Federal Assembly of Switzerland
Swiss Federal Council
Federal Department of Foreign Affairs (FDFA)
Federal Department of Justice and Police
Task Force on Asset Recovery, FDFA
Swiss Agency for Development and Cooperation, FDFA
ON PRINCIPLES OF RESPONSIBLE ASSET REPATRIATION TO UZBEKISTAN
Proposal by Uzbek civil society activists
We are a group of Uzbek activists
citizens concerned about the future of our country. We write because we understand
the Swiss government may soon negotiate the return of several hundred million
Swiss francs that were stolen from the people of Uzbekistan through corruption in
the country’s telecom sector and hidden in Swiss banks. We urge the government
of Switzerland to ensure that these assets are returned responsibly to
Uzbekistan. By responsibly, we mean in ways that provide remedy to the citizens
of Uzbekistan as the true victims of grand corruption and that ensure concrete
measures to eliminate systematic corruption in Uzbekistan.
The
need for responsible repatriation of stolen assets
Repatriation of
stolen assets is responsible if it contributes to the fight against corruption
on global and national levels.
Grand corruption in
Uzbekistan is endemic and penetrates all levels of the political,
administrative, legal, corporate, and financial institutions. A clear
correlation is seen between corruption at the highest levels and the gross
violations of human rights.
Therefore, a cautious asset
return policy should be pursued. There is a high risk that without anti-corruption
mechanisms in place unconditional return of the proceeds of corruption will
very likely lead to funds being stolen once again and recycled anew into
western and offshore jurisdictions.
Unconditional return of the
proceeds of corruption would violate both the word and spirit of the UN
Convention against Corruption, international human rights law, and the global
framework for fighting corruption. UNCAC requires the return of stolen assets
to the countries of their origin. States however cannot selectively comply with
UNCAC provisions; they must comply with all provisions, including those requiring
the establishment of anti-corruption prevention mechanisms.
If asset-holding states fail
to ensure that the return of the proceeds of corruption serves as a deterrence
to future crimes, this must be considered irresponsible, if not complicit with
perpetuating corruption in the requesting state.
Proposed
principles of asset restitution to Uzbekistan
At the Global Forum on Asset
Recovery in December 2017, representatives of the Swiss and other governments
agreed to principles to guide asset return. Among other things, the governments
pledged that the return should guarantee transparency and accountability in the
use of the assets returned, benefit the citizens of the receiving country,
strengthen the receiving nation’s ability to fight corruption, and ensure civil
society participation in the return and use of funds.
Below we propose eight guidelines
that emphasize and elaborate on these principles in the context of any return
of stolen assets to Uzbekistan:
1. The asset return should be considered in the context of, and
contribute to, the fight against corruption and human rights violations on
global and domestic levels and meet standards of transparency and accountability.
2. Therefore, there should be no return of assets without sound
conditions and mechanisms to prevent the assets being stolen again.
3. The asset return should serve as remedy for the victims of
corruption, i.e., the entire population of Uzbekistan.
4. The best remedy for victims is not compensation – which
would neither be practical (with Uzbekistan’s specific institutional
environment) nor have any lasting long-term, systemic effect – but rather
measures to reduce the scale of corruption and human rights abuse in Uzbekistan.
5. Asset return should
be used as an incentive to reform those specific institutional conditions that
were and remain driving factors for the original crimes – bribery and extortion
in the telecom sector of Uzbekistan – and consequent money laundering. Below we
identify five such institutional conditions.
6. Reforms cannot be
implemented overnight, and will require several years. Therefore, the
restitution process should be extended over time and follow a step-by-step
approach, rewarding Uzbekistan for making benchmarked and measurable progress
in realizing reform.
7. Provided the
principles 1-6 are satisfied, the Government of Uzbekistan should have control
and ultimate approval over the disposition of the assets, following a
transparent process inclusive of consultations with relevant international
stakeholders and Uzbek civil society. The implementation of these principles would
create a completely new reality in Uzbekistan in terms of the governance system
and, as such, give sufficient assurances that the assets are not re-stolen and
re-laundered.
8. In case the government of Uzbekistan rejects these
guidelines, the Swiss government should keep the assets until the Uzbek
leadership accepts the deal and has shown legitimate and verifiable progress.
