Prices of grains in Nigeria have not been stable over the years. In fact, at some points in time, prices of grains have been outrageous. Anecdotal evidence suggests that a number of macroeconomic variables have been responsible. However, the relative importance of each macroeconomic factor is still uncertain. This study, therefore, determines the factors influencing the price variability of grains in Nigeria. The data for the study, covering the period 1981–2014, were obtained from publications of the Central Bank of Nigeria, National Bureau of Statistics, and Federal Ministry of Agriculture and Rural Development. Econometric analysis was used to establish determinants of price volatility of grains in Nigeria. Results indicate that certain macroeconomic variables (inflation, crude oil price, interest rate, exchange rate, broad money supply, credit to the grain subsector, and domestic grain production) were the major determinants of varying prices of grains in Nigeria. The findings suggest the need for policies that will buffer the grain subsector of Nigeria’s agriculture from the effects of inflation. Also, policies that decrease inflation rate will minimize price variability among grains and consequently reduce inefficiency, distortions and misallocation of resources in this important subsector of Nigeria’s agriculture that might be caused by inflation.INTRODUCTION
The grain sub-sector in Nigeria plays an important role in economic development of the country and it contributes a larger portion of staple food stuff in the country (Agboje et al., 2013). Grains accounted for about half of total food supply in Nigeria. The commonly grown cereal grains in Nigeria are maize, rice, sorghum, and millet. While some of it is actually consumed as food, most is converted into animal feed, ingredients for processed food or feedstock for ethanol (Mansharamani, 2012 cited in Agboje et al., 2013).
One of the determining factors to how much an average poor Nigerian can consume these available energy giving food is price, and the nominal price of the individual grains has continuously fluctuated over the past years (Agboje et al., 2013). Major grain crops in Nigeria have shown broad variations in nominal prices or producer prices over the decades (Akpan and Udoh, 2009). The price of rice increased by more than 100% between 1975 -1979 from that obtained between the previous five years averages (1970 -1974), similar trend was obtained for maize, millet and sorghum not until between 1994 -1999 the prices were lowered by less than 40% of their preceding prices (Akpan and Udoh, 2009).
Prices of food commodities on world markets, adjusted for inflation, declined substantially from the early 1960s to the early 2000s, when they reached a historic low (FAO, 2011). They increased slowly from 2003 to 2006 and then surged upwards from 2006 to the middle of 2008 before declining in the second half of that year, and the sudden increases led to increased concern over the ability of the world food economy to adequately feed billions of people, presently and in the future (Agboje et al., 2013). Although various observers attach differing degrees of importance to assorted factors, there is a relatively strong consensus that multiple factors had a role in the price increases that began in 2003 (FAO, 2011).
The current price volatility of grains presents a puzzle, thus raising concerns among researchers and policymakers. A number of macroeconomic factors have been identified for this price volatility in Nigeria. However, the relative importance of each factor is still unclear in Nigeria. Moreover, the quantitative estimates vary from one country to another (Olomola and Adejumo, 2006; Blanchard and Gali, 2007; Aksoy and Isik-Dikmelik, 2008; Dessus et al., 2008; Ivanic and Martin 2008; Wodon et al., 2008; Aliyu, 2009; De Janvry and Sadoulet, 2009; Kumar, 2009; Asian Development Bank, 2011; Fezzani and Nartova, 2011; Freire and Isgut, 2011; Heady, 2011; Ortiz et al., 2011; Olomola, 2013). The large variations in quantitative estimates of impact of price variability of grains on food security can be explained by differences in methodologies and assumptions – which often are not clearly understood by policymakers. Against this background, this paper develops a state-of-the art dynamic macroeconomic model that will effectively link the volatile prices of grains to the various interconnected factors. In this study, the link between grain price volatility and the various interconnected macroeconomic factors are modelled.
Data were obtained from the publications of Central Bank of Nigeria, National Bureau of Statistics, and Federal Ministry of Agriculture and Rural Development. The data covered the period 1981–2014. Time series data on Gross Domestic Product, inflation, crude oil prices, interest rate, exchange rate, broad money supply, credit, total savings, annual output of the various grains (rice, maize, sorghum, and millet), and prices of the grains were collected from 1981 – 2014. Data collected were estimated by ordinary least squares method. The augmented Dickey–Fuller (ADF) test was carried out to determine the time series properties of the variables. The implicit model is Y = f(X1,X2,X3,X4,X5,X6,X7,X8,X9, X10,e).
Y = Price of each grain (Naira)
X1 = Gross domestic product (Naira)
X2 = Inflation rate (%)
X3 = Crude oil prices (Dollars/barrel)
X4 = Interest rate (%)
X5 = Exchange rate (Naira/Dollar)
X6 = Broad money supply (Naira)
X7 = Credit to grains sector (Naira)
X8 = Total savings (Naira)
X9 = Domestic production of grains (Tonnes)
X10 = Time period
e = error term
The Augmented Dickey Fuller (ADF) Test was also used to test for the number of cointegration vectors in the model. Johansen technique was suggested by Maddala (2001) not only because it is vector auto-regressive based but because it performs better in multivariate model. However, for this study, ADF was found to be best. If two-time series variables, pt and qt, are both non-stationary in levels but stationary in first-differences, i.e., both are I(1), then there could be a linear combination of pt and qt, which is stationary, i.e., the linear combination of the two variables is I(0). The two-time series variables that satisfy this requirement are deemed to be cointegrated. The existence of co-integration implies that the two co-integrated time series variables must be drifting together at roughly the same rate (i.e., they are linked in a common long-run equilibrium). A necessary condition for co-integration is that they are integrated of the same order. To check whether or not two or more variables are co-integrated, it is necessary to first verify the order of integration of each variable by performing unit root tests (Granger 1986; Engle and Granger 1987). RESULTS AND DISCUSSION
The results of the Augmented Dickey Fuller (ADF) unit root tests for the climate variables for each of the crops are summarized in Table 1. According to these results when price of maize, price of sorghum, price of millet or price of rice is used as the endogenous variable, the null hypothesis of a unit root cannot be rejected at conventional (10%, 5%, or 1%) significance levels for price of maize, price of sorghum, price of millet or price of rice; GDP, inflation rate, crude oil price, exchange rate, output of maize and output of rice in level, but is rejected at the 1% significance level for all of the time series in second difference. These results imply that each series is non-stationary in level but stationary in the second difference.Having established that all the variables are I(2) (i.e. the necessary condition for cointegration is satisfied), the results of the cointegration tests are reported in Table 2. The numbers in parentheses for the ADF test are the optimal lag lengths. Since the results of the unit root tests on the OLS residuals of the cointegration regression does not reject the null hypothesis of a unit root in favour of the stationary alternative even at the 10% significance level, we conclude that the series are not co-integrated. In other words, they are not linked in common long run equilibrium. Having established the fact that the variables are not co-integrated, the regression analysis was performed and a summary of the result is presented in Table 3. Table 3 shows the regression result of the macroeconomic determinants of price volatility of selected grains (maize, sorghum, millet, and rice) in Nigeria. The coefficient of multiple determination (R2) for each regression shows a good fit. For instance, that of price of maize against the macroeconomic variables was 0.524 (52.4%) (Table 3), implying that 52.4% of the variation in the price of maize in Nigeria is explained by macroeconomic variables like GDP, inflation, exchange rate, interest rate, crude oil price, broad money supply, savings, credit to the grain sector, and output of maize. Again, the significance of the F-ratio (p