Introduction The Lahore Journal of Economics 21 : 1 (Summer 2016): pp. 99–122 Electricity Consumption Patterns: Comparative Evidence from Pakistan’s Public and Private Sectors Karim Khan,* Anwar Shah** and Jaffar Khan*** Abstract This study examines the behavioral aspect of Pakistan’s energy crisis by comparing electricity consumption in the public and private sectors. Specifically, we compare consumption patterns of electricity across a sample of student hostels at two public sector universities and privately run student hostels. In addition, we collect household data for a sample of students at Quaid-i-Azam University (QAU) in Islamabad and compare their average electricity consumption with that of the public sector university hostels. We find that the latter’s average consumption of electricity is significantly higher than among private hostels and households. In assessing the moral hazard problem of the public sector in this context, we test the energy conservation behavior of QAU students and the university administration. The results show that students are largely indifferent to conserving electricity, while the administration pays little attention to the use of energy-efficient lights and equipment. Keywords: Electricity consumption, public sector, private sector, moral hazard, conservation of electricity, organizational inefficiency. JEL classification: H83, D12, D00, D03, D04. 1. Introduction Pakistan’s severe energy crisis has prevented the economy from reaching its potential in recent years. At the same time, the demand for energy is increasing continuously. For instance, from 2001 to 2008, the demand rose by almost 6 percent per annum (Kessides, 2013). This surge in demand has led to continuous growth in the gap between demand and supply over the last 30 years. In May 2012, the shortfall was estimated at 6,000 MW (National Electric Power Regulatory Authority, 2011a, 2011b).1 * Assistant Professor, Pakistan Institute of Development Economics, Islamabad ** Assistant Professor, Department of Economics, Quaid-i-Azam University, Islamabad. *** MPhil Student, Department of Economics, Quaid-i-Azam University, Islamabad. 1 See also http://tribune.com.pk/story/154420/countrywide-energy-shortage-as-pepco-increases- loadshedding/ and http://www.theguardian.com/global-development/2012/may/29/pakistan-energy- shortfall-coal-power-plants. Karim Khan, Anwar Shah and Jaffar Khan 100 This declined slightly to 4,250 MW in June 2013 with the demand for power at 16,400 MW and power generation (supply) at 12,150 MW. However, the gap is still alarming.2 Predictions for the near future indicate that the gap between demand and supply is likely to increase to 8,000 MW by 2017 and to 13,000 MW by 2020 (Shahbaz, 2011). The energy crisis can be studied from two perspectives: the supply side and the demand side. The supply side incorporates issues related to the production and distribution of electricity. The demand side relates to the inefficient use of electricity, which in Pakistan, is the dominant problem. According to Ullah, Khan and Akhtar (2014), 52 percent of the increase in energy intensity since 1972 was caused by the inefficient use of energy. This evidence is supplemented by the fact that Pakistan lags far behind most developed and many developing countries in terms of efficiency. For instance, for each dollar of GDP, Pakistan consumes 15 percent more energy than India and 25 percent more than the Philippines. In addition, Pakistan’s energy consumption per unit of GDP is fivefold that of the average for developed countries and twice that of the world average. The potential saving from the efficient use of energy in Pakistan is estimated at 18 percent or 11.16 million tons of oil- equivalent (MTOE), which would result in a 51 percent reduction in net oil imports (Pakistan Energy Sector Taskforce, 2010).3 All these trends imply that Pakistan has the potential to save electricity through demand- side measures, but has failed to do so thus far. At the global level, researchers and policymakers have paid special attention to the trends and dimensions of the energy crisis, including higher energy prices, the instability in the supply of different components, rapid energy depletion and global warming. While there is substantial research on energy problems in the context of Pakistan, most studies have looked at changes in energy prices and their relation to economic growth, inflation and other macroeconomic indicators (see, for instance, Jamil & Ahmad, 2010; Javid, Javid & Awan, 2013; Shahbaz & Feridun, 2012). To our knowledge, no original work has been carried out from a micro- perspective in Pakistan. This study endeavors to initiate this angle by discussing the behavioral aspect of the demand for energy in Pakistan. In particular, we are interested in examining whether electricity consumption behavior differs between the public and private sectors and, if so, why. 2 In 2008, the shortfall reached 4,000 MW for the first time. The Pakistan Ministry of Planning, Development and Reform (2015) reported a peak shortfall of 7,000 MW on average as of July 2014. 3 The Pakistan Energy Sector Taskforce (2010) estimates that 6.1 MTOE (or 15.4 percent of the total energy consumed in the country) could potentially have been saved in the fiscal year 2008 alone. Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 101 In order to support our hypothesis, we have collected data on the electricity consumption of two public sector universities and a sample of private hostels and households in Islamabad. Specifically, we compare the average consumption of electricity per capita between the university hostels and the private sector hostels and households. We find a significant difference in the consumption of electricity between the two sectors. The average consumption across the public sector hostels, however, is not significantly different. Our findings support the hypothesis that public sector universities are characterized by inefficient energy use. We examine the behavioral aspect of this inefficiency based on a survey conducted at the Quaid-i-Azam University (QAU) hostels, which documents the number of rooms left locked with electrical appliances still running. The results show that a significant number of students leave their rooms locked without having switched off the electrical appliances inside. This implies that, as consumers, students in the public sector are indifferent to the conservation of electricity. We have also collected data on the types of electrical lights used, which shows that little attention is paid to switching to energy-efficient devices at public sector universities. Section 2 describes the literature. Sections 3 and 4 provide a theoretical background and methodology. Section 5 describes the data collected. Section 6 analyzes the results and Section 7 concludes the study. 2. A Review of the Literature The literature on the behavioral aspect of energy consumption focuses on three strands: users’ behavior at the institutional level, household behavior and the use of electricity-efficient appliances. The first strand includes considerable research asserting that the provision of free electricity in institutions makes people less likely to conserve electricity (see Siero, Bakker, Dekker & van den Burg, 1996; Scherbaum, Popovich & Finlinson, 2008; Zhang, Wang & Zhou, 2013a, 2013b, 2014). Different factors shape such behavior. Scherbaum et al. (2008) focus on individual-level factors, but find that the public environmental consciousness is an important predictor of personal environmental norms. These, in turn, affect self-reported energy-saving behavior at the workplace.4 In contrast, Zhang et al. (2013b, 2014) gauge the role of variables such as social and individual benefits, and the organization’s 4 Scherbaum et al. (2008) do not find any direct effect of the environmental worldview on self- reported energy-saving behavior. Karim Khan, Anwar Shah and Jaffar Khan 102 own energy-saving norms. They suggest that environmental, personal and organizational benefits as well as the organization’s energy conservation patterns induce employees to conserve electricity. The second strand of the literature examines households’ energy- saving behavior.5 Abrahamse, Steg, Vlek and Rothengatter (2007) look at the effect of tailored information, tailored feedback and goal setting on energy consumption and saving behavior. They show that households that have been exposed to interventions save 5.1 percent in energy consumption compared to households in the control group, which saved only 0.7 percent. Ek and Söderholm (2010) argue that, in Sweden, providing specific information on energy saving plays an important role in energy conservation. Moreover, the willingness to save electricity differs by age group – retired people appear more willing to save than the average individual. In a similar study, Brounen, Kok and Quigley (2013) examine the role of awareness and energy literacy in the behavior of Dutch households. They find that households that are more aware of their energy consumption are also more efficient because they conserve and organize their energy consumption better. Gyberg and Palm (2009) show that information in the residential sector can help achieve sustainable energy systems and control excess demand for energy in people’s daily lives.6 Other studies analyze the role of socio-demographic and economic variables. Abrahamse and Steg (2009) assert that family size and income have a significant effect on households’ direct and indirect energy consumption.7 Wang, Zhang, Yin and Zhang (2011) find that smaller expenditures, subsidized energy conservation, social norms and the experience of energy shortfalls enhance electricity-saving behavior among Chinese households. However, the discomfort caused by energy- saving activities still has a negative effect on conservation behavior.8 5 See: Barr, Gilg and Ford (2005); Lindén, Carlsson-Kanyama and Eriksson (2006); Sardianou (2007); Abrahamse, Steg, Gifford and Vlek (2009); Thøgersen and Grønhøj (2010); Martinsson, Lundqvist and Sundström (2011). 6 Feng, Sovacool and Vu (2010) argue that economic benefits and awareness play an important role in energy conservation. 7 Abrahamse and Steg (2009) find that, among Dutch households, socio-demographic variables are more important than psychological variables in determining energy consumption. 8 The other control they use – similar to the level of education, income and gender – has no impact on people’s willingness to reduce electricity consumption. Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 103 The third strand of the literature focuses on the use of electricity- efficient appliances (including lights, air conditioners, washing machines, refrigerators, TV sets, etc.). Guan, Mills and Zhang (1997) examine the problems and prospects of energy-efficient lighting in China. They find that lack of awareness, the cost-effectiveness of power efficiency projects, primitive manufacturing processes and strong incentives to export energy-efficient products constrain the local use of efficient lighting. Government policies also have an impact on the adoption of energy-efficient appliances. Ma, Andrews-Speed and Zhang (2011) find that energy-efficient appliances and the government’s energy efficiency polices have positive implications for energy conservation in China. Accordingly, they suggest that energy conservation can be enhanced by subsidizing and promoting the use of energy-efficient appliances. Ma, Andrews-Speed and Zhang (2013) find that households prefer comfort and convenience over energy conservation, but that this behavior can be changed through economic incentives such as charging higher prices for electricity or offering discounts on the use of energy-efficient appliances. 