Faculty of social studies department of economics supply response of tobacco output to price changes in

Figure 1: The Cobweb model
Source: Adapted from Kaldor (1934)
7

Assuming static producer expectations, given current a high producer price, producers will increase production, however due to excessive supply prices will fall. Realising low prices, farmers will reduce production and supply, but due to a market shortage (excess demand) price will rise and the process continues.
The theory assumes producers have static expectations and thus they base their production plans on past experiences on prices. Hence this study also assume tobacco farmers in Zimbabwe have static expectations and they do make their production decisions basing on past experiences on prices
2.2 Empirical Literature
A lot of studies have been conducted throughout the world modelling the supply response behaviour of certain agricultural products and which include studies by Leaver (2004), Dean
(1966), Muchapondwa (2008), Sankor (2009), Townsend and Thirtle (1997), Thiele (2002) among others.
Leaver (2004) carried a study to measure the supply response function of tobacco in Zimbabwe. The main aim was to estimate the price elasticity of supply for tobacco output in Zimbabwe using an adapted Nerlove (1958) model using time series data for the period 1938 to 2000. Dummy for average rainfall, sales quotas, time trend, time trend squared, real price of tobacco and lagged tobacco output were included as explanatory variables. A logarithmic equation was estimated using the ordinary least square (OLS) approach. Short-run and long-run price elasticities of supply were found as 0.34 and 0.81 respectively indicating that farmers are unresponsive to prices. Time trend, the lagged tobacco output and lagged real tobacco prices were found to have a positive influence on tobacco production whilst sales quota dummy, dummy for average rainfall and the quadratic time trend were found to have a negative impact. The time period of 1938 to 2000 represents both the pre and post-independence era in Zimbabwe therefore the data set may not be adopted in this study since the data of interests is that of the post-independence era.

Another study by Sankor (2009) used time series data for the years 1968 to 1984 in analysing factors responsible fora decline in tobacco production in Trinidad and Tobago. The study utilised the Nerlove (1958) distributed lag model using acreage planted as the dependent variable. Independent variables included were lagged price, competing crops, agricultural wage
8

rates, rainfall and lagged acreage. Results were examined within the framework of various a priori hypotheses about supply responsiveness of tobacco farmers. Although some factors such as price of competing crops will be adopted, agricultural wage rates will not be considered because the data is not readily available.
Furthermore, Muchapondwa (2008) estimated aggregate tobacco supply response in Zimbabwe using time series data for the period 1970 to 2000. The autoregressive distributed lag (ARDL) approach to cointegration capturing both short-run and long-run dynamics was employed. Variables included were current prices, lagged tobacco prices and output. Short-run and long-run price elasticities of supply were estimated at 1.21 and 0.18 respectively indicating that tobacco output is responsive to prices in the short-run only. The study concluded that agricultural price policies are blunt instruments for effective growth in agricultural supply and thus provision of non-price factors must play a key role in reviving the agricultural sector in Zimbabwe. Although tobacco output was found to be highly responsive to price incentives in the short-run, the study did not include most of the non-price factors such as extension, population of farmers, credit and research.
An empirical investigation on the supply of maize and tobacco for commercial agriculture in Zimbabwe employing the Error Correction Model (ECM) was presented by Townsend and
Thirtle (1997). Annual time series data for the period 1970 to 1989 were collected on expected maize price, real tobacco price, fertiliser prices and government intervention. Area planted to tobacco was treated as the dependent variable. Real price of tobacco, expected price of maize and institutional factors were found significant whilst fertiliser price was found insignificant. Short run and long run price elasticities for maize were 1.44 and 1.76 whilst for tobacco were
0.28 and 1.36 respectively showing that commercial farmers are unresponsive to output prices in the short-run. The results were consistent with a priori expectations. However, the use of area planted as a dependent variable may not be a correct measure of output since farmers can also increase output by farming their land more intensively rather than by utilising more land.
Making use of a logarithmic supply response equation, Dean (1966) carried a study in Malawi with an attempt to find out whether African tobacco farmers are responsive to price changes in

the same manner as farmers in developed countries. Time series data for lagged tobacco prices, lagged wage rates and population of tobacco growers for the 35 years from 1926 to 1960 were
9

used. A long-run price elasticity of supply for tobacco was found to be 0.48 implying that tobacco farmers are unresponsive to changes in prices. All variables except for wage rate were found significant at 1% level. Therefore in this study, similar results to Dean’s (1966) findings for Malawi are expected since both Malawi and Zimbabwe are Southern Africa developing countries.
Thiele (2002) carried a study attempting to answer the question of how responsive are farmers in
Sub-Saharan Africa to both price and non-price incentives. The study employed the Johansen's multivariate approach to cointegration using time series data for the period 1965 to 1999. The main objective was to investigate the long-run effect of pricing policies, macroeconomic distortions and non-price factors for ten selected countries. A double log-form equation was employed and variables included were output, real prices, real exchange rate and dummy variable for drought and irrigation. Estimated supply elasticities were found to lie between 0.2 and 0.5 and also the study found out that drought episodes have a negative significant impact on agricultural growth of six of the ten sample countries. The deterministic time trend used as a proxy for technical progress was found to have a small positive significant impact on agricultural output. Although macroeconomic distortions were included by Thiele (2002), they will not be used in this study because of lack of data on macroeconomic distortions.

Download 1.11 Mb.

Share with your friends: