Abstract
Recently ISI and IAOS launched a joint statement on El Salvador’s decision to dissolve DIGESTYC. ISI/IAOS expressed its grave concern about the risks involved. The risks particularly relate to the mission of El Salvador’s official statistics as well as their professional independence, the qualification of staff and the confidentiality of statistical data.1 Javier Caranza, director of GeoCensos (See box below), an NGO that develops, studies, and implements technology trends to support National Statistical Offices, has worked with DIGESTYC, the NSO in El Salvador. In this interview, held on 11 October 2022, Misha Belkindas invites Javier Caranza to give his view on the developments with official statistics in El Salvador.
The national decree that suppressed DIGESTYC took the statistical community by surprise especially because it did not involve the decision-relevant internal local or international stakeholders. GeoCensos shared from the start this preoccupation with ISI and IAOS and methodically analyzed as an affected stakeholder the undesired consequences that the abolishing decree brought to the businesses, academia, NGO community, and citizens of El Salvador. DIGESTYC inherits a history of producing statistics for the country starting in 1881 and its legal status was established almost 70 years ago.
As active stakeholders of the statistical ecosystem, GeoCensos staff has closely monitored the situation keeping up with the present and former officers of DIGESTYC. After three months of an ill-discussed legislative decision, most of the present information about the transition is speculative or informal, against recommendations of good governance. Most of the involved statisticians claim that from the beginning of the transitioning process a roadmap for the transference of functions was neither agreed upon nor planned with the appointed authorities from the Central Bank. Until now, there is no official confirmation about how former officers will be involved in the new statistical system, although normally well-informed sources estimate that in the best-case scenario one-third of personnel could be employed by the BCR, mostly those coming from a technical background. Also, as the transition delays its implementation, some statistical surveys are being undertaken by former field personnel of DIGESTYC. This unexpected situation provokes that, some of the individuals being surveyed are questioning the validity of the authority to request replies. The rationale of this attitude reduces to asking, “Why should I give answers to an abolished organization and how the information I provide will be used?”
All this chaotic context of a forced transition naturally produces an important gap in the regular stream of statistical operations for the country, now trusted to the Central Bank.
Abrupt changes in the statistical system may jeopardize several of the fundamental principles of official statistics advocated by the United Nations. In the case of El Salvador – among other shortcomings – laws, regulations and operational measures of the reformed statistical system have not been made available to the public. Many of the documents or procedures that are implicitly guiding the post-demolition plans have not been disclosed and are still unclear for all the involved actors.
The case of abolishing the NSO of El Salvador diverges in general from the natural tendency shown by the good governance of official statistics in most exemplary countries. In the last 10 years, the institutional responsibilities for the collection, processing and dissemination of official statistics have been increasingly entrusted to statistical offices with an autonomous functioning, oriented towards demographic and social statistics, among other measurements. Although Central Banks have always been relevant in the production of national statistics, their role in the field is much more closely linked to macroeconomic statistics producing indicators with data mostly collected by national statistical institutes. Their area of expertise is far removed from social and demographic statistics.
We understand that all governments, especially in democratic settings, are entitled to sovereign decisions regarding institutional arrangements within the public sector. As the decree that launched DIGESTYC was created out of a foundational law, it is understandable that the legislative assembly needed to subscribe to the abolishment of the decree. However, the timing and resources that were made available to decide the abolishment showed as insufficient. The parliamentary commission appointed to study this far-reaching governance change did not refer to any in-depth expert report by national or international experts in official statistics and did not even have the point of view of international agencies.
The launching of new governance with limited consensus from the users of statistics sets a disadvantageous precedent for the future cohesion of the statistical system. National statistical systems are not just mere infrastructures for unilateral decision-making. They are also living organizations that can produce public goods whether for timely informing government decision-makers but also to offer accountability to all actors of the economy and the whole array of decision-makers within a society. The data revolution that the UN’s 2030 Agenda envisions, highlights the need of leaving no one behind, including all those who benefited from public policies, creating new spaces for participation and activism.
We have worked in El Salvador through local communities for eight years now, establishing strong bonds with the government, private sector, academia and civil society, especially with groups related with open data and open-source developments.
Our first active commitment was to convene with local information technology associations, universities and nascent start-ups to join the Mesoamerican Apps Hackathon in 2013. In that event, databases and software development packages were used by participants to solve challenges that were formulated by relevant stakeholders in the data ecosystem.
