector is
its extraordinarily wastefulness–waste in fuel production, waste
in transmission and distribution, waste in end uses of
electricity, and waste of scarce skilled labor. The DPRK’s power
sector is badly organised and managed. It cannot operate
efficiently due to obsolete equipment and procedures. It is hard
to imagine it operating effectively a modern nuclear power plant.


From an economic perspective, the DPRK’s priorities for public
investment in increasing energy services obtained from its energy
sector probably should be (in order of most to least important):

1. Improve energy efficiency in end uses, especially in large
and centralised consumers such as industrial plants and

2. Reduce energy losses in generation, transmission, and
distribution in the existing power system;

3. Increase the quality and quantity of domestic energy
resources (coal and water storage);

4. Provide new energy service capacity based on integrated,
least cost power planning which puts marginal supply options on
an equal footing with marginal end use efficiency options.

5. Construct new generating capacity as needed after all the
above priorities have been achieved. This analysis suggests that
constructing a nuclear power plant in the DPRK is likely to be a
high cost, low priority way to fulfil energy demands. The
demonstration effect of the Japanese and South Korean nuclear
power programs make it difficult to argue this case effectively
with North Koreans–but the fact that these two countries have
overinvested in a costly energy option should not disguise the
fact that the DPRK can ill-afford to waste money on a nuclear
power plant when many other options exist to supply energy
services at far lower cost, faster, and with less risk. Indeed,
continuing to divert a large fraction of North Korea’s scientific
and technological talent to a nuclear power program may worsen
significantly the chronic and pressing problems of the
conventional power sector described above. Technical Problems:
In addition to the opportunity cost of foregone energy services
that a nuclear power plant will impose on North Korea’s economy,
such a plant would also pose formidable technical challenges
including: maintaining system reliability; following load
patterns with a base load plant; safe operation; delay; and

A nuclear power plant may also be technologically ill-suited for
the DPRK power system. First, it is unclear whether a 1 GWe
plant at Sinpo (or elsewhere) will be small enough to not
threaten the power system’s stability (crudely, no generating
unit should exceed more than about 10-20 percent of the total
system capability–or the available system reserve–or the
operation of the whole system may be threatened due to unexpected
outages) (22). Detailed review of the DPRK transmission system
would be necessary to answer this question. Inspection of Table
3, however, indicates that the DPRK barely meets the reliability
criterion–assuming that its total actual generating capacity of
10,000 MWe feeds into one national, highly interconnected
transmission grid. Conversely, by the time that the DPRK might
bring an LWR on-line, the grid may have grown enough to
accomodate a large LWR.

Second, a nuclear power plant is usually operated as a baseload
plant and cannot be quickly powered up and down to follow peak
demand cycles. Ascertaining whether a nuclear Table 3:
Relationship between installed capacity and size of plant

Installed Capacity Must To Accommodate A Single
Be At Least Plant of
________________________ _________________________

850 MWe 100 MWe 3,300
MWe 300 MWe 9,200 MWe
600 MWe 20,000 MWe 1,000 MWe


R.J. Barber Associates, LDC Nuclear Power Prospects, 1975-1990:
Commercial, Economic and Security Implications, ERDA-52 UC-2, p.

power plant would be technically appropriate in relation to
demand patterns would require access to data either as yet
uncollected, or not released by the DPRK Government.

Third, it remains an open question as to whether a nuclear power
plant could be operated safely and its output dispatched, given
the parlous nature of the current power operating infrastructure
described in the previous section. Admittedly, it would take 5-7
years (if South Korea were to be the supplier and architect-
engineers) before an LWR could be built in the DPRK which would
provide some time to train power system and nuclear plant
operators. Nonetheless, the status of the current power system
does not inspire confidence that safety and operational
objectives would be achieved in a DPRK nuclear power program.
Attempting to operate an LWR (especially a Russian LWR) in the
DPRK may pose an environmental threat to domestic populations as
well as to neighbouring states already sensitive to radioactive
fallout issues in the aftermath of Chernobyl and Russian
radwaste dumping in the Sea of Japan.

