Funferal

Census question

I’ve vague memories of heading this issue being raised back in 2001 or 2002, but can’t remember the detail of the discussion, so here goes:
The Irish census is generally held in years ending in a 1 or 6. From 1841 – the pre-Famine census that showed an all-island population of 8 million, or 6.5 million in the area now constituting the Irish state – to 1911 the census was held every 10 years. The 1921 census never took place – it’s difficult to do that sort of thing during a war of Independence, and the next census was 1926, with the 10 year interval continuing until 1946. From 1951 to 1971 the census was held every 5 years. The 1976 census was delayed (I believe because of concerns about cost) and a census was held in 1979, with the 1 and 6 pattern being re-established just two years later in 1981.
One would expect that the census after 1996 would be held in 2001, therefore. However, for some reason it wasn’t held until 2002. Now, again, we see a reversion to the usual pattern, with a 2006 census to take place this April.
OK, this is a somewhat arcane question, and perhaps trivial compared to issues of greater import. However, the historical patterns revealed over censuses cannot be properly tracked when we get quasi-random intervals between data points. Given that 2001 came at the end of the greatest financial boom in the history of the state, what sort of justification is there for not gathering the census data in 2001? The question doesn’t seem to be addressed on the CSO FAQ (not that I expected it to be) – any knowledge or theories on this?