Scud variants (Al Hussein and Al Abbas). We have been told that Iraq later declared to the United Nations that it had produced 17 Al Abbas and 387 Al Hussein missiles between 1987 and 1990. Thus, if the Iraqi figures are taken at face value, while the JIC paper was approximately correct in its estimate of the overall number of about 400 Scud missile variants produced by Iraq, it was inaccurate in the ratio of production between Al Abbas and Al Hussein. A possible explanatory factor is that the JIC’s performance estimate for one of the two versions of the Al Abbas missile was not greatly different from that of the Al Hussein. The episode illuminates, however, the complexities surrounding estimates of Iraqi ballistic missile stocks, against which later JIC estimates should be considered.
By contrast to pre-war assessments, JIC assessments prepared in April and May 1991 on the residual Iraqi ballistic missile stockpile did declare explicitly that they had been prepared on a worst case basis and in the absence of any direct intelligence. On the basis of somewhat fewer than 100 Iraqi missile firings during the war, the JIC concluded that Iraq:
. . . may have up to 600 left (but probably less),both standard Scud and exten ded- range variants.
[JIC, 17 April 1991]
On that basis, the JIC said that:
We cannot be precise,but we are confident that the Iraqis have substantiall y under- reported the numbers of missiles.
[JIC, 9 May 1991]
As in the chemical and biological weapons fields, we detect a risk here that, by making comparisons with worst case assessments (especially those not declared as such), analysts and policy-makers may have come to conclude that there were shortfalls in Iraqi declarations, with suspicions being exacerbated by Iraqi prevarication, concealment and deception.
A further JIC assessment in January 1992 described Iraqi declarations and included a substantial downwards revision in its estimates of Iraq’s ballistic missile stockpile. The JIC reported that:
Although we do not know the true figure,we assess that around 100 Scud B remai n concealed.
[JIC, 16 January 1992]
The JIC did not show fully the basis on which it derived that calculation. It has not therefore been possible for us to investigate whether the assumptions that underpinned it might have had an impact on assessments in later years about whether Iraq was concealing ballistic missiles and, if so, how many.
The JIC also noted in the same assessment that there might be: . . . as many as 250 complete Soviet built SCUD B guidance and engine packages which cannot be accounted for,and would be critical for future production . Provided the raw material was available,Iraq could build its own replacement mid-b ody sections and assemble new missiles from this stockpile. [JIC, 16 January 1992]