Nigeria Content Online

Which is the more
remarkable? The fact that in its statistical news for July 2010, the
National Bureau of Statistics (NBS) describes the revision of the basis
for computing the consumer price index (CPI) or the fact that on the
strength of this restatement, previous estimates of domestic price
movement may have been slightly understated (as has long been suspected
by most commentators on these data)?

At issue with the
former fact is why the NBS chose July 2010 to revalue the consumption
expenditure data from the May 2003-and September 1985-based indices,
and why it opted for November 2009 as the new basis for the index? Are
there triggers for such adjustments, and what pray tell, might these
be? Or is this a periodic thing, a key part of the statistical bureau’s
calendar? In which case, when might we expect the next revaluation?

No less important a
line of enquiry is why the NBS did not think as part of this
revaluation of the basis for computing the domestic inflation rate, to
re-weight the makeup of the basket of goods on which its computation is
based.

A good number of
commentators have argued that a basket heavily weighted (65%+) in
favour of food items (including non-alcoholic beverages, and food of
the imported variety) does not sufficiently reflect the spending
patterns of the most economically active segments of the Nigerian
economy. Thus, if the quantum of spending is what matters to the
movement of prices in the economy, we shouldn’t be looking to the
all-in costs of the average Nigerian to explain such movements.
Instead, we should look to an index based on a basket that reflects the
spending patterns of the financially significant sections of this
economy.

In a sense, one
cavils a bit here. For, evidently, the NBS’ adjustment is driven by its
sense of a change in the mix of goods and services purchased by the
“typical Nigerian consumer”. To gauge the trajectory of this change,
one need only look at the composition of the basket over its many
incarnations, and the weights attached to the respective components.
For the September 1985-based index, the food component alone accounted
for 69% of the basket. This index also included a category “drinks,
tobacco, and kola”, which accounted for a little under 5% of the
basket. With the May 2003-based index, “food” alone had a weight of
63.76%. Add “non-alcoholic beverages”, and this weight rises to 64.41%.

With the new base
(November 2009), however, the bureau has introduced a new category
(“imported food” with a weight of 13.25%), on account of which the old
“food” category now has a much reduced weight of 50.70%.

The question is
what changes do these new categories tell of? Do they reflect a change
in the composition of domestic shopping baskets? Do they reflect the
entrance of new goods and services into the shopping basket, the way a
properly reflective index ought to have in the earlier parts of this
decade when telecoms spend entered household balance sheets to an
unusual extent? Or do they just show how granular the bureau can be
now, in the light of new skills and/or software?

The balance of
evidence favours more granular number crunching competences. The makeup
of the measured basket remains basically unaltered. But should we then
replace the one measure of price inflation (that of the man in the
street, plenty of numbers both, no spending weight) with the other
measure (where the size of the purses, and the items on which spending
goes are statistically significant)?

Not necessarily!
For each measure tells a different story about both the character of
the economy and its likely direction. This latter observation leads on
to another question: why do we have just one measure of inflation in
the country, when in some countries they make do with up to five?

Would, for
instance, monetary policy be better designed if we had some measure
that excluded more changeable items in the current basket? What would
we exclude from our own version of the “personal consumption
expenditures price index” to compensate for those short-term price
changes that could interfere with proper estimates of future long-term
inflation trends in the economy?

Point is that the NBS’ recent adjustments raise a few more questions than they do address.