Five
institutional sectors to be reformed as prerequisite for asset return
1) Establish a transparent and competitive tendering process,
including in the spheres of allocation of frequencies and licenses, the public
procurement of goods, public works, and contracted services. Publish all material
related to public procurement and concession contracts in the public domain.
2) Reform the public administration to make sure civil servants
carry out the duties of their role, abide by the law, and work in the public
interest instead of acting out of personal loyalty to high-ranking officials.
With this in mind, implement a conflict of interest regulation for all civil
servants and government officials in Uzbekistan and provide full transparency
of their income.
3) Establish independence of judiciary and legal profession;
enforce the rights for fair trail and due process;
4) Ensure transparency of public and corporate finance,
including: obligatory auditing of state and its major corporate contractors by
reputable international auditors and the full publication of such audit reports
in the public domain; disclose government books to the public, with an emphasis
on export revenues; and administer a transparent system of corporate
disclosure, with a requirement that all corporate entities operating in Uzbekistan
participate in a public register of beneficial owners.
5) Provide oversight including through the establishment of an independent
public anti-corruption agency; and allow civil society and media outlets to freely
operate and conduct journalist investigations without fear of harassment and
repressions.
We note there
are many widely recognized international indicators available that would
measure governments’ progress in meeting these objectives (see annexed below discussion
of Indicators for Gaging the Progress of
Reform in Uzbekistan, prepared by members of the Working Group on Responsible
Asset Repatriation).
Why specifically
these five institutional conditions?
The complete absence of
institutional integrity in the five sectors mentioned above were key enabling
factors for the original corruption case. As such, these areas should be
addressed as prerequisites for returning the proceeds of corruption to the
Government of Uzbekistan. But what is
right for the telecom industry can also be applied to other sectors of
economy.
First, Uzbekistan did not at
that time and still does not have an established system for transparent and
competitive tendering processes, neither in allocating frequencies and licenses,
nor in public procurement. Given the absence of transparent public procurement,
it is very likely that unless appropriate reforms are implemented, the returned
assets will be redistributed via opaque lucrative procurement contracts, a
frequent practice if not the rule for use of public funds in Uzbekistan.
Second, Gulnara Karimova did
not act alone in extorting bribes from telecom companies in exchange for
issuance of operating licenses. Karimova would have been unable to carry out
the original crimes without involvement of a number of government institutions.
A number of high ranking officials were directly involved in issuing these
licenses, and these individuals acted out of personal loyalty to herself and
her father, former president Islam Karimov. This highlights the systemic impact
of the patronage system upon how public office was and still is run, and the
total absence of conflict of interest regulation in Uzbekistan. It also
demonstrates the lack of a civil service which works in the public interest and
can withstand pressure from high ranking corrupt officials.
Third, it is notable that
from the very beginning, western telecom companies sought back-door deals with
local officials, instead of relying on judicial measures and legal means to engage
in and protect their business and property rights. We understand western
companies were well aware of the absence of rule of law and independent
judiciary in the country.
Fourth, the public and
corporate finance systems in Uzbekistan are extremely opaque. Neither the state
itself nor its major corporate contractors are under obligation to open their
finances to scrutiny by reputable auditors. Additionally, the state budget does
not reflect income from export revenues the government controls, such as
revenues from the cotton sector – linked to forced labor and human rights
violations. Instead, these revenues go into hidden extra-budgetary accounts
controlled by a few individuals around the President, and are used at the
discretion of political elites on expenditures unknown to the public. It is
unfortunately very likely that an unconditional return of assets to Uzbekistan
will leave the state susceptible to repeating such practices; with the funds remaining
absent from the state budget.
Fifth, essential monitoring,
oversight and compliance institutions are absent. Uzbekistan does not have an
independent anti-corruption state agency capable of independently detecting and
preventing cases of government corruption. Uzbekistan also lacks an independent
organized civil society – following a large-scale crackdown between 2004and2007
– or an independent press that would be able, without fear of repression or
reprisal, monitor, document and report on corrupt practices, especially on
cases of grand corruption. Activists and journalists who tried in the past to
report on such cases were detained, tortured and imprisoned. While some of
these journalists and activists have been recently released from prison, which
constitutes a positive sign, they are still subject to reporting requirements,
harassment, and other conditions inconsistent with full rehabilitation by the
state. Nor have they been compensated by the state for their unjust
imprisonment. In fact, legally, these individuals are still considered
criminals according to the judicial system of Uzbekistan, and the original
miscarriage of justice has not been addressed.
