I originally posted this as a comment on Kiwiblog, but it is worth elaboration. I am not so much interested as why  sensitive documents somehow managed to be dropped on a public street into the path of a journalist, which, if interesting, is inconsequential in the larger scheme of things. The real issue is the proposed, or at least potential merger of NZ intelligence agencies. From a democratic standpoint, I believe that centralising all intelligence-gathering and analysis in one agency is a recipe for disaster, or at least political manipulation. A core tenet of democracy is the decentralisation of power, evident in a system of checks and balances, particularly in its security component. I fear that NZ has lost sight of this tenet. In that light, here is my brief (excerpted)  thought on the matter of NZ intelligence agency mergers:

(With regard to the potential merger of the GCSB and NZSIS) I shall limit myself to pointing out two problems, one external and one internal to the intelligence agencies involved. Externally, the GCSB manages the Echelon stations in NZ and passes along foreign derived signals intelligence (SIGINT) to the SIS and Police where necessary, as well as monitor NZ signals traffic where required (this is a minor part of its operation). It is therefore more of a foreign-oriented intelligence collection agency rather than a NZ-oriented one. That spells potential conflicts of interest with larger intelligence liaison partners in the event that it is subsumed under or within the SIS. NZ intelligence requirements do not always run in concert with those of its larger partners, although it gains a measure of insurance and protection for providing its soil for the eavesdropping stations (another reason why NZ will never be invaded without a fight, since the stations are extremely valuable to the Echelon partners).

Internally, the SIS already has to handle external and domestic espionage and intelligence analysis along with counter-intelligence duties. This with a total complement of less than 200 people, a quarter of whom are clerical staff. That means that all of the human intelligence that gives NZ primary source or primary-derived information, plus the analysis of intelligence derived from the GSCB, NZDF, NZ Police, contract assets and liaison partners, has to be done by 150+/- people. It is a tall task already, and adding the SIGINT duties to it can complicate the management of intelligence flows and result in turf battles between the SIGINT and HUMINT branches and their respective analytic units (to say nothing of the fact that foreign nationals are heavily involved in the operation of the Echelon stations and therefore answer first to their foreign masters. Allowing them into the SIS could therefore compromise NZ national security even if they are erstwhile allies).

It is also generally believed that in a democracy it is best to separate domestic from foreign intelligence gathering, and SIGINT from HUMINT so as to avoid the monopolisation of intelligence flows and advice in any one agency, which could be politicised to deliver “intelligence” that is more politically-motivated spin than actual fact (as occurred with the Zaoui case under the previous SIS Director). Unified intelligence agencies can operate in democratic systems (such as in Canada), but that requires strong parliamentary oversight authority, something that does not exist in NZ.

The EAB is an intelligence client that undertakes foreign-oriented assessments rather than a collection agency, so a move to merge simplifies the intel streams coming its way. The same goes for the Police and the NZDF (which have their own collection branches), Treasury, other Ministries as well as the Combined Threat Assessment Group (CTAG). But one of the good points of having different sources of intelligence collection and analysis is that it avoids “group think” (and mistakes) by getting independent vetting of sources, methods and interpretation. Under the merger plan intelligence will be reduced but not completely centralised, although the question remains as to whether a merged agency can competently handle all of the responsibilities that entails.

All of which is to say that the merger idea may be economical but it may not be efficient.