One addition to avoid inconsistencies: when the tied/untied marker is specified on the transaction level, it should NOT be specified on the activity level . The other way around: when the tied/untied marker isspecified on the activity level: it should not be specified on the transaction level. This should in my opinion be part of the guidelines.

The tied/untied merker on activity level should in my opnion NOT be depreciated, since very often individual transactions are not earmarked to tied/untied, the activity as a whole is though.

Thanks Herman,

I agree and will add your proposed guidance into the proposal.

There is no plan to remove or deprecate the activity-level element.

Bill

Having raised this issue repeatedly, we are disappointed with this outcome.

We do not agree with the Secretariat's conclusion that the existing standard is adequate. In early 2012, the OECD DAC Working Party on Development Finance Statistics (WPSTAT) and the International Aid Transparency Initiative (IATI) developed a framework for the common standard, which was endorsed by the Working Party for Aid Effectiveness (WP-EFF) in June 2012. This framework sets out a clear process for the governance of data fields and associated definitions to deal with overlap, specifically:

"Where CRS and IATI overlap (e.g. same data fields) IATI has and will continue to use CRS definitions and classifications. Should there be a need to amend/change/add/remove a code methodology, definition, structure or format in the CRS/FSS domain, such action shall be in the form of a proposal to be considered for ratification, as per current procedures/practices, by the WP-STAT. IATI shall not create codes, standards, definitions, structures or formats that will create a divergence for said overlap"

Therefore, according to the common standard agreement, IATI must comply with the standard set out by the OECD. While we can appreciate that the IATI standard can potentially provide more details on the tying status at the transaction level, the proposal of splitting the commitment across two (or three) transactions, each with the appropriate tied-status code is not feasible for most donors, given that their systems are not designed to accommodate this kind of coding and reporting.

We strongly encourage the IATI community to ensure the Busan common standard agreement is respected. For the tying status, this means that tied, untied and partially tied commitments are recorded as separate amounts in separate columns.

There are other parts of the standard where IATI provides a different, yet logically consistent, representation of  CRS++ guidelines. Policy Markers are a case in point.

Our proposal has been made in the same spirit: how can we accurately reflect CRS requirements with an activity-based, transaction-based framework.

 I would like to suggest three options and would ask all DAC reporters to provide feedback on this.

1. Separate commitment transactions for different tied statuses as currently proposed.
2. A new <tied-aid> element at activity level with three attributes mimicking CRS++ columns 36-38: "amount-untied", "amount-partially-untied" and "amount-tied". These values would be reported in addition to current requirements. This would also involve the removal of the <tied-status> element in the current standard as having both could lead to confusion with different publishers adopting different approaches.
3. A similar element at transaction level

While I personally think our proposed solution is ideal as it attempts to integrate CRS requirements into an IATI framework, I have no strong objection to Option 2. Option 3 I think would be confusing.

Yohanna is Option 2 acceptable?

What do others think?

In my opinion what is important is that information content can be represented adequately. The current IATI standard enables publishers to report the tied status of activities in an identical manner as reporting to the OECD DAC in CRS++: the meaning (definition) is exactly the same, so the same facts can be represented in both IATI and CRS++ format.  The current IATI format fullfils all information needs and seems to be adequate. So no changes are nessesary.

But maybe I am missing something here. If the current IATI standard is not enough to capture the CRS++ information content, please explain the problem from the information perspective. In my opinion the format is less relevant: IATI is a different format than CRS++. It would be a mistake in my opinion to try to mimick formats. Having compliance on the information level and not on the format level should be the goal of the common standard.

There was another point I meant to make ...

Most users of IATI data analyse transactions to make sense of the financials. If information on tied statuses is important is it not best for it to be embedded into transactions, rather than sitting as a detached piece of data that is less likely to be used?

I will answer to your comments on behalf of Yohanna as we have been working together on IATI implementation and I am the statistical correspondent for Canada at the OECD-DAC.

Our concern has more to do with the additional expectations of the standard measuring the tied status at a transaction level, when reality this is very difficult to measure in current donors' systems. As Herman put it in his earlier intervention: “very often individual transactions are not earmarked to tied/untied, the activity as a whole is though”.   Furthermore, this is coded at the commitment stage when activities are approved, and this is why the OECD-DAC measurement and stats on untying are done at the commitment level.

Therefore, Option 2 is the best and most acceptable, but it would still have to be attached to the commitments somehow.  Thanks,

Regarding the current intended usage of tied-status at transaction level:

• The intention was never to expect all transactions to have a tied status, only Commitments where an activity is partially tied.
• If an activity as a whole is Tied or Untied then the Default-Tied-Status code at activity level suffices.
• If an activity is Partially Tied two transactions, both commitments, can itemise the Tied and Untied amounts.

This is what the current version does. There is no additional expectation beyond what the CRS requires, only a different layout.

Is this explanation sufficient?

Ok, thanks for the explanation. The explanantion will be part of the guidelines and the data quality checks in the IATI dashboard?

Yes

Thanks for all the comments and the clarification that only the commitment is the expectation. It wasn't very clear to me at first, and agree with Herman's suggestion that this should be in the guidelines.

It is proposed that the following guidance is added to guidance here - http://iatistandard.org/activity-standard/overview/classifications/

• If an activity is either wholly "tied" or "untied" it is recommended that the appropriate default-tied-status code is used.
• If an activity is "partially tied" it is recommended that the reported commitment/s are split into "tied" and "untied" amounts and tied-status is reported at transaction level. (NB that tied status should be reported for commitments only.)

This change might be misunderstood since “3 - partially tied” has a different meaning than just a mixture of “5 - untied” and “4 - tied”. An activity is either wholly “tied”, “untied” or “partially untied”, or a mixture of these three.

In the CRS it is possible to split the sum reported in “Commitment” between three columns, therefore I wouldn’t say the IATI standard with this change would be more detailed than the CRS.

While discussing “Partially tied” I would like to suggest a change of the definition (in the DAC directives it’s called “Partially untied”, but that´s not a big issue). What's important here is that essentially all partner countries are included, and that it is not enough with the donor country, which the “or” in the definition might lead the reader to think.

Current definition:

“Official Development Assistance for which the associated goods and services must be procured in the donor country or among a restricted group of other countries, which must however include substantially all aid recipient countries”

Proposal:

“Official Development Assistance for which the associated goods and services must be procured from a restricted number of countries which must however include substantially all aid recipient countries and can include the donor country.”

Definition in the DAC directives:

“Partially untied contributions

148. Partially untied aid is defined as loans and grants which are tied, contractually or in effect, to procurement of goods and services from a restricted number of countries which must include substantially all developing countries and can include the donor country.”

What do you think?

IATI should follow the DAC on this. We will make the changes as you suggest. Thanks

The proposed change to guidance modified on 28 August (above) should be further modified as follows.

It is proposed that the following guidance is added to guidance here - http://iatistandard.org/activity-standard/overview/classifications/

• If an activity is either wholly "tied", "partially tied" or "untied" it is recommended that the appropriate default-tied-status code is used.
• If an activity more than one tied status it is recommended that the reported commitment/s are split into the relevant tied, untied and/or partially-tied amounts and tied-status is reported at transaction level. (NB that tied status should be reported for commitments only.)

See: https://github.com/IATI/IATI-Guidance/issues/139 for implementation 

Great! Would it be possible to also change the first sentence to include "partially tied"?

If an activity is either wholly "tied", “partially tied” or "untied" it is recommended that the appropriate default-tied-status code is used.

Oops, yes