GeoCensos has organized since then six unconference-format events in the country involving more than 20 base communities. One of our last assignments was to work in the preparatory stages of the next Census of El Salvador in 2019, together with DIGESTYC, civil society with the support of the Inter-American Development Bank. We learned by then that the statistical system has been struggling to function properly for decades already given many restrictions that are faced. The reasons that prevent the effective functioning of the system can be found in the lack of autonomy of DIGESTYC, scarce resources, technical skills and, especially a shady culture of data silos in the country.
El Salvador hasn’t run a census for already 15 years now. Before the decree was enacted, the parliament had approved the request of a 44 USD million loan to the Inter-American Development Bank for this purpose. The loan included a technological component and a timetable for the execution of the population and housing census and other census activities such as an agriculture and livestock census, and an economic census. Reference dates for the execution of the census were already available by the first semester of 2022. Currently, and until the end of the year, there is no formal date or clearly defined methodology for the development of these censuses. Incumbent sectors of the country have demanded for a long time a prompt update of census statistics for credible decision-making.
Nevertheless, the present government has been declaring a strong favoritism to embrace information technologies since the beginning of its mandate, with a decisive encouragement to involve what their plans mentioned as “the radical power of citizen creativity” to produce changes. This idea was embossed in the Cuscatlán plan, a manifesto of the to-be-elected government, proposed to voters to support radical changes to address mismanagement in the country before 2019. The promised plan formulated a set of subcomponents to support technical teams that would trigger previously unseen changes at the country’s national level and in some cases the local ones. Among other, the consolidation of an integrated national monitoring system was proposed, without mentioning the national statistical system. This plan called for the implementation of the so-called "Owl" system, which would combine big data with localization data to target relevant problems in a nowcasting fashion. In addition, the Cuscatlán Plan mentioned the need to achieve an open data government and interoperability strategy that “would integrate communications between government entities and private enterprise, (
Up to now, the government has not been able to keep up to these electoral promises, and the confusing signals to provide certainties to involved stakeholders don’t seem to follow a predictable path for the future of the census. Although following the elimination of DIGESTYC as a full-fledged statistical institution in the country, the authorities of the Central Bank have only offered diffuse signals as to what the concrete actions in preparation for the census will consist of. At present, there is only evidence that certain sectorial surveys have been announced, according to what is published on the still active web page of the former NSO.
I believe that deciding to modernize the national statistical system was correct. In fact, the approved development loan to undertake the next census included in the pre-demolishment stage had an important component for the modernization of the former NSO. However, reviewing performance figures and indicators to back the outreach of the involved decisions could have resulted in a less confrontative decision to enrich an inclusive dialogue, as data-driven decisions are strongly advised for a modern governance. Performance indicators, based on formal evidence are great ways to rationalize these kinds of transcendental decisions that affect accountability and credibility in countries.
The Tool for Assessing Statistical Capacity (TASC), adopted by the Inter-American Development Bank to support all regional NSOs in their self-assessments, has been administered since 2012 to the abolished NSO in El Salvador. Its results regarding strengths in data collection were not expressly considered in the final decision of the legislative commission. Surprisingly, the assessment that TASC offered through the last years revealed important advances in statistical production. This report has also highlighted some flaws in DIGESTYIC’s statistical performance, related mostly to the lack of technological capacities. TASC, along with other kinds of assessments, should be more seriously considered in carefully planned modernization processes for official agencies. Also, an independent evaluation could have nurtured a carefully administered discussion and hence favor a transparent decision-making. This has not been the case in El Salvador, at least based on the public discussion reflected by the media and the general considerations of the officialized decree.
Furthermore, according to the enacted law, the attributed mastery that the Salvadorian Central Bank has developed over the years to manage statistics is a matter still to be proven. The organization has certainly ingrained a role to produce, analyze and disseminate macroeconomic statistics but has no practical experience in producing demographic, socioeconomic or environmental statistics. This raises a crucially relevant question as well of how the upcoming demands for the calculation of sustainable development goals indicators will be fulfilled in the coming years.
