Over decades, there has been no commodity price that has received so much attention predictions and speculations than crude oil price. However, intellectuals always fail in their predictions to produce the expected results. Crude oil in its original state has no value to the consumer until it is transformed into useable products such as Liquefied petroleum Gas (LPG), Kerosene, Gasoline, Diesel and other petrochemicals.  In most countries, motor fuels (Gasoline and Diesel) are always dominant than other petrochemicals. But Liquefied Petroleum Gas and other natural gas are becoming more visible in today’s global oil and gas industry.  Like any other commodity in the market, the prices of these products as well as their demand, keep responding to changes in crude oil price and other market conditions. This movement in fuel prices causes worry among pundits, politicians and everyone who uses these products. Petroleum products get to the final consumer or to the market through different channels. The sales of these products also vary from one market or geographical location to another. Refiners either sell to wholesalers or sell to company-owned or company franchised retail sites.  For example, in the United States of America, the open market dominates and has a greater share of the market. However, in countries such as Malaysia, Mexico and most developing countries National Oil Company (NOC) own retail outlets and hypermarkets control majority share of the petroleum product market. In most West African countries one can even see by the roadside, these products filled in bottles and jars for sale. 
This shows how important these products are in the day to day activities of consumers.
Ghana imports crude needs from other countries to meet almost all of its crude needs and as such face the consequences of price volatilities. Ghana like most other developing countries is faced with the problem of a pricing scheme of petroleum products that will be acceptable to the population but which will not also cripple the economy. This is because the government of the day always subsidise the prices of these products as a way of reducing the impact on the ordinary Ghanaian consumer  . Since 2000, there has been a consistent rise in the consumption level of petroleum products even though the prices of these products keep increasing  . Many have tried to analyse the effect of changes in the price of these products on the final demand and on the welfare of the consumer in other countries. But, there seems to be little literature on that of Ghana.
This study therefore seeks to analyse the effect of changes in crude oil price on the demand of petroleum products in Ghana. The underlying assumption is that, any change in crude oil price manifest itself in the prices of the derived products such as petrol, diesel, LPG, kerosene and other petrochemicals. The study employed an econometric method as a way of examining and analysing the effect of changes in price of the various products on the total demand of petroleum products (petrol, diesel, LPG and Kerosene) in Ghana from 2000 up to 2011. This study is organised into four chapters. Chapter one covered the Introduction. The background as well as the demand for petroleum and the theoretical framework are captured in chapter two. Chapter three contained the analysis and discussions of the findings. Chapter four is the concluding chapter. It also captures some recommendations.
2.1 Global Demand for Petroleum Products
The oil and Gas industry is one of the largest and complex global industries which touch almost every part of the human life with products such as gasoline, kerosene, gasoil, lubricants and thousands of petrochemicals products. The demand for crude oil is derived demand  . As indicated earlier on, crude oil in its original state has no value to the consumer until it is transformed into useable products such as Liquefied petroleum Gas (LPG), kerosene, Gasoline, Diesel and other petrochemicals  . The demand for these products over the past decade has been increasing in both industrial and developing economies  . There is no product price that has received so much predictions and speculations than crude oil price. But these predictions in most cases failed to produce the expected results. Like any other commodity in the market, the prices of these products as well as their demand, keep responding to changes in crude oil price and other market conditions  . According to Suleiman 2009, consumption of petroleum products in Indonesia has grown significantly and as a result, by 2004, Indonesia had become a net-importer of both crude oil and refined products  .
2.2 Demand for Petroleum Products in Ghana
The Ghanaian economy depends solely on import for its petroleum needs. The Tema Oil Refinery (TOR) is the only refinery in the country charged with the responsibility of refining crude oil to produce petroleum products for domestic demand  . But, its inability to carry out this mandate effectively and efficiently has given rise to importation of some petroleum products from neighbouring country  . Approximately, 70% of the demand is met with products are produced by TOR and the remaining 30% from imports. The country imported its entire crude oil requirement as at December 2010, which is refined at TOR. However, this is looking to change as the country now produces oil in commercial quantities.