Fourth, transferring an LWR will take years–many years. The
tasks of financing, site selection, power system upgrade, fuel
cycle infrastructure, fuel supply contract, technology supply and
architect-engineering contracts, training of operators and
technicians, and actual construction and testing would all have
to be completed before a nuclear power plant would deliver the
first kWhe into the North Korean power grid.

A minimum of six years (assuming South Korean financing, and
South Korean or Russian LWR technology) would be required, being
one year to set up the deal, and five years to construct an LWR.,
(23) Given the difficulties of building a nuclear power plant in
North Korea where basic legal and administrative barriers exist
to the operation of foreign firms and in which the economic
infrastructure is so poorly developed that an architect-
engineering firm would have to import virtually all supplies and
much of the requisite skilled labor force, a more reasonable
estimate of the time to complete the plant might be 8-10 years.

Finally, a GWe-sized LWR will cost upwards of US$3 billion–money
that the North Koreans do not and will not have in the
foreseeable future, given their accumulated foreign debt of US$5
billion. If the North Koreans are serious about obtaining an
LWR, then they must assume that they can persuade another state
to provide financial guarantees to private financiers to bankroll
the project, or to directly finance the transfer with a loan.
Presumably, they have in mind that South Korea might finance
either a Russian LWR, or a South Korean LWR as doing so may be
cheaper than the political and military costs of responding to a
North Korean nuclear weapons program. The DPRK may also
calculate that obtaining external financing for an LWR on this
scale might help it to revive its sagging credibility with
foreign lenders still angry at its failure to reschedule its $5
billion debt.


Thus far in this paper, I have: 1) described the emergence of
the LWR transfer issue in the context of the nuclear weapons
issue; 2) compared the relative proliferation intensity of an LWR
relative to an indigenous North Korean nuclear power reactor; and
3), demonstrated that North Korea probably will incur significant
opportunity costs if it pursues a nuclear power program rather
than cheaper and less risky ways to meet its energy needs. In
this section, I turn to the concern which lies at the heart of
the LWR transfer issue: why did the North Koreans raise this
demand and is it sensible to meet it? North Korean officials
often repeat a slogan in international meetings: “We mean what
we say and we say what we mean.” In reality, fathoming the North
Koreans’ intention has been the most difficult aspect of the past
and on-going nuclear negotiations, and the LWR transfer issue is
no exception.

In sum, the following conclusions can be drawn from the preceding
four sections of this essay:

Conclusion 1: the North Koreans raised the LWR transfer issue to
keep their options open by defining a face-saving exit from the
NPT impasse that they have created and to create a battering ram
with which to break down the US closed door policy on trade,
investment and aid to the DPRK;

Conclusion 2: an LWR presents marginal advantages over the
indigenous North Korean reactor in terms of relative
proliferation intensity; but the critical issue is the
implementation of full-scope safeguards and compliance with NPT
obligations, not the relative technical characteristics of
nuclear fuel cycles;

Conclusion 3: an LWR is probably an expensive way to meet North
Korea’s energy needs and may be dubious from an economic
perspective. In any case, demanding an LWR along with
abandonment of the DPRK’s own reactors would delay the startup of
its nuclear power reactor program by at least five years;

Conclusion 4: the DPRK is likely to insist that it retain its
existing nuclear power program and operate it under safeguards
while an LWR is transferred, in order to retain backstopping
insurance against the whole deal going sour. Although the United
States will find this stance difficult to accept, it may conclude
that keeping the DPRK in the NPT with safeguards applied to its
fuel cycle is better than having it outside the NPT without
safeguards, especially if it judges that the actual transfer of
LWR technology is unlikely to be completed in the lifetime of the
Kim regime. The North Koreans who make decisions in Pyongyang
know these facts will have drawn their own conclusions. The
corollary of these conclusions is that they seek primarily to
realise intangible benefits such as prestige, the impression of
modernity, and symbols of external recognition of the durability
of their rule; and possibly more tangible gains in terms of
reopening trade and financial relations with the external world
(see the epilogue below).

