Creating a `one house` culture at the IAEA through matrix

Creating a `one house` culture at the IAEA through matrix
SIPRI Policy Brief
January 2014
robert e. kelley
The International Atomic Energy
Agency (IAEA) is charged with
promoting nuclear energy around
the world and with verifying that
nuclear materials in states that have
entered into safeguards agreements
are only used in peaceful activities.
The agency derives much of its
funding and worldwide reputation
from enforcing nuclear material
safeguards under the 1968 NonProliferation Treaty (NPT), but it
describes itself as being built on
three pillars: (a) safety and security,
(b) science and technology, and
(c) safeguards and verification.
The current IAEA structure
reflects this pillar structure: each
pillar is responsible for hiring its
own staff, setting its own goals and
establishing its own standards.
They have largely separate budgets
and missions within the IAEA
and often there is little overlap or
communication between them.
This leads to ‘stovepiping’: the
idea that each pillar is separate
from the other.1 Communication
between the pillars is not
encouraged, with the result that
resources are not efficiently shared
and creativity is not propagated
horizontally. Although the pillars
share some common resources,
such as budgeting, payroll,
travel, conference services and
similar administrative functions,
technical information is largely
compartmented. 2
This business culture is not
unique to the IAEA. For example,
large research laboratories often
have multiple programmes with
multiple sponsors, which operate
independently. However, the
most successful laboratories use
the multiple technical skills that
they hire and develop across all
programmes. This increases such
a laboratory’s competiveness and it
offers exciting technical variety to
its highly skilled staff.
1 The current IAEA stovepipe structure
is illustrated in the organizational chart at
2 See e.g. Findlay, T., Unleashing the Nuclear
Watchdog: Strengthening and Reform of the
IAEA (Centre for International Governance
Innovation: Waterloo, ON, 2012).
The former Director General of
the IAEA, Mohamed ElBaradei,
frequently expressed his hope that
the IAEA could end this stovepiping
to become ‘one house’—that is, that
w The International Atomic
Energy Agency (IAEA) is
currently structured along
largely programmatic lines,
with each programme
recruiting and training its own
staff. This process is wasteful as
it creates bureaucratic
disincentives to sharing people
and resources.
Many technical
organizations in the business
and government sectors deal
with this problem by using
matrix management. In matrix
management, programme
managers budget their funds
and set programmatic
objectives. This gives them a
great deal of control over their
programmes and the desired
outcome. Staff, who are
recruited as technical
specialists by technical service
organizations, not
programmes, are then able to
work on a multitude of
The IAEA acknowledges that
it would like to work as ‘one
house’ and this has been a goal
of senior management for some
years. Structural obstacles have
meant that there has been little
progress towards changing the
management culture to the ‘one
house’ goal. A thoughtful
reorganization of the IAEA
along programme–matrix lines
could go a long way towards
reaching this goal.
sipri policy brief
the whole of the IAEA would work
cooperatively together on all its
missions. 3 He and the other IAEA
staff frequently expressed their
disappointment that this did not
The ‘one house’ policy was feebly
implemented by asking departments
to cooperate, but there were few
incentives to do this
Mohamed ElBaradei frequently
and power continues
to reside in the hands
expressed his hope that the IAEA could
of managers who
become ‘one house’
control large budgets
and have a stranglehold over career
opportunities of their staff. Many
of the managers are recruited from
outside the IAEA for limited-term
positions. They have little incentive
to put long-term solutions in place
because they will leave in a few
years. Safeguards managers, in
contrast, are usually promoted
from within. They often have core
support from long-term tenured
safeguards staff who understand
that loyalty and compliance are
keys to promotion. They have little
incentive to innovate.
In the technical units of the
IAEA today, professional staff
are hired, trained, housed in
work units, evaluated and given
career guidance by line managers
who are subordinate to a de facto
programme manager. This creates
inefficiencies and conflict and
virtually ensures that resources,
especially human resources, are not
shared and that information is not
distributed reasonably across the
3 E.g. Elbaradei, M., Statement to the
IAEA General Conference, Vienna, 27 Sep.