The petroleum products marketed in Ghana are Premium Gasoline, Kerosene, Residual Fuel oil, Gas oil, LPG and Premix  . Bulk supply of these products is reliable and is done through an extensive infrastructural network comprising of storage depots located at strategic parts of the country, pipelines for the movement of petroleum products, Bulk Road Vehicles (BRVs) and also barges located on the Volta Lake  . The actual supplies however improved compared to the previous year and for that matter shortage of LPG and diesel were not as serious as in 2010. There are a number of factors influencing the consumption of petroleum products in the country. Among these factors are prices of the various products, real income of the consumer, changes in the number of consumers as well as the availability of these products. For instance, significant increases in retail prices of LPG for vehicular fuel have helped in reducing its consumption and on the other hand increased consumption of gasoline and invariably the sales of the latter  .
Inability to raise the required and adequate funds for the required LPG import due to cross-subsidisation enjoyed by the product might have also contributed to moderate LPG consumption compared to the Energy Commission’s forecast during the year  . It was projected that the requirements for the major petroleum products-gasoline, kerosene, gas oil and LPG will increase significantly from 1.62 million tons in 2005 to 2.49 million tons by 2015 representing an annual growth rate of 5.3% over the period  .
Access to petroleum products in Ghana is satisfactory, but has to be improved as the population grows.
As at 2010 there were a total of 1700 petroleum products retail outlets of which 37% are Service Stations (SS), 20% are Filling Stations (FS) and 43% are Reseller Outlets (RO)  . The total number of retail outlets represents an access ratio of 71 retail outlets per 1,000,000 people  . There are also vendors who sell mainly kerosene in rural communities, thereby increasing the accessibility of petroleum products in most part of the country.
The National Petroleum Authority (NPA) is a statutory agency responsible for regulating, overseeing and monitoring the petroleum downstream industry in Ghana to ensure efficiency, growth and consumer satisfaction  . It is also charged with the responsibility of monitoring and regulating petroleum prices in accordance with the prescribed pricing formula. With strong economic growth and steady population growth over the past years, Ghana’s total energy demand for petroleum products has grown substantially. Table 1 show the total demand for the various petroleum products in the country. It also included the Gross Domestic Product (GDP) of Ghana from 2000 to 2011. Figure 1 shows the trend of movement for the same period under study.
TABLE 1: PETROLEUM PRODUCT CONSUMPTION AND GDP
PETROLEUM PRODUCT CONSUMPTION
GROSS DOMESTIC PRODUCT (GDP)
SOURCE: National Energy Statistics 2000-2011
From the table above, the total energy demand for petroleum products 5,536.30 kilotonnes in 2000. This increased to about 8,877.10 by the end of 2011. It shows an average annual increase of 4.01% in energy demand. GDP which shows the real income of the country was US$4,977.50 billion as at 2000 and increased to US$ 9,890.90 billion by the close of 2011. This also shows an average annual growth rate of 4.5% within the same period understudied. This is a clear indication that, as the real income of the country increases, more petroleum products are demanded even though price of crude oil was on the increase as indicated in table 2 below.
FIGURE 1: TREND IN GDP AND TOTAL PETROLEUM PRODUCT DEMAND
From figure1 above, it is clear that, there is a strong correlation between petroleum product demand and the GDP of Ghana. The trend shows that as GDP grows, demand for petroleum products grows as well. Its further shows that as the real income of consumers increase, they tend to demand more of these products and hence movement of the two curves in the same direction.
TABLE 2: PRICES OF CRUDE OIL AND SOME PETROLEUM PRODUCTS
CRUDE OIL (Brent price US$)
DIESEL PRICE (US$/l)
PETROL PRICE (US$/l)
KEROSENE PRICE (US$/l)
LPG PRICE (US$/kg)
SOURCE: National Petroleum Authority
From Table 2 also, it can be seen that crude oil price has been volatile, but generally on the rising side from 2000 up to 2011. The price of crude oil was US$ 28.5 in 2000 which decrease to about US$ 25 in 2002. It picked up again in 2003 and rose continuously to a high of US$ 111.3 by the end of 2011. Also prices of the various petroleum products responded accordingly to changes in the price of crude oil as indicated in table 2 above. In 2000, diesel was sold at US$ 0.4 per litre, petrol US$ 0.3 per litre, LPG US$ 0.5 per kg and kerosene was US$ 0.3 per litre. However, as crude price increases, the prices of diesel, petrol and kerosene also increased to US$ 1.1, US$1.0 and US$0.6 per litre respectively by the end of 2011. LPG also increased to US$ 0.7 per kg over the same period.
FIGURE 2: TREND OF PRICES OF CRUDE OIL AND PETROLEUM PRODUCTS
From Figure 2 above, it is also clear that, the prices of the various petroleum products respond positively and directly to changes in crude oil prices. This can be explained by the fact that, these products are derived from crude oil and hence, any change in its price or quantity will have corresponding effects on its products.
2.3 Theoretical framework
The price of crude oil has been characterised by major swings especially in the 1970s and 2008 and has always impacted on the demand for petroleum products. This has increase the interest of many economists and other intellectuals and has since led to a number of studies that seek to investigate the relationships between real income, real prices and the consumption of petroleum products  .
Dahl (1994) conducted a survey of petroleum demand in developing countries  . The result showed that the average price elasticity of demand of petroleum products for developing countries was -0.36, and income elasticity was 2.20. This suggested that, the demand for petroleum products is more responsive to changes in income than changes in real prices. Suleiman S.2009, used selection criteria from various models as way of cointegration approach to estimate the price and income elasticities of demand for total petroleum products such as gasoline and diesel in Indonesia. The results suggested that both total products and gasoline share estimates are more responsive to changes in income than changes in the real price of petroleum products. Based on this, he suggested that “policy makers may need to use market-based pricing policies and other policies such as public enlightenment in addition to regulations like minimum energy efficiency standards to promote efficiency and conservation and curb the rising consumption of petroleum products in Indonesia”. 
Akin et al 2009 conducted empirical studies which sort to estimate the demand elasticities of petroleum products in Nigeria. The study employed a multivariate cointegration approach to estimate both the short-run and long-run price and income elasticities. The result showed that, energy consumption responds positively to changes in GDP and negatively to changes in energy prices. The study concluded that both price and income elastticities are inelastic even though the elasticities of demand vary according to product type. 
Maria et al 2011 employed an econometric method developed by Deaton to calculate the price and income elasticities of demand for domestic energy in Mozambique. The calculations were done for all households at the national level, differentiating urban household from that of rural folks. The result showed that the responsiveness to changes in both price and income were higher for high-grade energy sources such as gasoline, diesel and LPG and lower for low-grade sources such as charcoal and fire-wood. They concluded by indicating that, “the potential for energy transition in the domestic context is not directly visible from the price and income elasticities”. 
From the theoretical framework above, Dahl (1994), Suleiman S. (2001) and Akin et al all employed the cointegration approach in analysing the relationship between demand of petroleum products and changes in price. Maria et al (2011) applied an econometric method developed by Deaton in their study  . This study follows the approach adopted by Maria et al (2011). It is advantageous because it uses unit values as the prices of goods and hence provides the opportunity for the use of individual fuel prices. However, the usage of these unit values can lead to biased elasticity estimates  .
The total demand for petroleum products (share of gasoline, Diesel Kerosene and LPG) can be specified as a function of real per capita GDP and real prices of Gasoline, Diesel, Kerosene and LPG. This can be modelled as follows:
ED= f (GDP per capita, Price of Petrol, price of Diesel, price of Kerosene and price of LPG)â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦ (1)
Forming a linear regression equation from the function gives;
Ed= a0 + b0GPD/p + c0pP + d0pD + e0pK + f0pLPG + u0â€¦â€¦â€¦â€¦â€¦â€¦â€¦ (2)
Both the dependent variable and the independent variables are transformed to natural logarithms and hence double log estimation equation. This is written as follows
lnEd= a0 + b0lnGPD/p + c0lnpP + d0lnpD + e0lnpK + f0lnpLPG + u0â€¦â€¦â€¦â€¦â€¦â€¦â€¦ (3)
Edâ€¦â€¦.. Demand for petroleum products
GDP/pâ€¦â€¦.. Gross Domestic Product per capita
pPâ€¦â€¦â€¦â€¦..price of petrol
pDâ€¦â€¦â€¦â€¦..price of Diesel
pKâ€¦â€¦â€¦â€¦â€¦ price of Kerosene
pLPGâ€¦â€¦â€¦â€¦price of Liquefied Petroleum Gas
3.0 ANALYSIS AND OUTCOMES
Using the double log estimation equation as;
lnEd= a0 + b0lnGDP/p + c0lnpP + d0lnpD + e0lnpK + f0lnpLPG + u0â€¦â€¦â€¦â€¦â€¦â€¦â€¦
Edâ€¦â€¦â€¦â€¦.. Demand for petroleum products
GDP/pâ€¦â€¦.. .Gross Domestic Product per capita
pPâ€¦â€¦â€¦â€¦.. price of petrol
pDâ€¦â€¦â€¦â€¦. price of Diesel
pKâ€¦â€¦â€¦â€¦..price of Kerosene
pLPGâ€¦â€¦â€¦.price of Liquefied Petroleum Gas
TABLE 3: REGRESSION ANALYSIS
Source: Arthur’s construct
3.1 Discussion of outcomes
3.1.1 Gross Domestic Product per capita
From the analysis summarised in table 3, GDP per capita as a variable shows a positive sign in its coefficient against demand for petroleum products. This implies that, as real per capita income of consumers’ increases, their demand for petroleum products increase as well. Its p-value of 0.022904 is also below the significant level of 0.05 (95% confidence) which is an indication that, the effect of changes in GDP per capita on the demand for petroleum products is statistically significant.
In terms of elasticities, a coefficient of 0.081134 for petrol shows that, the demand for it is inelastic. This implies that changes in the price of petrol leads to less than proportionate change in demand. The positive sign however does not follow the a priori expectation. But it shows that, over the period under study, there is a direct relationship between demands and price of petrol. Thus, any increase in the price petrol leads to an increase in its demand. The p- value however confirms that, this direct effect is not significant at 0.05 significance level since the value 0.822455 is greater than 0.05 (significance level).
From the table, the coefficient of diesel is quite revealing as it follows the a priori expectations in it demand. It has the expected negative sign which confirms that any increase in it price will result in decrease in it demand and hence decreasing the level of total demand for petroleum products in Ghana. But this is not significant at 95% confidence interval as the p-value of 0.925098 is greater than 0.05.
Also from the table, the coefficient of LPG revealed the a priori expectations in it demand. It has the expected negative sign which confirms that any increase in it price will result in decrease in it demand. This is an indication that, decrease in the level of total demand for petroleum products in Ghana can be the result of increases in the price of LPG.
But this is also not statistically significant at 95% confidence interval as the p-value of 0.276798 is greater than 0.05. Thus, inelastic demand for LPG.
The variable price of kerosene shows a positive sign. This sign again does not follow the a priori expectation. This result implies that, there is a direct relationship between demands of kerosene and its price and hence, any increase in the price kerosene leads to an increase in its demand. But this is not the case or the expectation as the demand for kerosene decrease due to increases in it price. However, the p- value shows that, this direct effect is not significant at 0.05 significance level since the value of 0.732547 is greater than 0.05.
The R2 of the analysis is measures the overall goodness of fit of the regression. The test result indicates that, the R2 =0.978 = 97.8%. It shows that the independent variables (GDP per capita, price of petrol, price of diesel, price of LPG and price of kerosene) explain the dependent (demand for petroleum products) variable to the tune of 97.8%.
This study analysed the effect of oil price change on the demand for petroleum products. This effect is analysed through the changes in the various prices of the major petroleum products (petrol, diesel, LPG and kerosene) consumed in Ghana from 2000 to 2011. The study employed a double log equation in its analysis. It included the Gross Domestic Product per capita as a measure of the real income of consumers. The results show that aggregate demand for petroleum products in Ghana follows with the a priori expectations of negative price elasticity and positive income elasticity. The results also showed that, the demand for petroleum products in Ghana has been on the increase from 2000 to 2011 and future still promise to show more increment if the economy maintain its strong growth pattern.
The analysis of the results show that the real income of the consumer measured by GDP per capita has a positive sign which indicates that, as real per capita income of consumers’ increases, their demand for petroleum products increase as well. Its p-value of 0.022904 is also below the significant level of 0.05 (95% confidence) which shows that, the effect of changes in GDP per capita on the demand for petroleum products is statistically significant.
The test result also shows that, for the period under consideration, the demand for the major petroleum products (petrol diesel LPG and kerosene) is inelastic even though their individual coefficients vary. Their prices continue to increase throughout the period under consideration as oil price increases. Total demand for these products increase as well.
Notwithstanding the values of the study, it should be noted that energy demand might assume different implications in different economic sectors. Different sectors of the economy have different consumption behaviours and as such, have different optimal demand decisions under the same constraints and do not necessarily demand the same services from the various energy forms. Sectoral analyses of the demand for petroleum products will therefore, offer an in-depth analysis in future research works.
In terms of policy implications, the study recommend that policy makers should design policy frameworks that will promote energy efficiency and conservation. These policies can be market-based pricing and taxation policies, policy that can ensure minimum energy efficiency standards and has the tendency of informing the public the consequences of the present trend in consumption and importation. Otherwise, the consumption of petroleum products is likely to continue to grow at a significant rate as long as per capita income continues to grow.