The critical issue is whether provision of an LWR will induce the
North Koreans to abandon their reprocessing plant (and possibly
their own reactors) and allow full-scope safeguards to be
implemented. If so, then providing an LWR is a cheap way to
preserve the peace and restore the nuclear non proliferation
order in Northeast Asia. If not, then the transfer issue is
simply a diversion introduced by North Korea to stall for time
while they pursue a nuclear weapons program or seek other

Given that an LWR would not exist under the most optimistic
scenario until after Kim Il Sung has passed from the scene, the
abandonment of its 200 MWe reactor and reprocessing plant, and,
by returning to the NPT fold, the resolution of outstanding
ambiguity as to the North’s residual nuclear weapons capability
arising from past reprocessing, would be a major concession by
Pyongyang. Indeed, the DPRK’s current rulers would have no
assurance that they would ever receive an LWR given the long lead
times involved. It follows that however politically important an
LWR transfer agreement might be to ensuring that full-scope
safeguards are applied to the DPRK’s nuclear fuel cycle, an LWR
cannot substitute for other benefits sought by the regime which
may have an immediate and tangible impact on its survival
prospects. These include negative security assurances, an end
to Team Spirit, and a general upgrading of US-DPRK relations.

By demanding that LWR technology be transferred, North Korea has
set a high price for complying with the NPT. But in doing so, it
has at least defined a specific way to resolve the standoff that
might be acceptable to all parties and against which progress can
be measured quite precisely. Striking this deal would also
symbolise that the United States, and by implication, the rest of
the world, recognises the political autonomy of the North Korean

It is difficult to be optimistic at this late stage in the
endgame. North Korea has barely fulfilled the two conditions
that it agreed to in Geneva–starting a serious dialogue with the
IAEA to resolve the discrepancies identified by the IAEA as to
past plutonium reprocessing, and entering into substantive talks
with South Korea. Indeed, it has backtracked by asserting that
compliance with IAEA safeguards should follow, not precede
striking a deal to transfer an LWR–a position it knows to be a
non-starter with the United States. It has also done nothing to
date to resolve the outstanding issues with the IAEA and has
refused to allow the IAEA to conduct uninhibited routine
inspections (although it did offer to let inspectors refurbish
monitoring equipment at the end of October).

Until now, the DPRK has been able to curb moves to increase
pressure on it by allowing the international community to
maintain the transparency of its current nuclear activities (24).
However, the IAEA inspectors who went to North Korea at the end
of August were unable to conduct even routine inspections, and
were barely able to maintain continuity of monitoring at declared
sites. Now that, as the IAEA puts it delicately, continuity of
observation has been “damaged,” the patience of the international
community will be tested to the limit and time will run out for
North Korea.

Shortly, therefore, we will know whether the LWR issue is simply
another siren song to seduce the naive, or if it is a strategic
commitment on the part of the DPRK intended to enable it to
reenter the international community.


Fortunately, “shortly” is an elastic word. It could be some time
before the IAEA and the DPRK identify enough common ground to
permit the United States and the DPRK to reconvene high level
talks. Also, US national technical means can substitute for IAEA
ground monitoring, at least for a time and to some extent.

In my October 19, 1993 interview with Kim Yong Sun in Pyongyang,
he made a number of significant points relating to the LWR issue.
“The LWR issue,” he stated, “will be crucial to the success or
failure of the next round of US-DPRK high level talks.”

“If the LWR issue is solved successfully,” he added, “then the
DPRK will stay in the NPT. If not, then we have no alternative
but to seek to supply energy from our own nuclear technology.”

“The DPRK doesn’t care where the LWR technology comes from,
whether it is American, Russian, South Korean.”

“But whatever the source,” he said, “the arrangement must be made
via an agreement between the DPRK and the United States.” North
Korea, he explained, fully understands that for the United States
to provide LWR technology, for example, by allowing US LWR
technology licensed to South Korean companies, to be exported to
the DPRK will entail clearing away political and legal barriers
that apply to all aid, investment, and trade between the two
countries. Indeed, that is the major reason that the LWR issue
is so important and why the high level talks will succeed or fail
according to the way that the LWR issue is handled.

“It is crucial,” he said, “that the next round of high level
talks with the United States happen very soon. Only a
comprehensive solution will work that declares that the United
States and the DPRK will together bring about the LWR transfer.
This could ease a lot of tension. If such a deal is made, the
NPT issue will no longer be a big deal and it would contribute to
the normalizing of relations between the DPRK and the United

Presuming that the immediate issues relating to the reactivation
of routine inspections are overcome and US-DPRK high level talks
are reconvened, what obstacles to and opportunities for
cooperation arise with respect to the transfer of LWR technology
to the DPRK?

In this epilogue, I analyse these obstacles and opportunities in
a hierarchy starting with high and ending with low order
questions. I conclude with some suggestions as to practical
steps toward cooperating with the DPRK that would be entailed by
LWR technology transfer, including roles that non governmental
organisations can play. 1. The overarching quid pro quo US
Objectives: Will the United States facilitate this transfer in
return for merely reactivating routine inspections; or must the
DPRK also allow special inspections to proceed? modified special
inspections? dismantle its reprocessing plant? and allow it to
be kept if inspected, but not insist that it be dismantled?
dismantle its 200 MWe indigenous reactor? or allow it to be kept
if inspected, but not insist that it too be dismantled? DPRK
Objectives: Will the DPRK insist that the United States actually
supply LWR technology (including the hardware)? commit to
ensuring another supplier transfer the technology? merely
facilitate discussions with another supplier? finance the
transfer? over what time frame? and what will the DPRK give up in
terms of fuel cycle capabilities that the United States wants
dismantled and which represent fallback insurance if the LWR deal
and related normalisation of relations go sour? What Does the
DPRK Mean by “LWR Technology Transfer”? Does the DPRK mean the
term to cover merely the supply of hardware, software, and
peopleware required to plan, construct, operate and decomission
an LWR in the DPRK? Or does it include equipping the DPRK with a
full LWR fuel cycle facilities? And/or transfer of LWR
manufacturing capabilities? In the rest of this paper, I assume
that only the first, most narrow definition of transferral is
under discussion with the DPRK. However, it is important to
clarify this point at the appropriate time with the North
Koreans. What Price is the US Willing to Pay to Keep the DPRK in
the NPT?: Is the effort worth it for the United States? For the
North Koreans, it is evidently necessary to transform their
external political and economic relations if they are to commence
the delicate process of internal reform, economic transition, and
structural adjustment. The stakes for the North are regime
survival. The nuclear lever is the only one available to it in
which domestic and external factors converged.

But for the United States, the calculus is not so loaded in favor
of an LWR transfer: a nuclear pariah state that is the exception
that proves the rule of the NPT and forces allies back into US
arms for extended nuclear deterrence may be preferrable to a
creeping proliferator which retains residual nuclear options
under the nose of the IAEA.

