Geoscience Reference
In-Depth Information
consequences of an adverse climate change could, therefore,
have a significant negative effect on rice production.
11.6
Data and methodological approach
The districts for which data on rice productivity and irrigation
acreage are available are the primary spatial units of analysis in
this study. Of the total 75 districts of India (39 in the Hills, 20
in the Terai and 16 in the Mountains), 73 districts are included.
Two districts in the Mountain (Mustang and Manang) are not
included since rice is not grown there. The choice of the district
as the unit of spatial analysis is further justified because it is the
smallest administrative unit that contains the full complement
of government services. For example, in the agriculture sector,
every district has a government-run Agricultural Development
Office (ADO) that employs agricultural extension workers
responsible for promoting improved technologies. Each district
is also supplemented by the office of the Agricultural Input
Corporation (AIC) and the Agricultural Development Bank
(ADB); government subsidiaries established to market agro-
technologies to the farmers. In addition, the Department of
Irrigation (DOI) has its offices at the district level, which are
responsible for developing irrigation infrastructure. All these
agencies are pivotal in the development of specific agricultural
technologies needed in various agro-climatic regions of India.
This study is based on secondary data obtained from the
various agencies of the government of India. The data concern-
ing rice yield (productivity and yield are used interchangeably
to indicate mean output per unit of land) and irrigation were
obtained from the India Agricultural Database (NAD) of the
Ministry of Agriculture and Co-operatives (MOAC). The aver-
age monthly rainfall data were obtained from the DOHM. The
DOHM has compiled the average monthly precipitation for the
period between 1968 through 1997 from the records of various
meteorological stations throughout the country, and has used
the data to represent the monsoon rainfall in this analysis.
Methodologically, following Barro and Sala-I-Martin (1992),
convergence can be understood in two ways: convergence in
terms of the level of productivity across time, that is, sigma
(σ) convergence and the rates of productivity growth across
space and time, that is, beta (β) convergence. Conceptually, the
two measures used in the literature to test for convergence are
related and provide alternative ways to examine similar phe-
nomenon. In this chapter, convergence in terms of the level of
