Lopes et al. (2017), correctly, and helpfully, draw attention to the fact that the CITES source code ‘I’ should represent legal (re)exports of seized wild animals and their derivatives using a CITES permit, rather than seizures relating to illegal trade activity, as we had used it. They correctly state that we followed an earlier version of the official CITES guidelines [Version 8 (UNEP-WCMC 2014)] to interpret the CITES Trade Database. This was, of course, our only option insofar as it was only several months after our paper was published that the latest clarifying guidelines for the preparation and submission of CITES annual reports became publically available (via CITES Notification 2017/006), followed one month later by a new guide to the application of CITES source codes (UNEP-WCMC 2017). Together, these documents added new clarity, revealing ambiguity in the guidelines at the time we wrote and published our paper. In passing, we observe sadly that instances of such misleading failures of clarity detract seriously from the usefulness of the CITES database, and the one exposed here is not the only such case [one of us has recently noted unhelpful ambiguity in the figures relating to legal lion trophy exports, and recommended a reform of this (Macdonald 2016)].

Like Lopes et al. (2017), we think it helpful to excavate further the intricacies of the CITES database and its interpretation. As they correctly state, the CITES trade database should only contain legal trade data, reported via granted permits and certificates (as specified in Article VIII of the Convention, paragraph 6 and 7). However, it has become clear that this is not always the case. For example, Parties are sometimes also required to report seizures involving illegal wildlife trade [e.g. Pangolins (Pholiodota) as requested through Decisions 16.41 and 16.42 in 2013 (CITES 2013)]. Indeed, according to Heinrich et al., (2016) [also published subsequent to D’Cruze and Macdonald (2016)] the USA is one of the few countries reporting such trafficking incidents, and we understand that it does this by reporting the seizures under source code ‘I’ in the source column [their study reports 98% of pangolin records using source code ‘I’ made between 1975 and 2014 proved to be seizures rather than re(exports)].

Acknowledging that source code ‘I’ has been, and is mostly likely still being, misinterpreted and used by different Parties in different ways, further shakes confidence in the CITES Trade Database that has already been subject to criticism due to claims of inconsistent and incomplete reporting by Parties (e.g. Harrington et al. 2014; UNEP-WCMC 2014; Heinrich et al. 2016). These misgivings, and the explicit acknowledgement that data assigned to Code ‘I’ have been and remain a mixture (of unknown proportions) of both illegal seizure and legal re(export) incidents, seem to strengthen the conclusions and recommendations we offered in D’Cruze and Macdonald (2016) and we repeat our gratitude to Lopes et al. (2017), for affording us the opportunity to point this out.

In light of the clarity provided via CITES Notification 2017/006, and given the increased ambiguity regarding what type of incidents have been assigned to CITES source code ‘I’, we conclude that currently it remains impossible, using the CITES Trade Database alone, to establish accurately how many seized wild animals have re-entered commercial trade. Rather, as exemplified by Heinrich et al., (2016) in the case of pangolins, we argue that to establish this number it is necessary to compare these records with information in other databases that contain data regarding illegal wildlife seizures [e.g. the Law Enforcement Management Information System (LEMIS) database maintained in USA]. Similarly, as highlighted by Lopes et al. (2017), it is now clear that the true role of the importer and exporter is also impossible to determine without similar comparisons to other existing databases. Transparency is further obscured insofar as these additional data may be hard to access.

Standing by our original conclusion, and further strengthened by ambiguity in interpreting the CITES data, the point remains that national enforcement agencies have had to detect and quickly deal with illegal live shipments involving a diverse array of vertebrate species, and (as reaffirmed by Lopes et al. 2017) that numbers reported are likely the ‘tip of a far bigger iceberg’. It also remains true that CITES Trade Database records are inconsistent and incomplete. Indeed, data regarding the disposal of confiscated live animals is unavailable as providing them is not currently a formal CITES requirement. We repeat our conclusion that this lack of information impedes the proper allocation of resources and prevents the effective monitoring and evaluation of management outcomes. We add now, although we had imagined that it was obvious, that this is detrimental for both conservation and animal welfare.

We welcome the news that the first CITES annual illegal trade report will be made public in October 2017. Lopes et al. (2017) are surely right that this document will provide valuable information that will aid efforts to combat illegal wildlife trade. However, this diminishes neither the veracity of our main conclusions, nor the relevance of recommendations provided in D’Cruze and Macdonald (2016) which we presented formally at the 17th Conference of the Parties during a side event opened by the CITES Secretary General in September 2016. We are pleased to consider that our work may have helped to prompt, soon after, the adoption of Resolution Conf. 17.8 on Disposal of illegally traded and confiscated specimens of CITES-listed species (CITES Notification 2017/045). This led in turn, again pleasingly, to the development of a questionnaire intended to review existing guidelines and evaluate current practice in the disposal of confiscated live wild animals.

Both of these policy-focused developments represent significant leaps forward, but alone they will be insufficient to ensure the accuracy of reporting required to monitor illegal wildlife trade or to guide the allocation of available resources to address it. The publication of D’Cruze and Macdonald (2016) and subsequent documents thereafter (CITES Notification 2017/006; Heinrich et al. 2016; UNEP-WCMC 2017) have also contributed, we think usefully, to the accumulating evidence of opportunities to improve the already unquestionably valuable CITES data and to reduce confusion amongst Parties and researchers regarding reporting requirements.

However, given past evidence of inconsistent and incomplete reporting by Parties, and indeed the comments made by Lopes et al. (2017), future records using source code ‘I’ should be subject to scrutiny before judgement can be reached on whether such initiatives have been successful in preventing further misinterpretation. Meanwhile, none of the foregoing detracts from our general assertion that a very large number of Threatened wild animals is confiscated each year, that they must inevitably place a heavy burden on confiscating authorities, that they constitute a detriment to both conservation and welfare, with many of their fates remaining unknown and unrecorded. While it is appropriate that specialists (like Lopes et al. 2017, and ourselves) delve into the minutiae of these figures, our shared quest for precision should not obscure the starkness of this situation.