Conversely, the United States may be willing to pay a very high
price to preserve the regional and peninsular peace, to keep the
DPRK in the NPT in order to protect the 1995 NPT Extension
Conference, and to avoid a chain reaction of Asian nuclear
proliferation. (It should be noted that there appears to be
relatively little technical advantage in terms of proliferation
proneness of an LWR versus North Korean indigenous reactor
technology; the issue is how to keep the DPRK in the NPT/IAEA
system versus withdrawal rather than one versus another
technology.) Is There a Better Alternative than an LWR Transfer?
Is there another deal which makes more sense than LWR transfer?
Should the United States propose instead to facilitate a major
renovation of the DPRK energy sector, with particular emphasis on
coal mines, power system, and boiler technology? Such a package
deal would also entail overcome the same legal barriers; would be
more in the US Government’s purview; could be done incrementally
in smaller, faster chunks; and would have a much bigger impact on
the DPRK’s prospects for economic survival, attracting foreign
investment etc. Conversely, would the DPRK see this as a as
losing face? as hooking up its economic train too fast and too
much to an external locomotive? as foregoing its residual nuclear
option to proliferate? 2 The difficulty of moving forward
together but separately: Who Moves First? Can the two sides
edge forward together toward normalisation of political and
economic relations without admitting it? Or will the United
States insist that the DPRK fulfil its IAEA/NPT obligations down
to the last letter before any formal upgrading occurs and it
commits to facilitating an LWR transfer? Conversely, will the
DPRK accept US “concessions” (such as cancelling exercises,
declarations of no first use, negative security guarantees, and
the like) as surrogates for formal upgrading of relations, or
will it insist that the two move strictly in tandem (creating
problems for the United States with its allies)? Will it insist
that the LWR transfer be realized before it reimplements full
scope safeguards? 3 Political issues that arise include:
Political and Ideological Opposition: Overcoming the political
barriers in the United States and key allied states to allowing
LWR technology to be transferred to a proliferation-prone state.
In particular, the “non proliferation at all costs” school will
have to be overcome as well as hardline hawks who relish the
prospect of a confrontation with North Korea, their perfect
adversary. The ROK’s Reaction: Most important, how will the ROK
react? What domestic political factors will come into play in
Seoul that will affect the ROK’s support or opposition to
transferring LWR technology to the North? The IAEA’s Role: Can
the IAEA play a productive role in the transfer given its recent
history with the DPRK? (It continued to assist the DPRK on non-
politicized projects until very recently to keep the door open to
Pyongyang.) Is a US Commitment Credible to the DPRK? Given the
problems adduced above and below, is a US commitment to effect
the transfer credible to all players in Pyongyang? Is this issue
amenable to external inputs of any kind? 4 Obstacles to Transfer
include: Obtaining Congressional Approval: Negotiating a deal
that is acceptable to not only the DPRK, but to all parties that
must be consulted and agreeable inside the United States,
especially in Congress. The relevant acts are quite stringent in
this regard, particularly with respect to the legal obligations
of the Nuclear Regulatory Commission (25). The Legal Barriers:
Skirting the thicket of legal barriers to allowing a strategic
technology to be transferred to North Korea, including COCOM, the
London Suppliers Group/Zanger List, Terrorism Act, Trading with
the Enemy Act, Nuclear Non Proliferation Act, and many (twenty
plus) other US laws; and, in the ROK–the only likely supplier of
LWR technology to the DPRK (see below), what legal obstacles have
to be overcome given its own nuclear export controls, both for
ROK nuclear technology, and for US-licensed technology exports?
Who Might Finance the Transfer? Financing the deal given that
the DPRK is bankrupt and owes banks and creditors about $5
billion. The only conceivable source for the $2 billion+ that
would be required is the ROK. The United States has virtually no
manufacturing plant on line for making LWRs (although some
components or parts of a second hand reactor from a US utility
might be available cheap, or the BNPP in the Philippines) and
even less political will to finance such an export., (26) Russia
could supply the technology but not the finance, and it is
difficult to conceive of barter trade on a scale that would meet
the bill. Japan must resolve the reparations problem before the
DPRK will entertain a specific deal like the LWR transfer. No-
one wants France to be involved. What Role Might South Korea
Play in the Transfer? That leaves South Korea. What kind of
financing package might be involved? Apart from a major
government-financed grant-in-aid, what kind of loan-cum-in-kind-
repayment deal might be negotiable? Could the DPRK repay the
loan in raw materials? by exporting electricity from the plant
via a linked grid across the DMZ? Does the ROK actually have the
full complement of LWR-related manufacturing capabilities that it
claims? Building an LWR in the DPRK: Constructing an LWR in the
DPRK would be a nightmare. There is almost no supporting
infrastructure. Materials and services are of very poor quality,
so all steel and concrete as well as every nut and bolt of
machinery, plus all the supporting suppliers of incidental and
routine goods and services for large scale power plant
production, all of this and more will have to be imported. A ROK
supplier will have advantages in this regard: its management,
skilled and construction labor speak the same language as their
compatriots; they have large stocks of the relevant nuclear-
specific materials and items produced up to US nuclear
engineering and manufacturing standards, plus a well developed
set of supporting suppliers of goods and services. Time Horizon:
DPRK decision makers may not fully realise the time required to
plan, construct, and complete an LWR. The DPRK has never
undertaken an industrial project on the scale and complexity of
an LWR plant. Its industrial and construction culture is attuned
to massively engineered, low technology, labor-intensive
approaches that have no or negative bearing on nuclear power
plant construction techniques. It will take at least 5-6 or more
likely 8-10 years before the DPRK sees the first kWhe from an
LWR. This time horizon is beyond the political lifetime of the
current generation of gerontocrats in Pyongyang. It is not clear
that the decision makers who will inherit this legacy will want
to complete the project. If so, then the suppliers and financiers
will incur additional risk of project non-completion and DPRK
non-payment of the financing. The rulers-in-the-wings may also
not thank the suppliers for locking them into a nuclear white
elephant (the Aquino precedent is relevant here). Operation and
Maintenance: North Korea’s electric agencies are singularly ill-
equipped to operate a nuclear power plant. Also, their grid may
be technically inappropriate for a large (GWe) LWR due to the
reliability criterion (density of interconnection and peak load
relative to size of biggest generation unit). Although the
performance of system operators can be upgraded during the LWR
construction period, there are practical limits on what can be
done in this regard, even in 5-6 years. Technical constraints
include: deteriorating fuel supply, generation, transmission and
distribution, and end use equipment; and almost non existent
system control and dispatch capabilities in real time. Cultural
and institutional constraints include organisational pathologies
associated with forty years of command and control economics,
standard operating procedures that are incompatible with safe and
economic operation of an LWR etc. Uneconomic Front and Back End
Fuel Cycle Facilities: Also, with only one LWR (which is all it
could ever hope to obtain, whatever the pretensions of its
Ministry of Atomic Energy Industry), the DPRK would not be able
to operate economic front (uranium supply and fuel fabrication)
and back end (storage except racked on the LWR site, and
disposal) fuel cycle facilities. Perhaps it is best to assume
that by the time an LWR comes on line, the two Koreas will be
merging and South Koreans would staf and operate the DPRK’s LWR;
if there are still two separate states, perhaps South Koreans
could be seconded to the North Korean nuclear agency. Until a
couple of years ago, DPRK nuclear officials assumed that they
could reexport the spent fuel to the former Soviet Union; now,
they don’t know what they will do with spent fuel any more than
their ROK counterparts. Safety: There probably isn’t much
difference in the relative hazard of the indigenous DPRK nuclear
reactor (higher chance of catastrophe with less technological
isolation from the biosphere, but less curies of radioactivity in
a smaller core) versus an LWR (less likely to crash with more
barriers against release, but bigger load of radioactive
materials). Training and technical assistance in site selection,
operating procedures, radiation monitoring, and accident and
emergency response procedures would be important aspects of a
cooperative approach to an LWR program in the DPRK. Also, the
DPRK would need to set up from scratch a sound regulatory
framework including an independent nuclear regulatory agency. 5
Practical Cooperation with the DPRK