Our demand for serious
changes in these five areas as part of the process for returning assets to
Uzbekistan should not be seen as confrontational towards the government. The
Uzbek political leadership has itself committed to reforms in these areas and
has even made a few steps forward. For instance, a draft law on public
administration has already been published, which still needs to be adopted and,
more importantly, implemented. There has been also one case of fair trial open
to the public (in March-April 2018, trial of Bobomurod Abdullaev and two
others). So far, this is a single such precedent. Besides, there are some indications
of opening for Uzbek press to report on controversial issues, though the press does
not still dare to conduct journalist investigations into corruption cases. These
positive signs do not yet represent fundamental or irreversible change. What is
needed in Uzbekistan is not promise and rhetoric of change, but a sustained
pathway to reform and the adoption of standards of good governance.
It still remains to be seen
whether the government will adopt a full package of anti-corruption reforms
and, more importantly, implement new legislation into actual practice. We
believe that each significant step and improvement in these five areas should be
rewarded, with a corresponding portion of assets returned to the government.
Such a sequenced, benchmarked approach to restitution would be perfectly in
line with the course of reform to which the new Uzbek political leadership has
publicly committed, notably within Uzbekistan’s Development Strategy
(2017-2021). The return of assets should reinforce the process of reforms,
offering a guarantee that they will become sustainable and, over time,
irreversible.
Why not a
Bota Fund-like solution?
We believe that the creation of the Bota
Foundation in 2009 as a channel for restituting Kazakhgate money to the people
of Kazakhstan was the right decision. The latest restitution by Switzerland to
Kazakhstan of $48 million, in 2012, was, by contrast, much worse, from the
point of view of principles of responsible return.
An investigation was conducted into USD 21.76m
of the restituted assets earmarked for a Youth Corps Program. It revealed that
the Project Coordinator for the Youth Corps Program is a consortium made up of
Government Organized NGOs (GONGOs). The consortium is headed by the President’s
eldest daughter, Dariga
Nazarbayeva, who is a politician in her own right. This
GONGO consortium will receive approximately USD 465,000 for its services as
Project Coordinator. “From the limited available procurement data, - write the
authors of the summary report, - examples have been uncovered of lavish
spending on promotional material, awards to the ruling party’s youth wing,
expenditure on materials that have a propagandistic function, and the selection
of ‘host organizations’ that are predominantly GONGOs, some of which are run
directly by public officials and politicians espousing strong commitment to the
President’s national ideology.” [1] This resulted in restitution
benefiting the ruling party and senior officials of Kazakhstan, but not the
victims of corruption.
Under the
current state of affairs in Uzbekistan, the Bota Fund-like solution cannot be
applied without violating the principles of transparency and accountability and
responsible return, because the conditions for operation of such a fund,
independent from the government, do not exist.
First, the Bota
Fund in Kazakhstan operated through allocation of grants to local NGOs which, in
turn, delivered financial and technical assistance to low-income families. In
Uzbekistan, such independent NGOs simply do not exist. Between 2004 and 2007,
the relatively independent NGOs were suppressed, leaving Uzbekistan with no independent
organized civil society. Due to repression inflicted by the Karimov regime,
there are currently no stakeholders in Uzbekistan who could either provide Bota-like
services to the population or conduct independent monitoring of how these funds
are being used.
Second, in the current
absence of a transparent tendering and public procurement process in Uzbekistan,
there is a very high probability that the money will be channeled through
lucrative contracts to cronies of political elites, and thus stolen once more.
Examples of such shadowy deals on the allocation of contracts on construction
and reconstruction in Tashkent are multiple nowadays.
Third, without a
transparent system of public finance, the government can manipulate the use of
the returned assets. For example, it could cut funding it had planned to
provide for health, education, and other social projects, divert these funds to
private interests, and employ the returned assets to cover the gap in public
funds. By shifting funds this way, the intended restitution would bear no
positive effect for the people of Uzbekistan.