1999, <
aspx?orig_q=RN:30044544>; and International
Atomic Energy Agency, General Conference,
The Agency’s Programme and Budget for
2004–2005, GC(47)/3 (IAEA: Vienna, Aug.
2003), ‘Major Programme 7’.
4 Findlay (note 2).
Most of the IAEA’s mid-level
safeguards managers have a
background in nuclear material
accounting and verification. They
have little or no experience in,
for example, the skills required
to do state-level safeguards
analysis such as exploitation of
open sources, analysis of satellite
imagery and synthesis of data. As
a result, they view the staff who
carry out analytical functions
as ‘support staff’, subordinate
to safeguards inspectors. If the
‘support staff’ were to support all
of the IAEA’s programmes, and
not just safeguards, their overall
contributions would soon become
‘One house’ is not achievable in
the IAEA’s present organization
because the sharing of resources,
skills and staff on the agency’s
technical side is not rewarded;
indeed, the current programmatic
goals actively discourage sharing
of resources. This situation can
be remedied by changing the
organization of the IAEA technical
structure to a programme-based
matrix structure supported by
technical disciplines managed by
technical experts.
In matrix management, programme
managers budget their funds and
set programmatic objectives. Staff
are recruited as technical specialists
by technical service organizations
and can be called on to work on a
multitude of programmes. This
gives the programme managers
a great deal of control over their
programmes and the desired
Because technical specialists
can work for a number of
creating a ‘one house’ culture at the iaea
programmes, they can use their
technical skills to benefit the whole
organization. They are trained
to support all programmes to
which they could contribute and
are certified to conduct multiple
tasks for more than one internal
programme. Each member of
staff resides in a technical unit, so
the unit’s management can have
appropriate technical training
to evaluate the staff member for
critical assignments and can make
more informed decisions about
promotion and retention than if
staff are supervised by a generalist
with no experience in a particular
discipline. This is a powerful
morale-building factor for staff who
want to associate with professionals
in their own fields and be evaluated
fairly by people they respect.
A programme–matrix
organization is used by many
large technical organizations and
companies. 5 They see it as the
best way to provide a technically
satisfying work environment for
highly trained staff. It also drives
bottom-up innovation and allows
maximum flexibility in the use of
all qualified employees across the
organization, assigning them to
work where they are most needed.
Under a programme–matrix
organization, staff gravitate towards
managers who are forward-looking
and programmes that provide
challenges. If the IAEA were
to adopt such a structure, then
IAEA managers who treat skilled
professionals as merely support
staff would find that they must
5 E.g. Johnson, R., ‘Advantages &
disadvantages of matrix organizational
structures in business organizations’,
Demand Media, <http://smallbusiness.>.
restructure their programmes to be
competitive and innovative if the
best of the agency’s technical staff
want to work on their programmes.
Matrix management in the IAEA
The IAEA currently has several
clear programmes: (a) nuclear
materials verification under
the NPT; (b) nuclear safety
standards; (c) nuclear security;
and (d) technical cooperation.
In the future there could even be
new mandates such as verifying
a possible fissile material cut-off
treaty (FMCT). 6
Each of these programmes
requires different training and
standards. Currently, each function
uses many people
A programme–matrix organization
with similar skills
allows maximum flexibility in the use of
who are tied to
their specific
all qualified employees across the
programme and
not encouraged to
cooperate with other programmes.
Under a programme–matrix
organization, the engineers and
scientists currently recruited
individually for separate IAEA
programmes would instead be
employed in work groups with
shared skills and interests, allowing
them to be managed more efficiently
and favourably.
