Page 433 - Invited Paper Session (IPS) - Volume 1

IPS175 Pietro Gennari et al.
agency in one year, the custodian agency should validate new estimates every
year even though they may have been produced using the exact same
methodology and data source. Once again, one is easily left wondering on
what grounds the validation could be denied one year after it has been
granted in the past, all else being equal.
The fourth reason that this provision is problematic is the end result that it
foresees, i.e. the nonpublication of data in case a mutually acceptable solution
has not been found and a country still disagrees with the value. Effectively, this
represents a severe limitation on the autonomy and independence of
international organizations, which have been assigned the function, by
member countries, to collect and disseminate statistics in their relevant
domains. Disallowing international organizations to publish their estimates,
especially when a country has none of its own, also blindfolds the international
community, which in such cases may not have any other instrument to
objectively assess a country’s situation, or to compare national figures that
may have been influenced by nonstatistical considerations. Without
independent international statistics, efforts to improve transparency and
accountability worldwide are undermined. It is for these reasons that the CCSA
strongly recommended that in case a mutually acceptable solution were not
found, an international organization could still publish these country estimates
under certain conditions (together with the national estimate, when available).
This recommendation, though, was eventually rejected by the IAEG-SDG.
This outcome becomes even more controversial as in many countries the
independence of the NSO may be vulnerable and its capacity to produce
objective estimates of SDG indicators may be challenged, especially when
these indicators are politically sensitive. There are known cases where the NSO
was not even consulted in the preparation of the national SDG indicator
framework. Even where the NSO enjoys a higher degree of autonomy, it still
struggles to fulfil the role of “coordinator of the national statistical system”.
This role may not be formally recognized in national legislation or it may be
difficult to implement in practice. Communication between the NSO and other
data producers may be an issue and there may not be appropriate
mechanisms of coordination and data transmission across data producers.
NSOs are often not aware of already existing data flows to custodian agencies,
especially for non-statistical indicators. More importantly, NSOs may lack the
capacity and experience to certify the quality of datasets produced by other
10
institutions (public or private) . An often-cited reason for not validating
estimates is that “we [the NSO] are not in a position to validate these
estimates”. There is therefore a need to develop systematic data transmission

10 MacFeely, S. (2018). Op. Cit.
422 | I S I W S C 2 0 1 9