Estimating the Demand for Energy Products in Nigeria

Estimating the Demand for Energy Products in Nigeria

Dr. Emmanuel Ojameruaye

emmaojameruaye@yahoo.com

A recent report reveals an ongoing controversy between the Federal Ministry of Finance (FMF) on the one hand and and the Petroleum Products Pricing Regulatory Agency (PPPRA) and the Nigerian National Petroleum Corporation (NNPC) on the other hand over the projected quantity of fuel consumption used for the determination of subsidy requirements for 2012. The report indicates that the FMF is insisting on 19 million litres a day while the PPPRA and NNPC are insisting on 33 million litres a day (Aderinokun & Alike, Thisday, May 28, 2012). Responding to the difference, the Senior Special Assistant to the Minister of Finance stated that “the issue of underestimation does not arise because the NNPC and PPPRA which had the responsibility did not provide any basis for estimation of consumption. They both gave very different estimates which could not sustain scrutiny. The ministry did the best thing in the circumstance which was to use an economic estimation method to arrive at the projected amount of subsidy for 2012 based on growth in income and growth in demand. This amount was based on the consumption for 2008 which was adjudged as a good base year from an assessment of various relevant factors”. He further revealed that the FMF has “commissioned the globally respected consulting firm, Mckenzie, to use their worldwide resources to help us estimate the consumption… to provide the authentic figures”.

In their response, a spokesman of the NNPC stated that “the Ministry of Finance was about to plunge the country into fuel crisis by misleading the government on the daily consumption of petrol… How can anybody say that Nigeria with a population of 167 million consumes 19 million litres of PMS daily? The Ministry of Finance made provision for subsidy based on 19 million litres pay day, instead of over 33 million litres, which the country is consuming”.

This controversy between the FMF and the PPPRA/NNPC over the projected fuel consumption level reveals a fundamental problem that has bedeviled energy planning in Nigeria. This problem is the estimation of the demand for various forms of energy, particularly electricity and petroleum products, under conditions of disequilibrium and imperfection. It is known fact that both the electricity market and the petroleum product market are characterized by chronic disequilibrium as evidenced by frequent electric power outages and load shedding or rationing, and frequent fuel scarcity resulting in long queues at the filing stations. In the case of petroleum products, there are also problems of smuggling, adulteration, over-invoicing and round-tripping. In the case of electricity, there are also the problems of illegal tapping of electricity, the use of private generators and use of “estimates” by PHCN in the billing of many customers who do not have electric meters due to the shortage of meters or the long time it takes to provide meters to consumers. Under such conditions of disequilibrium, there is usually a significant difference between the consumption and “ex-ante” demand - the quantity that consumers are willing and able to pay for at a given price. When there is shortage or rationing, the actual (metered) consumption is usually less than “ex ante demand” because some consumers are forced to do without the product or to use alternatives. For planning purposes, it is ex-ante demand that matters. As a matter of fact, using raw consumption data (which is essentially the supply or energy sales) as a proxy for demand in the process of planning when the past is characterized by disequilibrium will result in perpetual or chronic disequilibrium. Unfortunately, this appears to have been the situation in Nigeria over the years because in making consumption or demand projections the planners of electricity supply and petroleum products supply have always failed to take into consideration the disequilibrium and imperfections in the markets. For instance, the PHCN uses metered consumption (electricity sold) as the basis for making projections of future demand or consumption of electricity. Similarly, the NNPC/PPPRA uses metered consumption (petroleum products sold at filling stations/outlets or domestic production + quantity imported) as the basis for making projection of future demand and fuel subsidy requirements. Such a crude approach will perpetuate disequilibrium by making the future a mirror image of the past. Furthermore, both the NNPC/PPPRA and the PHCN use simple time trend in projecting future consumption, i.e. they assume that consumption is a linear function of time, instead structural econometric models that specify demand or consumption as function of several “explanatory” variables such as disposable income, industrial production, number of consumers or population, unit price of the product, price of substitutes, and stock of energy-consuming equipment or machines.

Under conditions of disequilibrium, an “econometric disequilibrium” approach is the gold standard for projecting future demand for planning purpose. The econometric disequilibrium approach is based on the theory that “in the absence of an equilibrium condition, the ex-ante demand and supply quantities cannot in general be equated to the observed quantity traded in the market. The observed quantity is then equal to minimum of ex ante demand and supply quantities. Thus in attempting to estimate the supply and demand schedules, the sample should be separated into demand and supply regimes..Alternatively, the observed quantity should be adjusted for the effects of rationing and then both supply and demand supply schedules fitted over the entire period using the adjusted quantities” (Fair & Jaffe: Methods of estimation for markets in disequilibrium. Econometrica, Vol. 40: 497-514, 1972)

With respect to the Nigerian electricity market, about 30 years ago, a leading Nigerian energy economist noted that “Effective planning requires a good knowledge of the future level of the pattern of demand…Underestimation of demand in an environment of relatively high growth often leads to underinvestment, and consequently, an in-built disequilibrium between demand and supply. This situation has been characteristic of Nigeria’s power situation (Iwayemi, A: Planning in the electric power sector in Nigeria, in Development Economics and Planning. Unibadan Publishing Consultants, 1982). Also in 1984, another Nigerian energy economist also recommended that “NEPA must have an established index of reliability and all planning must be towards meeting this index” (Chuku, A. Electric power supply adequacy in Nigeria outlook for the 1980s and beyond. Journal of Issues in Development. Vol. 1 (1): 82-94, 1984). At about the same time, I constructed and estimated an econometric disequilibrium model of the Nigerian electricity market which I used to estimate the future demand for electricity (from 1983 to 1990) and to determine the “optimal” unit prices of electricity (Ojameruaye, E: An econometric disequilibrium model for electric power system planning in Nigeria. OPEC Review. Winter, 1988). In the model, I demonstrated that because of the disequilibrium that characterizes the electricity market, D = C/R where D = demand, C = metered consumption, R – power reliability index, defined as the ratio of the duration of time during which there was no power outage to the total accounting time. For instance, if we are using annual data, R = (8760 – Z)/8760 = 1 – Z/8760, where 8760 is the number of hours in a year, Z = the number of hours during the year when there was power outage. For instance, if during the course of the year, the duration of power outages add up to 438 hours or 18.5day, the reliability index R = 1 – 485/8760 = 1 – 0.05 = 0.95 = 95%, which means that consumers on the average had power supply 95% of the time. If the metered consumption (quantity of electricity sold to consumers) is that year is 20,000 GWH (20 billion kWh), then the estimate of “ex-ante demand will be 20,000/0.95 = 21,053 GWH. I used NEPA (now PHCN) power outages/rationing data and a sample of residential electricity consumption and billing records to estimate R which yielded about 0.95 in the mid 1980s. I then divided past consumption data by 0.95 to obtain estimates of “ex-ante” demand which I then regressed on some explanatory variables such as national disposable income at constant price, industrial production, five-year cumulative capital formation in machinery and equipment, number of customers and unit price of electricity. I estimated separate demand functions for residential, commercial and industrial customers of NEPA. The results were very interesting and I was able to simulate the model to study the impact of changes in the unit price of electricity on the “equilibrium” condition of the market and the profit/loss account of NEPA as well as in making long-term forecast of electricity demand and the capacity investment requirements.

Given the disequilibrium and imperfect conditions of the petroleum products market in Nigeria, an econometric disequilibrium approach is the best method for projecting future demand for and optimal prices of petroleum products. The first step in this approach is to adjust the “metered” consumption (supply) data to obtain estimates of ex-ante demand. The second step is to regress the ex-ante demand data on explanatory variables that determine the demand for each type of petroleum product. The major challenge is how to adjust the consumption data. Unlike the electricity market, we do not have estimates of reliability index for petroleum products supply. The way to estimate the ex-ante demand (D) would be to use extraneous information from surveys to estimate the impact of scarcity (disequilibrium) on the quantity consumed during an accounting period (Cd) and add this to the metered consumption (C) and then subtract estimates of the quantity smuggled out of the country, round-tripped and over-invoiced (Cs) and then add the quantity produced/supplied by illegal refineries (Ci). In order words, D = C + Cd – Cs + Ci. It may also be necessary to compare the per capita estimated D data with those of other countries at same level of development or with similar demand partners as Nigeria (e.g. Indonesia, Pakistan, Iran, Ghana and Egypt) and make further adjustments. The final D can then be regressed on appropriate explanatory variables for each product. For instance the demand for (petrol) PMS can be regressed on disposable income, population, the price of PMS, the price of substitutes, the number of registered vehicles, estimates of the number of other PMS-consuming machinery such as small generators, etc. Similarly functions can be formulated for the demand for kerosene (HHK), diesel (AGO), liquefied petroleum gas (LPG) and other products. Other functions can be specified including those for supply and subsidy, and the model can be simulated to determine the level of subsidy associated with different price levels and the price level required to establish equilibrium over time.

Finally, I must say that it is shameful that the FMF has engaged a foreign consulting firm to estimate the consumption of petroleum products in Nigeria in order to have “authentic figures” for planning purpose. What has happened to Nigerian energy economists and econometricians at home and abroad, who can do the same job, and even better? The FMF cannot claim that it is not aware of Nigerians who can do this job at a far lower cost. The situation reminds me of my experience in 1987 after completing my econometric disequilibrium study of the Nigerian electricity market. I approached the top management of NEPA and gave them a copy of the paper and request to sponsor my trip to Japan to participate in the Energy Conference to present it. NEPA refused to sponsor my trip and I was unable to attend the conference. However, the organizers were gracious enough to include my paper in the post-conference proceedings because of the seminal nature of the paper. I am sure NEPA would have been more than willing to pay 100 times the cost of my trip to a foreign consultant to conduct a similar study for them. Needless to say, NEPA was not interested in my work and they did not even invite me to discuss it, even though the Statistical Department of NEA provided me with much of the data I used in the study. Elsewhere in the world, government agencies issue requests for applications/proposal for local research organizations to apply for grants to conduct studies and implement program that support such agencies’ objectives. I think now is the time for NNPC, PPPRA, FMF and other government agencies to devise ways of utilizing the wealth of experience and expertise of Nigerians in addressing some of the planning and implementation challenges they face rather than turning to foreign consultants.