The IAEA already uses a matrix
structure for its administrative
tasks. Most non-technical
services are provided by a single
department, the Department of
Management. This department
provides key finance, travel,
personnel, conference, medical and
information technology services,
among others, and its legal and
6 See e.g. Kile, S. N. and Kelley, R., Verifying
a Fissile Material Cut-off Treaty: Technical and
Organizational Considerations, SIPRI Policy
Paper no. 33 (SIPRI: Stockholm, Aug. 2012).
sipri policy brief
Board of
Power and
Fuel Cycle
Safety and
UN Security
Department of Technical Services
Employee Health and Safety
Human Health
IAEA Marine Laboratory, Monaco
Imagery Analysis
Internal Security
Nuclear Industry Technology Specialists
Nuclear Materials Accounting
Nuclear Reactor Technology Specialists
Open Source Evaluation
Physical and Chemical Sciences
Radiochemistry and Sample Analysis
Technical Support
Technical Training
Seibersdorf Laboratory
Statistics and Trend Analysis
Department of Management
Budget and Finance
Conference and Document Services
General Services
Human Resources
Information Technology
Procurement Services
Public Information
Figure 1. A proposed programme–matrix structure for the IAEA
external relations divisions are
privy to virtually all of the IAEA’s
confidential information. The
ability, loyalty and discretion of
professionals in management
services to support technical
colleagues in all agency roles—for
example, in handling highly
confidential information related
to several programmes—is never
questioned. It should be the same
with scientists and engineers.
The IAEA should recognize that
its three pillars are programmes
in the management sense, not line
functions. The managers of these
programme activities should be
involved in programmatic planning,
financial planning and assessment
of results. If the programme
managers were required to work
together to plan and share resources
for all the programmes and to
reach compromises necessary for
the health of the whole agency, the
IAEA would become ‘one house’
out of necessity. The programme
managers need not be directly
involved in the hiring and staffing of
the technical services that make the
agency function. Figure 1 illustrates
how a programme–matrix
creating a ‘one house’ culture at the iaea
organization at the IAEA might be
Satellite imagery: an example of
the potential of matrix
Satellite imagery is an example
of an agency-wide capability that
would benefit all of the technical
programmes. Satellite imagery is
an extremely powerful modern tool
for geospatial awareness, analysis
and communication of complex
The first part of the IAEA to
develop satellite imagery was the
Department of Safeguards, which
has made a huge investment in
hardware, software and trained
personnel. However, it has isolated
these capabilities and did not
share or market them within the
agency until the extraordinary
circumstances of the Fukushima
Daiichi accident in 2011 led to
some reluctant sharing. If, instead,
staff for a satellite unit were hired
and supervised by a professional
with expertise in the imagery field
charged with providing services
to the whole agency, and not just
programmatic skills associated with
verification activities, this would
create a stronger team that serves
all parts of the IAEA. For example,
the satellite imagery section could
also play a huge role in the success
of the Nuclear Safety and Security
Not all of the barriers to sharing
are internal. Member states have
many preferences about how their
resources are used and can have a
very narrow outlook. The satellite
imagery section has been heavily
subsidized by one member state,
which has provided substantial
funding and guidance. Member
states should coordinate their own
efforts to ensure that their resources
are used to support all agency
missions and not just one.
Even within the satellite
imagery unit, there is subcompartmentation. This means
that some vital image indicators of
nuclear activities or proliferation
cannot be freely
If the programme managers were required
shared. But
to work together to plan and share
governments that
engage in satellite
resources, the IAEA would become ‘one
imagery analysis
house’ out of necessity
for their own
security needs long ago realized
that good cross-communication
within an imagery organization
is essential for success.7 While
the IAEA’s information security
is often mentioned as a problem,
this is really an excuse to avoid
making changes. The IAEA trusts
its administrative employees to
protect information from multiple
programmes; scientific and
technical employees deserve the
same respect. 8
The compartmentation of work
on satellite imagery is a reflection of
an existing management experience
base, trained in nuclear materials
accountability, that does not
understand the unique issues of this
technical area and consequently
treats it overcautiously. While
IAEA member states have security
concerns about the use of satellite
images of their territory, such
concerns are common to all levels
of safeguards verification, whether
it involves instrumentation,
inspection procedures, open source
analysis or imagery. If the false issue
of security is dealt with, satellite
7 Brugioni, D. A., Eyes in the Sky: Eisenhower,
the CIA and Cold War Aerial Espionage (Naval
Institute Press: Annapolis, MD, Mar. 2010).
