Food and agricultural production account for a substantial share of economic activity in developing
countries like Fiji. The relatively large size of the agricultural sector makes it all the more important to
base agricultural policy decisions on reliable data. Moreover, improvements in farm sector productivity
are an important driver of economic growth in developing countries. Good data provide governments
? and citizens and taxpayers ? with a more reliable basis for identifying policy issues and for assessing
the aggregate and distributional impacts of policy initiatives. Good data help to improve the quality of
both ex ante and ex post analyses and so help to make governments more accountable for their
decisions. In democratic settings, such improved accountability will help to ensure that to the extent
possible public policy does indeed promote high economic growth and development.
Despite the conceptual limitations of gross domestic product as a measure of economic activity, levels
of and changes in a country?s GDP do provide valuable information for government decision making.
However, there is some largely anecdotal evidence that the official GDP estimates for the Fiji Islands are
not sufficiently accurate. The potential errors reflect misreporting associated with the black economy and
problems in measuring agricultural activity in the informal sector. These informal sector problems come
from both sampling and non-sampling sources. To gain some insight into the size of the informal part
of the agriculture sector we estimate a model to explain the dependence of households on home
produced food. Based on this model, and other data sources, we conclude that official estimates of the
relative size of the agricultural sector in Fiji probably underestimate the true size of the sector and
underestimate its growth.