Having had a chance to digest the India safeguards agreement, I’ve become interested in what is left out. Specifically, how will the IAEA analyse and draw conclusions about Indian compliance?

How the Agency analyses information is, in many ways, just as important as the safeguards agreement itself. Moreover, it has changed considerably since full-scope safeguards were introduced in the early 1970s.

Prior to efforts to strengthen safeguards in the 1990s the Agency adopted what was termed a “facility-level approach”. It had a set of inspection goals for each facility in a state. If those objectives were met at each and every facility then, at the end of the year, the IAEA concluded that there had been no diversion of nuclear material in the state concerned.

Following the shock of discovering Iraq’s clandestine nuclear programme in 1991, the Agency pioneered a new “state-level approach”. The IAEA safeguards glossary defines it thus:

a safeguards approach… developed for a specific State, encompassing all nuclear material, nuclear installations and nuclear fuel cycle related activities in that State. A State level safeguards approach defines the application of safeguards measures at each facility and location outside facilities in the State and, where an additional protocol is in force, the safeguards measures … that would enable the IAEA to draw and maintain a conclusion of the absence of undeclared nuclear material and activities in that State….

(For a less terse discussion, but one still written in Agency-speak read this).

The Agency never gives examples so let me give one of my own which (I hope) illustrates the idea.

Say a state has a declared enrichment plant capable of producing 100 tonnes of LEU per year. However, the state declares it only uses the plant to produce 60 tonnes of LEU per year to feed two reactors. This might well be because the state built too much enrichment capacity and doesn’t need to use it all. However, it also raises the possibility that the enrichment plant is being used for some undeclared (“excess”) production.

Under the state-level approach the Agency will flag up this potential mismatch and will look into it before drawing any conclusion about the state’s compliance. Under the facility-level approach, it would have “blindly” applied safeguards at the enrichment plant and reactors without looking into the difference in capacities.

The state-level approach is, incidentally, applied in all states with a safeguards agreements in force, whether or not they also have an additional protocol in operation (although the Agency’s conclusions is modified accordingly).

Anyway, this all raises the question of whether the Agency will be applying the state-level approach in India.

Obviously it can’t apply a state-level approach per se since India’s military facilities will remain unsafeguarded. However, if India cleanly splits it civilian and military facilities, I can see no reason why the Agency can’t apply a state-level approach to the whole of India’s civilian fuel cycle: a state within a state, if you will. Using this kind of analysis in India will significantly enhance the credibility of conclusions drawn by the Agency.

I don’t think this question has ever come up before with the IAEA as it has never attempted to safeguard an entire fuel cycle in a state outside the NPT (safeguards are applied in such states but only to particular facilities).

However, Euratom must face a similar problem in applying safeguards in the UK and France. Like India, the UK and France have a military fuel-cycle but their civilian fuel cycles are comprehensively safeguarded (by Euratom). Does Euratom adopt a state-level approach in the UK and France? I don’t know, but I’d like to find out.

In any event, I think that the India safeguards agreement does give the IAEA the necessary authority to use the state-level approach. Specifically, article 11 specifies which facilities, material and equipment require safeguards. Its authority appears to be sufficiently broad to enable the Agency to ‘track’ all nuclear material around the fuel cycle, wherever that leads.

I don’t want to give the impression that the omission of the Agency’s safeguards “concept” for India from the safeguards agreement is a deficiency—safeguards agreements never specify it. But, I do think we need to know what this concept is before we can properly assess the value of safeguards in India.