3. Theoretical Framework Our starting point is the standard classification of goods shown in Table 1, which divides goods in a society into four categories. Private goods are competed for, that is, individuals can be excluded from their use. Common property resources are defined by two characteristics: it is difficult to exclude users and the use of the good by one user diminishes the benefits available to other users. Nonrival goods from which users can be readily excluded (the opposite of common property resources) are called “spite goods” because this exclusion does not enhance welfare.9 Common property resources are similar to public goods in that user exclusion is difficult, but they are also similar to private goods in that users must compete for them. By contrast, public goods can neither exclude users nor are users compelled to compete for them. Table 1: A taxonomy of goods Rival Nonrival Excludable Private goods Spite goods Nonexcludable Common property Public goods Source: Bowles (2006). 9 Examples include collecting a toll on a little-used highway or charging for admission to an uncrowded museum. Karim Khan, Anwar Shah and Jaffar Khan 104 In most Pakistani public sector universities, the consumption of electricity is divided into three categories of use: (i) the electricity that is used in common areas such as classrooms, libraries, laboratories and public access routes, (ii) the consumption of electricity specifically by employees in offices, and (iii) the electricity used by students in residence. In the first category, the consumption is paid for by the university as public university fees are subsidized,10 that is, electricity is a “public” good. This means that employees and students do not pay for the electricity they consume during working or class hours. Hence, it is likely that, in the absence of moral responsibility and the positive marginal cost of an extra unit of electricity, students and employees will be indifferent to whether their electricity consumption is efficient. Similarly, in the third case, there is the probability of moral hazard with regard to electricity consumption. In contrast, private hostels are run for profit and, as a consequence, resident students are fully liable for the costs they incur.11 As profit-maximizing agents, hostel owners will do their best to reduce their running costs, for instance, by monitoring the unnecessary or overuse of electricity. Moreover, the profit motive may induce them to use electricity-efficient appliances to reduce costs. All these factors will ensure the optimal consumption of electricity in private hostels. In the same way, households are fully liable for the cost of the electricity they consume. As neoclassical economic agents, they will equate their marginal cost of electricity with the marginal benefit of consumption. This is their incentive to avoid overuse and to use electricity-efficient appliances in their homes.12 Based on this discussion, we hypothesize that the consumption of electricity is inefficient in the public sector compared to the private sector. There are three justifications for this hypothesis. First, in the public sector, the marginal cost of using an additional unit of electricity is 0 while the marginal benefit is positive. This implies a higher level of average consumption in public sector university hostels than in private hostels and households. Second, the cost of electricity to individuals in the public sector is hidden and, therefore, we expect they will not bother to avoid using extra electricity. Third, in the public sector in Pakistan, there is no formal or informal punishment mechanism for the misuse of electricity, which significantly reduces the transaction cost associated with any misuse. 10 Universities are financed mainly by the federal government through the Higher Education Commission. 11 In particular, this payment must ensure that hostel owners have profit rates as an incentive. 12 For a detailed discussion, see Stern and Gardner (1981); Stern (2000); Howard (1997). Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 105 4. Methodology We carry out a comparative analysis of the public and private sectors by using two public sector universities – QAU in Islamabad and the University of Baluchistan in Quetta – as case studies. We have collected electricity consumption data for the university hostels, 18 private hostels in Islamabad and the households of a sample of QAU students. In comparing the data for the public sector and the private sector, we expect a significant difference in consumption patterns. Next, we investigate the behavioral causes for this difference. In particular, we are interested in showing that users in the public sector are indifferent to the conservation of electricity. For this purpose, we conduct two types of behavioral observations. First, we survey the QAU hostels to determine how many rooms are left locked but with the lights inside switched on. We expect this to be the case for a significant number of rooms. Second, we compare the types of electrical appliances used in both sectors. Again, we expect private sector users to be more concerned about electricity-efficient appliances than public sector users. 