The consequences that are expected by enabling a Central Bank - with no actual role in the real economy given the legal dollarization of the country- in the center of statistical governance are yet to come. Some actors have expressed stark preoccupations regarding how this case will affect attitudes toward the modernization of statistics in other countries, especially regarding technical soundness, accountability and transparency.
Dismantling official systems without alternative convincing solutions to fulfill their functions doesn’t make much sense. In an era where data is revolutionizing all domains of social and economic activities, it makes sense, nevertheless, that statistical systems integrate new visions and advances of data-related actors. Although revolutions are considered as the repudiation and the thorough replacement of an established system, in GeoCensos we believe that pushing forward a data revolution can also mean speeding up needed changes in the statistical domain to make available trustable data and reuse available statistics. This should be done by consolidating valuable dialogues but also actively discussing what could be done better mediated by constructive attitudes.
We believe in revolutionizing the data landscape with unedited partnerships, facilitating participative and inclusive commitments from organized NGOs with the official statistics processes and boost not only the availability of more and better statistics for a sustainable development, but also strengthening accountability and closer monitoring of public policies.
We have been part of the Global Partnership for Sustainable Development Data (GPSDD) since the beginning of its creation promoted by the UN Foundation. First, as data champions supporting many of the National SDG Indicators Roadmaps in Latin America and Africa. After that, actively participating in meetings as member of the GPSDD.
Our determination to provide new visions to the data revolution quest, took us to succeed in our outreach to the program committee of the World Data Forum, where we participated as formulators of sessions in the first two in South Africa and UAE, and to finally take part as members of the program committee for the virtual and in the third hybrid versions organized in Bern, Switzerland. Succeeding to find a voice in this global space should be inspiring for other NGOs to increase their influence in the global statistical ecosystem, whether with other international NGOs or partnering with more grass-rooted actors that feel the thirst of leaving no one behind for sustainable local development.
The space that the UN WDF is creating is beginning to draw the attention of governments that would like to take advantage of the benefits that data-driven policies bring to all economies, societies and ecologies.
However, considering that the organization of the Forum is strongly influenced by United Nations’ officers, its thematic areas are strongly affected by the perspective of their client members and local host countries. A certain bias influences the design of WDF, and consequently, priority is given to ideas and agendas that official organizations consider as “useable” or “politically correct” for new statistical developments. This approach is not facilitating a broader space to portray and seriously assess game-changing statistical practices that could build innovative data partnerships such as those that civil society can bring to the table through, for example, citizen-to-government data partnerships.2
We warmly applaud the increasing openness of IAOS to include in its publications not only public data providers, such as official statisticians but also other stakeholders in the statistical ecosystem in order to broaden perspectives, methodologies and practices in the field of official statistics for data provision. In terms of what we expect, this means broadening access to a whole new set of readers and reviewers, beyond those specialized in the field of data provided by NGOs.
The usefulness of this possibility of exchange is promising and tightly linked to the results of the peer-
reviewed scientific debate. Given the relatively low deepness of the debate among actors in the statistical ecosystem in the Global South, we would expect that one of the possible fruits would be the discussion of a data compact to co-create and co-produce data provision by agreeing on principles and standards, walking the path together with all sectors leaving no one behind, thus building up trust among the statistical community.
More changes in national statistical ecosystems are yet to come, not only due to the need of modernizing technical practices and governance of infrastructures but also provoked by a whole new variety of data producers. These changes could lead to a worrying divergence of perceptions of reality, but on the other hand, they could produce surprising synergies with governments that find themselves in the same situation as El Salvador in order to address bottlenecks in the production of statistics. We urge NSOs to be more open to consider alternative kinds of data. All those involved in the production of data and statistics, from the civil society to governments, should take more advantage of technical and institutional innovative data partnership arrangements in the spirit of collaboration of the data revolution. GeoCensos will very much support efforts to provide data for the wealth of official statistics and we are prepared to strengthen governance from the bottom up with citizen-to-government data partnerships.
Footnotes
ISI IAOS Joint statement by ISI and IAOS on El Salvador’s decision to dissolve DIGESTYC. !9 August 2022. See:
The terminology of Citizen-to-Government data partnerships (C2Gdp) refers to the partnerships of data between civil society communities with their governments. The C2Gdp concept refers to the long run venture of two different actors, civil society, and NSOs, that formally and voluntarily agree to undertake transformative actions to respond to the demands of monitoring sustainable development. See: