Abstract:
Tea has been cultivated and consumed in China for more than two thousand years and tea
growing in Rwanda started in 1952. By 2002 tea became Rwanda’s largest export, with
export earnings from tea reaching US$ 18 million equating to 15,000 tons of dried tea. The
problem addressed in this study is that tea farmlands increase has been constrained by
population growth in Rwanda, impacting on the quantity supplied and tea exportation is
volatile and very much influenced by real effective exchange rate, Income of major trading
partners, total investment as a proportion of GDP, tea world market price, the coffee worldiv
price among others. This is happening despite the effort made by the government of
Rwanda aiming to increase both tea farmland and the quantities exported in order to gain
from tea trading. In addition to the above highlighted issues, there are little recent empirical
researches investigating factors affecting tea export for a long-run period. The purpose of
the study was to analyse the export of tea in Rwanda for the period of 1982-2012. The
study aimed to establish whether the volatility in quantity of tea exported is related to
world price fluctuations and/or other factors affecting the quantity of tea exported. The
study applied a causal (explanatory) research design, with an objective to explain the
cause-effect relationship between variables. Since the study used a time series data
analysis, secondary data were collected from different sources. Data analysis was done
using ordinary least squares to estimate the model and establish trends of variables
included in the study. Recursive Residual test and Chow test were applied in analysing the
presence of structural change. To investigate the factors affecting tea export of Rwanda,
various tests were done: The unit root tests were conducted using the Augmented Dickey-
Fuller (ADF), the Phillips-Perron (PP) and correlogram tests. Co-integration test was
conducted using the Johansen’s procedure, as series were found to be integrated of the
same order in Phillips-Peron test and co-integrated; Vector Error Correction Model
(VECM) was estimated to adjust the series into their both short-run and long-run
equilibrium conditions. Heteroscedasticity test was done in order to assess whether
stochastic error terms were constant or not, autocorrelation test was done to check if
stochastic error terms were correlated for different period of time. The study used export
theory. A regression model was used as a production function of the quantity of tea
exported as endogenous variable and tea world price, coffee world price, Gross Domestic
Product of major importing countries, Real Effective Exchange Rate, investment as
proportion of GDP as exogenous variables. The study depicted that in the periods of 1988-
1989, 1992-1995 and 2001-2004 structural changes occurred in tea export due to Structural
Adjustment Programmes (SAPs) and decline in price of coffee, liberation war and Tutsi
genocide and decline in tea world price respectively. The research concludes that non
causality in the long term among all variables may be possible because prior to later 80s it
was pre-SAPs economy period thus export was totally controlled one. Short run causality is
possible because after 80s, in short period a decline in world tea price may lead to increase
in export in the sense that the demand increases. The study recommends the government to
help in improving investments in factories and all stakeholders to prioritise ease forward
selling or contracting and widening niche markets for tea products.