5. Collection and Characteristics of Data The data for the two universities’ electricity consumption was obtained from their monthly bills, to access which we submitted a formal application to the relevant department at each university. We also asked for information on the number of students and staff and the types of lighting used in each block.13 Next, we noted down the 14-digit reference number on the bill (see Table A1 in the Appendix for details) and used it to download the user’s annual billing data for 2013 and 2014 from the district electrical supply company’s website. For a comparative analysis, we calculated the per capita electricity consumed in units14 (the total number of consumed units divided by the number of actual users). As Table A1 in the Appendix shows, QAU has nine hostels, five annexes and six student residences. Collectively, these comprise 1,051 rooms, including accommodation for both men and women. QAU has 13 We needed the number of students and staff that actually consumed electricity in each block in order to find the per capita consumption of electricity. The enumerator also counted the number of each type of light in each block. 14 Per capita consumption of electricity can also be measured in monetary terms. However, electricity companies in Pakistan charge progressively higher rates for different units of electricity consumed, which makes the monetary measure impractical for analytical comparison. Alternatively, our measure of the per capita consumption of electricity is independent of pricing policies. Karim Khan, Anwar Shah and Jaffar Khan 106 approximately 8,000 students, around 33 percent of who live in its hostels. The University of Baluchistan has around 9,500 students and 17 hostels, one of which is for women. Around 1,700 students live in these hostels, which comes to almost 18 percent of the total student population. The second sample comprises 18 privately run hostels in Islamabad. The data collected included the 14-digit reference number obtained from each hostel owner’s electricity bill, the total number of resident students, and the types and number of lights used in each case. Informal discussions revealed that these hostels are administered primarily by the owners, which implies that it is in their interest to monitor the use of electricity. Finally, the household data on electricity consumption was collected through a questionnaire distributed among a random sample of QAU students living at the university hostels (see Appendix).15 Among other information, they were asked to record their 14-digit electricity bill reference number and the total number of energy savers, tube-lights, bulbs, air conditioners, etc., used in their homes. Based on this information, we used the reference numbers to download each user’s billing data and recorded the total number of units consumed by each household. The data represents households across 35 districts in Pakistan. Of almost 400 questionnaires distributed, only 260 were completed and returned. After accounting for those with missing information, the net number of usable questionnaires dropped to 106. 6. Empirical Evidence The empirical findings are based on a comparison of electricity consumption across the public and private sectors, and a behavioral explanation for the differences observed. 6.1. Comparative Analysis of Public and Private Sector Consumption Table 2 summarizes the electricity consumption data collected from the two universities and the private sector hostels and households. The average monthly consumption per student ranges from 26.41 to 104.12 units in the QAU hostels (or 26.41 and 79 if we remove the outlier). In contrast, the average monthly consumption per student ranges from 8.93 to 27.87 units in the private hostels. The level of energy consumed at 15 The data was collected from a sample of undergraduate, graduate and doctoral students. Although the questionnaire was also distributed among the women’s hostels, the response rate there was low. Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 107 the QAU hostels is significantly higher than that consumed at the private sector hostels (p < 0.05).16 Both types of hostels offer a similar level of services,17 which implies that the public sector hostels use energy less efficiently because they consume more while providing the same services. This characterizes the problem of moral hazard, where the findings indicate that students misuse or overuse electricity in the public sector. Table 2: Comparative electricity consumption A. Public sector Hostel name Average consumption per month in units Total number of students Average monthly consumption per student in units QAU Hostel #01 9,903 265 37.37 Hostel #02 6,513 245 26.59 Hostel #03 15,863 215 73.78 Hostel #04 17,460 221 79.00 Hostel #05 26,550 255 104.12 Hostel #06 14,363 378 38.00 Hostel #07 8,407 237 35.47 Hostel #08 10,087 245 41.17 Hostel #09 10,740 247 43.48 Three annexes 11,732 249 47.12 C-43 744 19 39.15 C-44 660 19 34.74 C-45 746 20 37.29 C-46 586 19 30.82 C-47 695 19 36.58 C-48 502 19 26.41 University of Baluchistan Hostel blocks 1–15 45,907 1,216 37.75 Hostel block 16 8,227 289 28.47 Girls’ hostel 19,373 339 57.15 Continued… 16 This p-value is the corresponding value of the t-statistic of unequal means while using the monthly consumption per student as the observation. See the simple regression for statistical inference in Table 4, in which the dummy for the public sector is highly significant. 17 In Islamabad, most private hostels offer services such as hot water in winter, cold water in summer, a fan in the room, a regular water supply and ironing facilities, etc. QAU offers the same services. Karim Khan, Anwar Shah and Jaffar Khan 108 Table 2: Comparative electricity consumption (Continued...) B. Private sector Hostel name Average consumption per month in units Total number of students Average monthly consumption per student in units Ali 257 18 14.30 Bilal 210 10 21.04 Chughtai 177 12 14.72 Danish 1,728 62 27.87 Ehsan 197 11 17.95 Fasal 836 28 29.85 Ghazi 203 12 16.88 Haris 410 18 22.78 Idrees 353 20 17.67 Jamsheed 286 32 8.93 Kurram 1,341 52 25.78 Liaqat 681 32 21.28 Moeen 310 20 15.52 Nashaad 343 15 22.85 Owais 273 20 13.65 Perooz 477 18 26.51 C. Household data Category Sum of average monthly consumption Total number of individuals in household Average monthly consumption per person Lowest 25% 1,962.58 263 7.46 Second lowest 25% 3,276.83 200 16.38 Second highest 25% 4,568.92 179 25.52 Highest 25% 9,316.17 286 32.57 Notes: Average monthly consumption = total annual consumption in a typical year / 12. Average monthly consumption per student = average monthly consumption of hostel / number of resident students. Total number of households = 106, divided into four quartiles based on average monthly consumption of electricity. Average monthly consumption per person for each quartile = sum of average monthly consumption by all households in that quartile / total number of household members in that quartile. Source: Authors’ calculations. In order to check the robustness of this result, we analyze the data for another public sector university, the University of Baluchistan, and the sample of private households.18 Table 2 shows that the average monthly consumption of electricity per student in the university’s hostels 18 The household data survey was conducted at QAU (see the questionnaire in the Appendix for details). Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 109 is not very different from that of the QAU hostels. This indicates that students’ energy use behavior is similar across the two public sector universities. A comparison with the household sample shows that average consumption in the public sector is greater. To check for the effects of outliers, we compare the average monthly consumption per person across all four types of consumers for a typical year.19 As Figure 1 shows, in most months (except August and September), the consumption per student at QAU is greater than that of the private hostels. However, given that universities are closed in the summer, it is not suitable to compare the average consumption in these months.20 In spite of this effect, the t-test of unequal variances shows a significant difference in the electricity consumption of the two types of hostels (p < 0.05). This anomaly is not robust with regard to the University of Baluchistan, where the average monthly consumption is larger than that of the private hostels all year round. Figure 1: Comparative electricity use per student in public sector and private sector hostels Similarly, if we compare the average consumption of electricity at QAU with the average consumption of the household sample, we find that the university consumes more energy than the households in all months except August and September. However, the households consume more electricity in the summer when the use of electrical 19 This, in turn, allows us to counter any billing errors on the part of the electricity company or the effect of an individual hostel that may be an outlier. 20 The regulations for running private hostels are not well enforced in Pakistan. It is possible that they are used for other purposes during the summer holidays, but we have no data to confirm this. 0 20 40 60 80 100 1 2 3 4 5 6 7 8 9 10 11 12 Units Months QAU hostels Private hostels UB hostels Households Karim Khan, Anwar Shah and Jaffar Khan 110 appliances such as exhaust fans, air conditioners and refrigerators increases. In comparison, the QAU hostels are usually closed during the summer. Again, this result is not robust with respect to the University of Baluchistan, which consumes more electricity per student than the households all year round. The highest level of average consumption at QAU falls in January, possibly because students use nontraditional means of heating during the peak of winter.21 However, if this is the case, it has an important behavioral policy implication: permitting students to use individual heaters or providing a centralized heating system might address the moral hazard problem of electricity consumption. The relatively low consumption per student during July, August and September reflects that the QAU hostels are closed, although this pattern does not emerge for the University of Baluchistan hostels. We apply the Mann–Whitney U test for a paired comparison of per person consumption across all types of consumers. As Table 3 shows, QAU’s average monthly consumption of electricity is significantly different from that of the private hostels and households; the same holds for average consumption at the University of Baluchistan. The average consumption of the two universities, however, is similar. Table 3: Results of Mann–Whitney U test Comparison p-value QAU hostels vs. private hostels 0.0027*** QAU hostels vs. households 0.0111** University of Baluchistan hostels vs. private hostels 0.0000*** University of Baluchistan hostels vs. households 0.0003*** QAU hostels vs. University of Baluchistan hostels 0.6442 Note: *** p < 0.01 and ** p < 0.05. Source: Authors’ calculations. The results imply that (i) electricity user behavior in the public sector is different from that of the private sector, and (ii) public sector organizations do not show any significant difference in their electricity use behavior. Table 4 confirms this finding, using a dummy variable for the public sector in a simple regression. 21 University hostels in Pakistan tend not to be heated in the winters and the use of individual heaters is not allowed. However, in the absence of administrative monitoring, students often use individual heaters. Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 111 Table 4: Simple regression results Dependent variable = per person electricity consumption Intercept 19.98*** Public sector dummy 24.99*** R2 0.43 Note: *** p < 0.01. Source: Authors’ calculations. 6.2. Behavioral Explanation for Differences in Public and Private Sectors This section analyzes the misuse of electricity by end-users and explains in part organizational inefficiency as far as the use of electricity- efficient appliances is concerned. 6.2.1. Misuse of Electricity in QAU Hostels Table 5 gives the results of the survey documenting the misuse of electricity in the QAU hostels. One indicator of students’ conservation behavior is whether or not they switch off the lights in their room when leaving. This is measured by the number of rooms left locked with the lights inside switched on. The survey was carried out at three different times: from 0900 to 1300, from 1400 to 1700, and from 1800 to 2100 hours. The three survey rounds were conducted on three different days. Karim Khan, Anwar Shah and Jaffar Khan 112 T a b le 5 : M is u se o f e le ct ri ci ty i n Q A U h o st e ls H o st e l n a m e N o . o f ro o m s F ir st s u rv e y 0 9 0 0 t o 1 3 0 0 S e co n d s u rv e y 1 4 0 0 t o 1 7 0 0 T h ir d s u rv e y 1 8 0 0 t o 2 1 0 0 M is u se % a cr o ss t h re e su rv e y s L o ck e d ro o m s L ig h ts o n M is u se % L o ck e d ro o m s L ig h ts o n M is u se % L o ck e d ro o m s L ig h ts o n M is u se % H o st e l # 0 1 1 0 5 7 5 5 7 7 6 .0 0 7 5 6 3 8 4 .0 0 2 6 1 7 6 5 .3 8 7 7 .8 4 H o st e l # 0 2 1 0 4 7 1 5 8 8 1 .6 9 6 2 5 4 8 7 .1 0 3 1 1 4 4 5 .1 6 7 6 .8 3 H o st e l # 0 3 1 0 3 5 6 4 9 8 7 .5 0 7 6 5 9 7 7 .6 3 2 7 2 2 8 1 .4 8 8 1 .7 6 H o st e l # 0 4 1 0 2 7 7 6 6 8 5 .7 1 7 2 6 1 8 4 .7 2 4 0 3 2 8 0 .0 0 8 4 .1 3 H o st e l # 0 5 1 1 6 6 6 5 4 8 1 .8 2 8 0 6 1 7 6 .2 5 3 3 2 1 6 3 .6 4 7 5 .9 8 H o st e l # 0 6 1 1 4 7 3 6 2 8 4 .9 3 5 7 4 9 8 5 .9 6 6 2 5 1 8 2 .2 6 8 4 .3 8 H o st e l # 0 7 5 3 4 3 3 1 7 2 .0 9 2 9 2 4 8 2 .7 6 3 1 2 6 8 3 .8 7 7 8 .6 4 H o st e l # 0 8 1 1 5 7 1 5 9 8 3 .1 0 5 1 4 4 8 6 .2 7 4 2 3 1 7 3 .8 1 8 1 .7 1 H o st e l # 0 9 1 1 5 7 8 6 7 8 5 .9 0 5 2 4 2 8 0 .7 7 4 4 3 5 7 9 .5 5 8 2 .7 6 F iv e a n n e x e s 1 0 0 6 1 4 7 7 7 .0 5 4 8 3 7 7 7 .0 8 2 7 1 9 7 0 .3 7 7 5 .7 4 S ix C -t y p e h o st e ls 2 4 2 2 1 4 6 3 .6 4 1 8 1 1 6 1 .1 1 5 3 6 0 .0 0 6 2 .2 2 A v e ra g e o f a ll h o st e ls 7 9 .9 5 8 0 .3 3 7 1 .4 1 7 9 .7 1 S o u rc e: A u th o rs ’ ca lc u la ti o n s. Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 113 In the first two survey periods, almost 80 percent of the hostel rooms were left locked with the lights inside switched on. This decreases to around 70 percent in the third survey round, which may be because most students remain in their rooms in the evening. Hostels 1, 2 and 5 and the C- type hostels are women’s hostels. We find no difference in the misuse of electricity where gender is concerned. Most students also tended to leave their fans switched on, but we did not collect data on this. Thus, students’ indifference to saving electricity might partly explain the higher average consumption at QAU relative to the private hostels and households. 6.2.2. Organizational Inefficiency in the Selection of Electrical Equipment In addition to consumer inefficiency, there is also organizational inefficiency in the public sector to consider. Table 6 compares the use of electricity-efficient lights across the public and private sectors. Table 6: Comparative use of efficient lighting appliances in public and private sectors Lighting appliance QAU hostels Private hostels Households No. As % of total No. As % of total No. As % of total Tube-lights 17,369 95.6 72 7.8 508 25.1 Bulbs 519 2.9 10 1.0 212 10.5 Energy savers 280 1.5 831 91.0 1,302 64.4 Total 18,168 100.0 913 100.0 2,022 100.0 Source: Authors’ calculations. Clearly, there is a considerable difference in the use of energy- efficient appliances in the two sectors. The private sector uses relatively more energy-efficient appliances than the public sector. Out of a total of 18,168 large and small lights installed at QAU, the number of tube-lights is 17,369 (95.6 percent), the number of bulbs is 519 (2.85 percent) and the number of energy savers is merely 280 (1.54 percent). In comparison, the percentage of energy savers installed in the private hostels and households is 91 and 64.4 percent, respectively. This shows that using energy-efficient appliances is still a low priority in the public sector compared to the private sector; it also reflects QAU’s organizational inefficiency in this context. Karim Khan, Anwar Shah and Jaffar Khan 114 7. Conclusion This study was motivated by the literature on the moral hazard and “free rider” problems associated with public goods. Based on this argument, we compare the consumption patterns of electricity across the public and private sectors, using a sample of public sector university hostels and privately run hostels to represent the two sectors. We also conduct a short survey to compare the average energy consumption of public sector hostels and households. Finally, we analyze the behavioral causes of the inefficiency that emerges – both on the part of users and the university administration – in the public sector. There are three main findings of this study. First, the average consumption of electricity per capita in the public sector is significantly higher than in the private sector, including both the private hostels and households. This result is robust with respect to different organizations in the public sector (QAU and the University of Baluchistan). This has an important policy implication, given Pakistan’s energy crisis. In order to offset the rising demand for electricity through conservation, policymakers must focus on the public sector. A twofold strategy is needed to enhance organizational efficiency in terms of monitoring the use of electricity, and addressing consumer behavior by developing norms that discourage the misuse of electricity. Employing both formal and informal institutional structures can help control the moral hazard associated with the misuse of electricity in the public sector. Second, the survey finding that almost 80 percent of hostel rooms at QAU are left locked with the lights inside switched on implies considerable indifference to energy conservation among students. We recommend devising an efficient monitoring and punishment mechanism to discourage this consumption behavior at public sector institutions. Finally, we find that the use of energy-efficient lighting in the public sector is a negligible 1.54 percent relative to 91 percent in private hostels and over 64 percent in the households sampled. This implies that the organizational structure at public sector institutions is significantly inefficient with regard to the use of energy savers. Expanding the scale at which energy-efficient appliances are used in this sector would help conserve electricity. However, future research is needed in this area to provide clear guidelines for overcoming the energy crisis. Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 115 References Abrahamse, W., & Steg, L. (2009). How do socio-demographic and psychological factors relate to households’ direct and indirect energy use and savings? Journal of Economic Psychology, 30(5), 711–720. Abrahamse, W., Steg, L., Gifford, R., & Vlek, C. (2009). Factors influencing car use for commuting and the intention to reduce it: A question of self-interest or morality? Transportation Research Part F: Traffic Psychology and Behavior, 12(4), 317–324. Abrahamse, W., Steg, L., Vlek, C., & Rothengatter, T. (2007). The effect of tailored information, goal setting and tailored feedback on household energy use, energy-related behaviors and behavioral antecedents. Journal of Environmental Psychology, 27(4), 265–276. Barr, S., Gilg, A. W., & Ford, N. (2005). The household energy gap: Examining the divide between habitual and purchase-related conservation behaviors. Energy Policy, 33(11), 1425–1444. Bowles, S. (2006). Microeconomics: Behavior, institutions and evolution. Princeton, NJ: Princeton University Press. Brounen, D., Kok, N., & Quigley, J. M. (2013). Energy literacy, awareness and conservation behavior of residential households. Energy Economics, 38, 42–50. Ek, K., & Söderholm, P. (2010). The devil is in the details: Household electricity saving behavior and the role of information. Energy Policy, 38(3), 1578–1587. Feng, D., Sovacool, B. K., & Vu, K. M. (2010). The barriers to energy efficiency in China: Assessing household electricity savings and consumer behavior in Liaoning Province. Energy Policy, 38(2), 1202–1209. Guan, F., Mills, E., & Zhang, Q. (1997). Energy-efficient lighting in China: Problems and prospects. Energy Policy, 25(1), 77–83. Gyberg, P., & Palm, J. (2009). Influencing households’ energy behavior: How is this done and on what premises? Energy Policy, 37(7), 2807–2813. Karim Khan, Anwar Shah and Jaffar Khan 116 Howard, G. S. (1997). Ecological psychology: Creating a more Earth-friendly human nature. Notre Dame, IN: University of Notre Dame Press. Jamil, F., & Ahmad, E. (2010). The relationship between electricity consumption, electricity prices and GDP in Pakistan. Energy Policy, 38(10), 6016–6025. Javid, A. Y., Javid, M., & Awan, Z. A. (2013). Electricity consumption and economic growth: Evidence from Pakistan. Economics and Business Letters, 2(1), 21–32. Kessides, I. N. (2013). Chaos in power: Pakistan’s electricity crisis. Energy Policy, 55, 271–285. Lindén, A.-L., Carlsson-Kanyama, A., & Eriksson, B. (2006). Efficient and inefficient aspects of residential energy behavior: What are the policy instruments for change? Energy Policy, 34(14), 1918–1927. Ma, G., Andrews-Speed, P., & Zhang, J. (2011). Study on Chinese consumer attitudes on energy-saving household appliances and government policies: Based on a questionnaire survey of residents in Chongqing, China. Energy Procedia, 5, 445–451. Ma, G., Andrews-Speed, P., & Zhang, J. (2013). Chinese consumer attitudes towards energy saving: The case of household electrical appliances in Chongqing. Energy Policy, 56, 591–602. Martinsson, J., Lundqvist, L. J., & Sundström, A. (2011). Energy saving in Swedish households. The (relative) importance of environmental attitudes. Energy Policy, 39(9), 5182–5191. National Electric Power Regulatory Authority. (2011a). Annual report 2010–11. Islamabad: Author. National Electric Power Regulatory Authority. (2011b). State of industry report 2011. Islamabad: Author. Pakistan Energy Sector Taskforce. (2010). Integrated energy sector recovery report and plan. Manila: Friends of Democratic Pakistan, Asian Development Bank and Government of Pakistan. Pakistan, Ministry of Planning, Development and Reform. (2015). Annual plan 2015–16 (chap. 18). Islamabad: Author. Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 117 Sardianou, E. (2007). Estimating energy conservation patterns of Greek households. Energy Policy, 35(7), 3778–3791. Scherbaum, C., Popovich, P. M., & Finlinson, S. (2008). Exploring individual-level factors related to employee energy-conservation behaviors at work. Journal of Applied Social Psychology, 38(3), 818–835. Shahbaz, M. (2011). Electricity consumption, financial development and economic growth nexus: A revisit study of their causality in Pakistan (MPRA Paper no. 35588). Retrieved from https://mpra.ub.uni- muenchen.de/35588/ Shahbaz, M., & Feridun, M. (2012). Electricity consumption and economic growth: Empirical evidence from Pakistan. Quality and Quantity, 46(5), 1583–1599. Siero, F. W., Bakker, A. B., Dekker, G. B., & van den Burg, M. (1996). Changing organizational energy consumption behavior through comparative feedback. Journal of Environmental Psychology, 16(3), 235–246. Stern, P. C. (2000). New environmental theories: Toward a coherent theory of environmentally significant behavior. Journal of Social Issues, 56(3), 407–424. Stern, P. C., & Gardner, G. T. (1981). Psychological research and energy policy. American Psychologist, 36(4), 329–342. Thøgersen, J., & Grønhøj, A. (2010). Electricity saving in households: A social cognitive approach. Energy Policy, 38(12), 7732–7743. Ullah, A., Khan, K., & Akhtar, M. (2014). Energy intensity: A decomposition exercise for Pakistan. Pakistan Development Review, 53(4), 531–549. Wang, Z., Zhang, B., Yin, J., & Zhang, Y. (2011). Determinants and policy implications for household electricity-saving behavior: Evidence from Beijing, China. Energy Policy, 39(6), 3550–3557. Zhang, Y., Wang, Z., & Zhou, G. (2013a). Antecedents of employee electricity-saving behavior in organizations: An empirical study based on norm activation model. Energy Policy, 62, 1120–1127. Karim Khan, Anwar Shah and Jaffar Khan 118 Zhang, Y., Wang, Z., & Zhou G. (2013b). Determinants and implications of employee electricity-saving habit: An empirical study in China. Applied Energy, 112, 1529–153. Zhang, Y., Wang, Z., & Zhou, G. (2014). Determinants of employee electricity saving: The role of social benefits, personal benefits and organizational electricity saving climate. Journal of Cleaner Production, 66, 280–287. Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 119 Appendix Table A1: Hostel reference numbers and addresses Public sector hostels Reference no. QAU Hostel #01 27-14131-0215600 Hostel #02 27-14131-0215700 Hostel #03 27-14131-0214200 Hostel #04 27-14131-0214300 Hostel #05 27-14131-0214500 Hostel #06 27-14131-0052540 Hostel #07 27-14131-0052541 Hostel #08 27-14131-0052542 Hostel #09 27-14131-0052543 Women’s annex 27-14131-0119431 Men’s annex 27-14131-0214400 Women’s annex H-5 27-14131-0214401 C-43 (women’s hostel) 02-14131-0304400 C-44 (women’s hostel) 02-14131-0304300 C-45 (women’s hostel) 02-14131-0304200 C-46 (women’s hostel) 02-14131-0304100 C-47 (women’s hostel) 02-14131-0304000 C-48 (women’s hostel) 02-14131-0304500 University of Baluchistan Hostel blocks 1–15 24481120977600 Hostel block 16 24481140658800 Women’s hostel 24481120977700 Karim Khan, Anwar Shah and Jaffar Khan 120 Private sector hostels Reference no. Main Murree Road, Barakho, Islamabad Ali 10-14131-3382100 Bilal 10-14131-3472700 Chughtai 10-14131-3472400 Danish 12-14131-3917300 Ehsan 10-14131-3472500 Fasal 12-14131-3912800 Ghazi 10-14131-3472501 Haris 10-14131-3471800 Idrees 10-14131-3470300 Sector F-6, Islamabad Jamsheed 14-14111-3167900 Chatta Bakhtawar, Chak Shahzad, Islamabad Kurram 09-14119-3606300 Liaqat 09-14119-3606200 Moeen 09-14119-3608600 Nashaad 09-14119-3607800 Owais 09-14119-4008600 Perooz 09-14119-4009000 Electricity Consumption Patterns in Pakistan’s Public and Private Sectors 121 Survey questionnaire Dear respondent, Assalam-o-Alaikum, This questionnaire aims to collect information on the electricity consumption of households. You are requested to spare the time to answer all the questions below to the best of your knowledge. Your cooperation will enable us to explore various dimensions of the current electricity crisis. The data will be used for research at the MPhil level at the School of Economics at Quaid-i-Azam University, Islamabad. Thank you in advance for your cooperation. MPhil scholar Personal information Age: Level of education: MPhil/PhD/MSc/BA/FSc/matric/below matric Gender: Male/female District: Household information Family structure: Joint/nuclear Where do you live? Official house/own house/rented/other Total number of family members: Below 10 years: Between 10 and 18 years: Above 18 years: Total number of males: Total number of females: Karim Khan, Anwar Shah and Jaffar Khan 122 House constructed: 1 year ago/2 years ago/5 years ago/10 years ago/more than 10 years ago Total number of rooms: Total number of energy savers: Total number of tube-lights: Total number of bulbs: Total number of air-conditioners (if any): Does your family receive an electricity bill regularly? Yes/no Please write down the 14-digit reference number of the electricity bill: Please tick the relevant answer to the following questions: My family is aware of energy-saving appliances. Yes/no/don’t know My family members turn off the lights when there is bright sunlight. Yes/no/don’t know All my family members know that conserving electricity decreases expenditure. Yes/no/don’t know The elders in my family encourage electricity saving. Yes/no/don’t know Do you think students conserve electricity at QAU in the same way they would at home? Yes/no/don’t know