This list of obstacles to successful transfer is daunting.
Equally, each obstacle represents an opportunity for possibile
cooperation and dialogue with the DPRK, even if the final outcome
is not a realised LWR transfer. I draw the following conclusions
from the preceding sections.

Conclusion 4: Governments must play the primary role if a
transfer is ever to be achieved. Only Governments can mobilize
the requisite resources to address a number of the critical
issues listed above.

Conclusion 5: Non governmental organisations have a role to
play, but to be effective, they must enter the field only in
areas where their flexibility, informality, and speed can help
the negotiating parties to come to grips with and resolve
critical issues. With a strategic approach aimed a key pressure
points, NGOs can complement official work in all three capital
cities involved in this question.

Conclusion 6: Governments are currently engaged in short term
maneuvering and hard bargaining on other critical issues that
will determine whether another round of high level talks take
place this year. Very little hard work on the core issues
involved in an LWR transfer has been undertaken within any of the
Governments. All three Governments with most at stake in the
DPRK nuclear issue must become much more informed about the
potential for and obstacles to cooperation if the LWR issue is to
become a practical plank of cooperation rather than another issue
of contention.

Conclusion 7: NGOs have a comparative advantage in their
ability to quickly address some of the critical issues that will
face Governments (see below). They are unlikely to have much to
offer in terms of defining legal and political barriers as legal
counsel in the State Department and the Pentagon have already
reportedly completed this analysis. Indeed, their main task may
be to educate and restrain the more ideological anti-nuclear
opponents who may take the US Government to court using NEPA,
mobilize Congressional and media opposition in order to block the
transfer, etc. Such educational meetings should be convened
sooner rather than later, especially in Washington DC, and could
be usefully undertaken by Carnegie Endowment; Nuclear Control
Institute; Natural Resources Defense Council; etc.