Finally, we
believe that using the assets to reinforce anti-corruption reforms will be most
beneficial for the victims of corruption and the development of Uzbekistan over
the long term, supporting systemic, transparent institutional reforms with
lasting effect.
Umida Niyazova,
on behalf of Uzbek-German Forum for Human Rights, Berlin, Germany
Nadejda Atayeva,
on behalf of Association for Human Rights in Central Asia, Le Mans, France
Jodgor Obid,
Uzbek political refugee, resident of Austria
Dilya Erkinzoda,
Uzbek political refugee, resident of Sweden
Ulughbek
Haydarov, journalist, former political prisoner, resident of Canada
Alisher
Taksanov, journalist, political emigrant, resident of Switzerland
Alisher Abidov,
political emigrant, resident of Norway
A number of
local Uzbek activists support these proposals, but their names are not
disclosed out of concerns over their safety.
Appendix
International Working Group on Responsible
Asset Repatriation
INDICATORS FOR GAGING THE PROGRESS
OF ANTI-CORRUPTION REFORM IN UZBEKISTAN
The Swiss and
Uzbek governments are negotiating a return of Gulnara Karimova’s assets, stolen
from the people of Uzbekistan and hidden in Swiss banks. Uzbek civil society
activists are urging the agreement to include reforms of Uzbek institutions and
ensure the assets are not again stolen, but are instead used to promote the
welfare of Uzbek citizens. The reforms
they seek are similar to those the United Nations, World Bank, and other donors
include when providing financial assistance to governments.
The
international community has developed a range of indicators to measure the
progress a government is making in realizing such reforms. The reforms Uzbek civil society collectively
urge, along with some of the key indicators used to mark progress in achieving
each, are listed below. We urge the Swiss and Uzbek governments to incorporate
these benchmarks into the asset return agreement. They should also agree on ambitious and realistic
targets for improving Uzbekistan’s score on achieving genuine reforms, as funds
are returned.
1)
Establishing a transparent and competitive tendering process.
The World Bank’s
annual Benchmarking Public Procurement reports how close national procurement
systems come to agreed upon standards of openness and competitiveness. Georgia’s Institute for the Development of
Freedom of Information has, with other organizations in Eurasia and Eastern
Europe, devised a method for measuring how open and competitive procurement
systems in the region are.
2)
Creating a capable, independent, apolitical civil service.
Published
annually, the World Bank Country Performance and Institutional Assessment rates
how closely countries adhere to the principles of a merit-based, politically
neutral civil service. Professors Peter Evans and James Rauch’s method for
measuring the quality of a nation’s civil service is widely employed by
scholars and policymakers alike to evaluate public service bureaucracies.
3)
Establishing an independent judiciary dedicated to enforcing fair trial rights
and observing due process of law.
Gothenburg
University’s V-Dem project rates countries annually on the extent to which the
executive respects the constitution and complies with court rulings; and extent
to which the judiciary can act in an independent fashion. The World Justice
Project and the Bertelsmann Transformation Index both publish a yearly
assessment of how closely governments adhere to rule of law principles.
4)
Ensuring a transparent public financial management system and transparency in
corporate ownership and finance.
A partnership
that includes the Swiss government, the European Union, and the IMF have
developed the Public Expenditure and Financial Accountability Framework (PEFA),
a now universally accepted system for assessing the transparency and
effectiveness of public financial management systems. The Open Budget Survey assesses the public
availability of budget information and other budgeting practices that
contribute to an accountable and responsive public finance system. The
Financial Secrecy Index administered by the Tax Justice Network provides
internationally recognized benchmarks for corporate and financial transparency
and compliance.
5)
Strengthening accountability through establishing an independent
anti-corruption agency and guaranteeing civil society and the media freedom to
operate.
The World Bank
gathers information from a variety of sources each year which it then compiles
into a scale showing “voice and accountability,” the extent to which a
country’s citizens can participate in selecting their government, how free they
are to express their opinions and create private, voluntary associations, and
whether the media faces restrictions on what it can publish.