8 Norris, P., Watching Earth from Space:
How Surveillance Helps Us—and Harms Us
(Springer: Dordrecht, 2010).
sipri policy brief
imagery could become a valuable
tool for the whole of the IAEA.
The current pillar structure trains
its captive personnel in only the
skills required for a particular
programme. For example, IAEA
safety experts carry out safety
inspections around the world at
the request of countries wanting
outside expertise; IAEA technical
specialists travel
The current pillar structure trains its
virtually everywhere
captive personnel in only the skills
in the world
required for a particular programme
dispensing technical
advice; and IAEA
safeguards inspectors travel to all
the countries that have declared
nuclear materials and perform
Even among safeguards
inspectors there are large variations
in tasks. Some perform mostly
auditing functions, examining
books with nuclear materials
accountancy records, while other
inspectors perform spot checks of
nuclear materials using specialized
instrumentation. These are two
largely separate tasks. Another
safeguards task is to verify that the
design of a nuclear facility is exactly
as declared by the state. Experts in
the Department of Nuclear Safety
and Security might well be better
qualified to carry out this task
than materials auditors from the
Department of Safeguards, but in
the rigid line organization of today
this does not happen because the
safety experts are not certified
safeguards inspectors and thus
cannot carry out safeguards-related
duties in member states.
Under a matrix system, any
professional could receive
basic training to reach a level of
proficiency that would allow them
to carry out simple inspections
and visits consistent with their
specialties. They would then receive
additional training for as many
areas as they are capable of and
needed for.
For example, all IAEA staff who
travel and carry out any kind of
inspection would receive basic
training in inspection procedures,
policy, ethics and safety. They
would then be additionally certified
in technical skills such as nuclear
materials auditing and accounting,
design information verification,
nuclear safety inspections or
nuclear security evaluation. They
may also receive specialized
certification in nuclear fuel
reprocessing or in enrichment (both
applicable to all the cases above).
There can be many variations, but
no one inspector is likely to need or
receive all of these certifications.
A training process like this could
hugely increase the pool of qualified
inspectors from among IAEA staff
and quickly identify those best
qualified for a task based on their
skills, experience and certification.
These inspectors would naturally
work on more than one programme
as needed, and the stovepipes that
impede the goal of ‘one house’
would be dismantled. Internal
communications would be naturally
In addition, a restructuring
would be an ideal time for the IAEA
to create employee health and
safety units, and an overarching
information security programme.
The IAEA must be unique among
high-technology organizations in
not having a visible and effective
employee safety programme.
creating a ‘one house’ culture at the iaea
Safety in hazardous work, such as
with nuclear materials, industrial
machinery, high voltages, and global
health and safety threats (including
disease and poor standards) dictate
that safety should be a high priority.
This is a chance to rectify that flaw.
Similar arguments can be
made about security. Information
security should be an agency-wide
activity with one set of standards.
It should not be the chaotic
province of individual programme
Rotation policy
The rotation policy has a goal of
ensuring that a majority of staff
leave employment with the IAEA
within seven years and that few
staff are given contracts longer
than seven years (essentially
tenure).9 Under a matrix system,
professionals would be supervised
and evaluated by other professionals
in a similar discipline. This
would allow for more objective
performance appraisal with a goal
of identifying the most promising
staff for longer-term contracts.
Under such a system the proportion
of staff who leave the agency might
A key component of any
evaluation system is judgment
and objectivity in managers: if
managers evaluate employees fairly
and objectively, the organization
benefits greatly and can retain
and reward the best performers.
Conversely, if evaluations are pro
forma or subjective, hiring and
rewards can be based on favouritism
and the avoidance of confrontation.
The existing rotation system makes
9 Statute of the IAEA, approved 23 Oct. 1956,
entered into force 29 July 1957, <http://www.>, Article VII(C).
decisions easier for managers who
want to avoid confrontation.10
Employees who do not conform
can be easily let go without any
formal justification, while those
who quietly comply with orders can
expect special treatment after seven
years and then benefit from job
Under a programme–matrix
system at the IAEA, managers
would be trained in the same
skill sets as
Under a programme–matrix system at the
their immediate
IAEA, managers would be trained in the
which would
same skill sets as their immediate
greatly increase
the chances of
impartial and accurate evaluations.