Conclusion 8: In particular, NGOs could enter into dialogue
with South Korean NGO and QANGO (quasi autonomous NGO)
counterparts to clarify what reaction might be expected from
Seoul to the ROK being the LWR supplier; and what issues and will
arise and obstacles have to be overcome should it become the
major source.

Conclusion 9: NGOs could also usefully enter into a dialogue
with the DPRK Government to provide it with a better
understanding of the critical economic, legal, and technological
issues surrounding LWRs; in particular, they could address the
relative economic and environmental performance of Russian versus
US LWR technology. The Center for Energy and Environmental
Studies at Princeton University; the Union of Concerned
Scientists; the Federation of American Scientists could all
supply such briefing missions at short notice. Such a mission
could include some experienced Korean American nuclear engineers
with construction experience in the ROK, likely from Bechtel or
from Westinghouse companies.

Conclusion 10: NGOs could also explain to DPRK decision makers
some of the opportunity costs and possible advantages of
switching its demand from LWR to energy efficiency and energy
supply technologies. The International Institute for Energy
Conservation with its Thai office; or the International Energy
Efficiency Initiative, with its Indian base in Bangalore, could
play an important role in sending
briefing missions to the DPRK on the latter issue.

The delegations of the United States of America (USA) and the
Democratic People’s Republic of Korea (DPRK) met from July 14-19,
1993, in Geneva for a second round of talks on resolving the
nuclear issue.

Both sides reaffirmed the principles of the June 11, 1993 joint
USA/DPRK statement.

For its part, the USA specifically reaffirmed its commitment to
the principles on assurances against the threat and use of force,
including nuclear weapons.

Both sides recognize the desirability of the DPRK’s intention to
replace its graphite-moderated reactors and associated nuclear
facilities with light water moderated reactors. As part of a
final resolution of the nuclear issue, and on the premise that a
solution related to the provision of light water moderated
reactors (LWRs) is achievable, the USA is prepared to support the
introduction of LWRs and to explore with the DPRK ways in which
LWRs could be obtained.

Both sides agreed that full and impartial application of IAEA
safeguards is essential to accomplish a strong international
nuclear non-proliferation regime. On this basis, the DPRK is
prepared to begin consultations with the IAEA on outstanding
safeguards and other issues as soon as possible.

The USA and DPRK also reaffirmed the importance of the
implementation of the North-South Joint Declaration on the
Denuclearisation of the Korean Peninsula. The DPRK reaffirms that
it remains prepared to begin the North-South talks, as soon as
possible, on bilateral issues, including the nuclear issue.

The USA and the DPRK have agreed to meet again in the next two
months to discuss outstanding matters related to resolving the
nuclear issue, including technical questions related to the
introduction of LWRs, and to lay the basis for improving overall
relations between the DPRK and the USA.

Source: Reuter’s wire service, July 19, 1993.
(remaining appendices and figures are available in hard copy
version of this paper. Please contact Nautilus Institute.)


Definition of ranking in factors

Source: J. Holdren, “Civilian Nuclear Technologies and Nuclear
Weapons Proliferation,” in C. Schaerf et al, New Technologies
and the Arms Race, St. Martin’s Press, New York, 1989, pp. 182-
185; cited by permission of the author.

Source: Author’s files

Source: Author’s files
Figure 2. DPRK Annual and Daily Load Curves, 1989
Figure 3. Transmission and Distribution System

1. R. Jeffrey Smith, “North Korea May Consider Reducing Atom
Program,” Washington Post wire service story, June 22, 1992.

2. P. Hayes, “Report on Trip to Pyongyang, May 8-11, 1993″,
Nautilus Pacific Research, Berkeley, California, May 1993.

3. Pacific Rim Intelligence Report, North Korea Studies Ways
to Make Nuclear Program ‘Transparent,'” Yonhap Friday June 11,

4. A. Higgins, “Korea, Reactor,” Associated Press wire
story, July 19, 1993.

5. Briefing from and interview with Kim Chol Ki, Director of
Science and Technology Bureau, Ministry of Atomic Energy
Industry, Pyongyang, October 4, 1991.

6. Kim Hong-muk, “Energy Institute Reports Status of DPRK
Nuclear Facilities,” Dong-a-Ilbo, Korean July 8, 1993, p. 2;
cited in FBIS-EAS-93-130, July 9, 1993, pp. 28-29.

7. J. Bermudez, “North Korea’s Nuclear Infrastructure,” Asia
Pacific Defence Review, volume 1, 1993, p. 4-8.

8. Nuclear News, “North Korea’s Nuclear Power Programme
Revealed,” July 1992, p. 2.

9. See Nuclear Energy Policy Study Group, Nuclear Power:
Issues and Choices, Ballinger, Cambridge, Massachusetts, 1977, p.
404; and American Physical Society, “Report to the APS by the
study group on nuclear fuel cycles and waste management,” Reviews
of Modern Physics, volume 50, no 1, part II, January 1978, p.

10. Nuclear News, “North Korea’s Nuclear Power Programme
Revealed,” July 1992, p. 2.

11. Nuclear Assurance Corporation, Nuclear Materials and Fuel
Cycle Services, Sources, Inventories and Stockpiles, report to US
Arms Control and Disarmament Agency, volume 1, September 1983, p.

12. D. Gurinsky and S. Isserow, “Nuclear Fuels,” in T.
Thompson and J. Beckerley, The Technology of Nuclear Reactor
Safety, Reactor Materials and Engineering, volume 2, MIT Press,
Cambridge, Massachusetts, 1973, p. 74; personal communication
from John Simpson, October 13, 1993.

13. Nuclear Assurance Corporation, Nuclear Materials and Fuel
Cycle Services, Sources, Inventories and Stockpiles, report to US
Arms Control and Disarmament Agency, volume 2, September 1979,
pp. IV-4, IV-5.

14. Personal communication, Frans Berkhout, September 14,
1993; D. Albright, F. Berkhout, W. Walker, World Inventory of
Plutonium and Highly Enriched Uranium, 1992, Oxford University
Press, New York, 1992, pp. 41-42.

15. D. Albright et al, World Inventory, p. 72; M. Hibbs,
“South Korea Renews Quest for Plutonium Separation Ability,”
Nucleonics Week, October 29, 1992, p. 7.

16. See J. Karsen Mark, “Explosive Properties of Reactor
Grade Plutonium,” Journal of Science and Global Security, volume
4, 1993, pp. 111-128; and J. Karsen Mark, “Reactor- Grade
Plutonium’ Explosive Properties,” Nuclear Control Institute,
Washington DC, August 1990.

17. J. Holdren, “Civilian Nuclear Technologies,” op cit, p.

18. S. Droutman, International Deployment of Commercial
Capability in Nuclear Fuel Cycle and Nuclear Power Plant Design,
Manufacture and Construction for Developing Countries,
Westinghouse Electric Corporation report to Oak Ridge National
Laboratory, ORNL/Sub-7494/4, October 1979, p. 6-122 and 10-9.

19. B. Ramberg, Destruction of Nuclear Energy Facilities in
War, The Problem and the Implications, Lexington Books,
Cambridge, Massachusetts, 1980, p. 90.

20. Briefing from and interview with Kim Chol Ki, Director of
Science and Technology Bureau, Ministry of Atomic Energy
Industry, Pyongyang, October 4, 1991.

21. W. Fawcett, Modernisation of Construction Design and
Calculation Centre, Pyongyang, DPR Korea, mission report to the
UN Centre on Human Settlements, October 1990, p. 17.

22. See R.J. Barber Associates, LDC Nuclear Power Prospects,
1975-1990: Commercial, Economic and Security Implications, ERDA-
52 UC-2, p. 11-8.

23. Reuters, “South Korea may help North convert nuclear
reactors,” wire story, July 24, 1993.

24. S.W. Cheong, “North Korea’s Nuclear Problem: Current
State and Future Prospects,” Korean Journal of National
Unification, volume 2, 1993, p. 102.

25. See V. Gilinsky and W. Manning, A U.S. Light Water
Reactor for North Korea: The Legal Realities, Northeast Asia
Peace and Security Network, Nautilus Institute, Berkeley,
December 2, 1993.

26. See S. Levy, Supply of Light Water Reactor(s) to
Pyongyang: Technological Issues and Their Possible Resolution,
Northeast Asia Peace and Security Network, Nautilus Institute,
Berkeley, December 2, 1993.