This would benefit both the agency
and the employees on the basis of
best value and fairness.
Country officer points of contact
Each existing IAEA department
has country officers for its
regions, countries and facilities.
Communication and coordination
across departmental lines is actually
discouraged. Thus, the departments
of Nuclear Safety and Security and
Technical Cooperation each have
their own country officers and the
Department of Safeguards has
at least two country officers for
each country, one for information
management and one for operations.
This is chaotic.
This can be addressed in a
programme–matrix structure,
albeit with difficulty. The goal
should be that, for each country, the
IAEA senior management would
be able to turn to a single country
officer who will be able to describe
10 IAEA, Board of Governors, ‘The agency’s
accounts for 2009’, Report of the Director
General, GOV/2010/20, 13 Apr. 2010.
SIPRI is an independent
international institute
dedicated to research into
conflict, armaments, arms
control and disarmament.
Established in 1966, SIPRI
provides data, analysis and
recommendations, based on
open sources, to policymakers,
researchers, media and the
interested public.
Göran Lennmarker, Chairman
Dr Dewi Fortuna Anwar
Dr Vladimir Baranovsky
Ambassador Lakhdar Brahimi
Jayantha Dhanapala
(Sri Lanka)
Ambassador Wolfgang
Ischinger (Germany)
Professor Mary Kaldor
(United Kingdom)
The Director
Professor Tilman Brück
all agency activities in that country.
This will be a huge cultural shift
for the agency. Such a person might
reside on the programme side of the
matrix and be called a project officer
(see figure 1).
In addition, the country officer
would control access to the agency
by outside organizations. This is not
a problem if outside organizations
do not coordinate their approaches
and funding, but the agency should
know what it is receiving and how
projects interact.
The IAEA acknowledges that it
would like to work as ‘one house’
and this has been a goal of senior
management for some years.
Structural obstacles have meant
that there has been little progress
towards this goal. A thoughtful
reorganization of the IAEA along
programme–matrix lines could go a
long way towards reaching the ‘one
house’ goal.
Re-creating the IAEA as ‘one
house’ requires intentional and
substantial management changes.
These changes must modify the
culture to the extent that technical
standards and challenges—not
historical loyalties and adherence to
tradition and seniority—define staff
If the IAEA can create an
environment where innovative
managers attract the best people
to their programmes, the poorer
managers will need to compete or
be left behind. Similarly, if the main
programme managers have to share
resources and compete for the best
brains among the staff, they will
have far more incentive to change
their programmes and offer new
choices and challenges.
For the IAEA to be chosen to
take on new mandates, such as
verifying an FMCT, it needs to
show willingness to adapt to new
circumstances and apply all its
resources efficiently to a problem.
The IAEA was not chosen to verify
the 1996 Comprehensive NuclearTest-Ban Treaty because it was too
inflexible and would have tried to
force new responsibilities into an
inappropriate existing structure.
In the case of the FMCT, under the
current line structure the agency
could be expected to create another
FMCT-focused department with
its own staff and resources. This is
exactly the opposite of what needs
to be done.
The failure to achieve ‘one house’
is not due to a lack of vision or
a failure to recognize the need;
it is caused by an obsolete and
stagnant management structure.
The IAEA needs to study other
management models for high-tech
organizations employing a variety
of well-educated technical staff. By
mimicking successful commercial
and government organizations,
it will also improve its ability to
compete externally.
Robert E. Kelley (United States) is a former IAEA director and deputy leader of the
Signalistgatan 9
SE-169 70 Solna, Sweden
Telephone: +46 8 655 97 00
Fax: +46 8 655 97 33
Email: [email protected]
Iraq Action Team. He is a veteran of over 35 years in the US Department of Energy
nuclear weapons complex, most recently at Los Alamos National Laboratory.
© SIPRI 2014
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF